Preposterous Universe

 
Vote tampering

I like to think I am a gracious loser, not a whiny sour-grapes conspiracy theorist. Which is why I'm hesitant to suggest that malicious tampering with the vote counts somehow contributed to Bush's win last Tuesday. But, at the very least, there is a significant pattern of irregularity going on, which should be cleared up if only for the sake of future elections.

We already know that one Ohio county gave Bush 3,800 extra votes, a mistake which has now been corrected. And there have been claims of systematic bias, some rather hysterical. But Brian Leiter points to a study that is provocative to say the least. The study compared the declared makeup of each Florida county in terms of registered Democrats and Republicans to the final votes. The results are startling -- a sizeable number of overwhelmingly Democratic counties, mostly smaller ones, where Bush came away with a big victory. Franklin county, 77% Democrat, went for Bush 3,472 to 2,400. Hamilton county, 79% Democrat, went for Bush 2,786 to 2,252. Holmes county, 73% Democrat, went for Bush 6,410 to 1,810. Lafayette county, 83% Democrat, went for Bush 2,460 to 845. Liberty county, where Democrats outnumber Republicans by 88% to 8%, went for Bush 1,927 to 1,070. And on and on.

Interestingly, all of the counties with these wild discrepancies between results and registrations have something in common -- they used optically-scanned ballots whose results were fed into a central PC. In counties where the much-maligned touch-screen voting method was used, the results conformed nicely to what the registrations would have you believe. I suppose the good news is that a recount should be straightforward in such counties -- although we all know how hard it is to get a recount done in this country.

The dramatic conclusion from this would be that there is a simple reason why the exit polls, which were indicating a substantial Kerry victory, deviated so dramatically from the final results -- they were right, the results weren't. (As Dick Morris says, "exit polls are almost never wrong.") But the minimum conclusion is that a grown-up civilized nation like the United States should be able to figure out how to count votes quickly, accurately, and verifiably. I would think that even the Republicans should be in favor making every effort to figure out what really happened, if only to shut up the sore losers once and for all.

Update: Blogging of the President has a list of links to other stories about possible tampering, including a very good piece by Keith Olbermann on how a story like this crosses from the blogosphere into the mainstream media. As I mention in the comments, I hope that there really wasn't any tampering going on; but there's every reason to investigate it as thoroughly as possible.

Update again: People who are more sophisticated than me (see, if I were sophisticated, it would be "than I") have pointed out that it's not really the results in small counties that are most problematic, as those can perhaps be explained by accidents of history creating misleading party registrations. But there is a discrepancy between the optically-scanned results and the touch-screen results even when you control for size of county. See also interesting commentary at Rhosgobel and Pharyngula.

Update yet again: Brian Leiter now points to an online discussion that seems to make sense, and gives good reasons not to take the study referenced above seriously as evidence of tampering. There are still some disturbing stories floating around, especially about Ohio, but nothing that seems to me to be extremely strong evidence of wrongdoing (much less "proof," as Atrios emphasizes). Worth re-checking everything as carefully as possible, of course, if only to give the appearance that the truth is what matters.

~ posted by Sean Carroll at 8:24 PM ~ link ~ ~ ~