There are now two big competing "media recounts" of the Florida presidential vote. The Miami Herald, the biggest paper in the state, plans one that may produce results in a few days. Meanwhile, a consortium of at least eight press organizations (including the New York Times, Washington Post, Wall Street Journal, Associated Press, the Tribune Co., CNN, and the Palm Beach Post) is planning another recount. The consortium's count got started late and will take much longer.
Which will be the crucial recount? Not the Herald's.
Why? Because the Herald will recount--visually inspect--only Florida's 60,000 "undervotes." Undervotes are ballots on which counting machines discerned no vote for president. The Herald decided not to recount the state's more numerous (110,000-120,000) "overvotes"--that is, ballots declared invalid because votes for two or more presidential candidates were detected. The press consortium will count both kinds of votes. The issue of whether to count both kinds of votes is one reason talks between the Herald and the consortium broke down.
The Herald's decision to focus only on the "undervote" might at one time have been a reasonable call. Until near the end of the Florida vote fight, press attention focused almost exclusively on these "no vote" ballots, especially in counties that use "punch-card" ballots with their attendant chad. Undervotes were the votes Gore lawyers said they wanted manually counted. And the statewide hand recount ordered on Dec. 8 by the Florida Supreme Court--the one the U.S. Supreme Court halted the same day it started, effectively ending the election--was also focused explicitly on undervotes. Many observers, including some Gore advisers, dismissed the possibility that valid votes could be hidden in overvoted ballots. After all, if a ballot is marked for two candidates, how is anyone supposed to guess what the intent of the voter was?
But that was before Lake County. In late December, the Orlando Sentinel examined 3,000 or so Lake overvotes and discovered more than 600 valid ballots--producing a 130-vote gain for Gore in this one, small, strongly pro-Bush jurisdiction! How did the Sentinel know they were Gore votes? Because the voters both darkened Gore's circle and wrote in his name. It was because they marked both the Gore circle and the "write-in" circle that their ballots were rejected by the machines as overvotes.
Lake County may have been an aberration. Some other counties inspected their overvotes on Election Day and counted such "write-ins" in their reported results. In punch-card counties, voters aren't given a pen with which to mark their ballots--as they are in counties like Lake that use "optical scanning" machines. Without a writing implement in their hands, punch-card voters are presumably far less likely to make the "write in" mistake.
But there's only one way to find out! What is clear after Lake is that any recount that doesn't include the overvotes is an incomplete recount. In an election decided by a few hundred votes, an undervote-only count ignores more than 100,000 ballots that may have hidden valid votes for Bush or Gore.
How does the Herald justify its undervote-only approach? Mark Seibel, the Herald assistant managing editor who is overseeing the paper's count, says he wanted a "real world" project, which was finding out "what would the result have been" of the Florida Supreme Court-ordered statewide recount "that was stopped." But Lake County officials say they would have included their overvotes in that recount! That "would have been outside the purview of the [Florida Supreme Court's] order," Seibel says. "Overvotes were never an issue in the election" controversy, he declares. True, the U.S. Supreme Court mentioned overvotes. "But," he says, "it's not part of the majority opinion."
"That's on the face of it wrong," counters Alan Murray, the Wall Street Journal's Washington bureau chief. Murray's right. Although Gore (foolishly) never asked to have overvoted ballots recounted, the possibility that they contain valid votes was an issue. Bush's lawyers raised it. The dissenting justices in the Florida Supreme Court raised it. And the U.S. Supreme Court--in the majority opinion--gave the following grounds for overturning the Florida Supreme Court's undervote-centric count:
A manual recount of all ballots identifies not only those ballots which show no vote but also those which contain more than one, the so-called overvotes. Neither category will be counted by the machine. This is not a trivial concern. At oral argument, respondents estimated there are as many as 110,000 overvotes statewide. As a result, the citizen whose ballot was not read by a machine because he failed to vote for a candidate in a way readable by a machine may still have his vote counted in a manual recount; on the other hand, the citizen who marks two candidates in a way discernible by the machine will not have the same opportunity to have his vote count, even if a manual examination of the ballot would reveal the requisite indicia of intent.