The case of the disappearing dead voters
The Washington Post's Glenn Kessler, who writes The Fact Checker blog, yesterday gave four "Pinocchios" – the most, in his scale for factual dishonesty – to South Carolina Attorney General Alan Wilson. The reason: Wilson last year had trumpeted reports of dead people voting in his state as "facts," then, when a subsequent probe issued a report discrediting the allegations, Wilson kept the report under wraps for more than a year, only releasing it pursuant to a reporter's Freedom of Information Act request. Allegations of dead voters have been used to fuel successful efforts to pass state laws creating barriers to access to voting booths, such as voter-ID laws – which, as reported in today's Washington Post, the Justice Department is preparing to challenge in the aftermath of a recent Supreme Court decision chucking out a key portion of the 1965 Voting Rights Act. What's ironic about dead-voter allegations providing impetus for voter-ID laws is that, when scrutinized, the allegations rarely if ever hold up. That, according to the Brennan Center for Justice (BCJ), in a friend-of-the-court brief filed in a U.S. Supreme Court case that, in 2008, ended with the court upholding Indiana's voter-ID law. City Paper, for instance, in 2005, did an exhaustive feature chasing down allegations in a New York Times article that there were "1,500 dead people listed as active registered voters," in Baltimore City, and that "fifty of those dead people somehow voted in the last election." The article, entitled "Election Nights of the Living Dead," was cited in BCJ's Supreme Court brief, and it found that the Times had gotten it wrong: 63 apparently dead people had voted in Baltimore City elections from the 1980s to 2004, not 50 in 2004. After getting a list of 62 of the 63 dead voters, City Paper pared it down to 26 whose deaths could be confirmed with certainty (seven of them were definitely alive), and found that three of them were recorded as having cast votes in elections in 2003 and 2004 – but the state's elections board, having reviewed the three instances, found that they hadn't, in fact, voted, but had been recorded as doing so due to clerical errors. For the remaining 22, the article concluded that "neither [clerical] error nor fraud can be ruled out as the cause" for records reflecting that they had cast votes. But even if some or all of the 22 instances of Baltimore City's dead voters were cases of voter-impersonation fraud, they wouldn't amount to a hill of beans, since they voted in elections spanning nearly 20 years, involving millions of votes cast. As Kessler pointed out regarding the South Carolina case, "the number of alleged dead votes was less than 2/10,000th of all of the votes cast in that election." The specter of dead voters confirms many of the suspicions people have of poll-rigging intrigue, but it disappears when examined. The real risk of fixing elections comes not from voter fraud, but from election-system fraud via exploiting security frailties in computer-based voting software – which City Paper also examined, just as Maryland was starting to commit to computer-voting technology, in an exhaustive 2002 feature. Rather than prompt legislatures to enact measures to minimize the large risks involved in computer-based voting – as opposed to the red-herring risks of dead voters – states and municipalities around the country have implemented computer voting again and again. The very real risks of these systems, meanwhile, have been chronicled again and again by the elections-watchdog group, Black Box Voting. Worrying about zombies tainting elections, and exploiting that fear to reduce voter access to polls, should itself be seen as a form of vote-rigging. The smart move, for those truly interested in full and fair elections, would be for legislatures to assess and address what BBV has relentlessly exposed – the real threat of unverifiable fraud in computer-based voting.