Before the election I’d already planned a trip to my wife’s parents’ place near Fort Lauderdale for mid-November. George Bush beat John Kerry in Florida by about 400,000 votes, a much larger margin than most polls had predicted. Closer contests, like the one in Ohio, had quickly prompted official recounts. The lopsided Florida loss hadn’t, but it was the one my in-laws and I had the most trouble stomaching.
Best of Chicago voting is live now. Vote for your favorites »
I’d picked Glades County for several reasons. It was only a little over an hour’s drive from Fort Lauderdale. More to the point, Bush’s performance there seemed suspiciously strong–he’d gotten 60 percent of the vote in a county where only 30 percent of the voters had registered as Republicans, and he’d increased his 2000 count by a third–much more than Kerry improved on Al Gore’s. Glades also had used machinery manufactured by Diebold, a company headed by Walden O’Dell, who’d pledged to “deliver” electoral votes for Bush. The machinery used optical-scan paper ballots, which, like punch-card ballots, can be recounted by hand–but more easily, because they list the candidate’s name next to the voter’s choice. And the small population of the county would make it possible to recount all of its 4,161 ballots before lunch.
There were plenty of them. The weekend before my trip I attended a forum hosted by Larry Quick, cofounder of the National Ballot Integrity Project. He said, “Numbers from Florida and a lot of places just aren’t credible,” and he put me in touch with Ida Briggs and Kathy Dopp, two mathematicians whose analyses of data from Florida and elsewhere had been widely cited on the Internet. Dopp, who lives in Park City, Utah, argued that results from counties using Diebold optical-scan machines were suspicious and that they didn’t seem to jibe with results from counties with similar demographics. She also told me that various computer models showed that it was extremely unlikely that so many exit polls had been wrong. Briggs, a computer analyst from the Detroit area, told me that Monday, November 15, would be a big day, saying, “We’ve got evidence of an ‘if-then’–it looks like if Kerry reached a certain level his votes started getting counted for Bush.”
“Uh, I don’t think so. I think only I can touch ’em.”
The man apologized briefly, then turned to me and introduced himself as Russell Michaels, a documentary producer doing an election inquiry. I told him I was doing the same thing.
“Oh,” he said. “Well, we shouldn’t be much longer. So you’re doing an actual recount?”
As we headed back to the storage unit, I asked her whom she’d called. I expected a guarded reply or none.
