Paper Trail Urged as E-Voting Fix 

By Kim Zetter   WiredNews   Sep. 23, 2005 PT

Efforts to secure the integrity of electronic-voting machines seemed to get a boost this week, but the debate over the best way to guard against election tampering remained at a fever pitch.

After five months of hearings and deliberations, a high-level election-reform commission led by former President Jimmy Carter and former Secretary of State James Baker recommended that Congress require electronic-voting machines to produce a voter-verifiable paper audit trail by 2008. 
 
But even while the Commission on Federal Election Reform expressed support for paper trails, it acknowledged that they might not be the best solution to address security concerns with e-voting machines. The commission urged researchers to develop new technologies that could resolve the issues more effectively.

The voter-verifiable paper audit trail, or VVPAT, is a printout of the machine's electronic ballot that voters can examine after they make their ions on the machine. The paper would scroll behind a glass partition, allowing voters to reject the ballot before casting it. Election officials would use the paper to audit machines and conduct recounts.

Daniel Calingaert, the commission's associate director, said the commission's concerns about paper trails don't mean they're the wrong answer for now.

"The deliberations of the commission were focused on the need for voting equipment to ... have a voter-verifiable record generated independently of the machine tally and that is available for recounts and audit," he said. "And looking at the technology that's available today, the voter paper trail met those criteria."

Few states are waiting for Congress to act on the paper-trail issue. Twenty-six states have already passed paper-trail legislation of their own, and another 14 are considering it, according to VerifiedVoting.org, which has led the movement for paper-trail laws.

But questions about the paper trails remain, in part because the question of how states implement them is as important as whether they implement them at all.

For example, someone determined to tamper with an election could program e-voting machines to record one vote on paper and a different vote inside the machine. Therefore paper trails are meaningless if officials don't conduct post-election audits to compare at least 1 percent of paper votes against their electronic counterparts.

The commission, organized by American University's Center for Democracy and Election Management, recognized this problem and recommended that states perform audits. But the commission balked at asking Congress to mandate them.

Opponents of paper trails, many election officials among them, raise other questions that the commission did not address. They're adamant that few voters will actually look at the paper record, negating its usefulness. During a test of paper trails last year in Nevada's primary and presidential elections, election observers estimated that fewer than 30 percent of voters bothered to examine the hard copy.

Critics also say the printers will jam, break down or run out of paper, creating more labor for poll workers. And they argue that an election involving numerous races and candidates would produce an unwieldy paper trail that would be time-consuming for voters to review and difficult for election officials to recount especially if the thermal paper used in the printers is tightly curled.

"All of these barriers that election officials talk about are just administrative obstacles that creative thinking could help us get around," said Kim Alexander, founder of the California Voter Foundation. She also said that while it would be great if all voters looked at the paper to verify their votes, it wasn't crucial.

"It gives voters the opportunity to verify their vote, but it also gives election officials a meaningful audit trail to verify software vote tallies, and it's that latter purpose that has made the paper trial a no-brainer," she said. 
 
A critical question, however, concerns which vote is counted if there's a discrepancy between the paper and electronic records. The commission would leave this up to states to decide. Calingaert acknowledged, however, that left to their own devices, states often had significant disparities in the ways they handled election issues.

According to Pamela Smith, spokeswoman for VerifiedVoting.org, of the 26 states with paper-trail laws, only 12 require a manual audit of the paper and electronic records. Of those, only about one-third specify the VVPAT as the official record when a discrepancy occurs.

"If you have a paper record verified by the voter, why would you ever choose to use an electronic record that the voter hasn't been able to inspect?" asked David Dill, founder of VerifiedVoting.org and a computer science professor at Stanford University. "Obviously, we have had some success but have a long ways to go before all states have strong audit laws."

Congress has so far shown little interest in any of the commission's recommendations. A spokesman for Rep. Bob Ney (R-Ohio), chair of the House Administration Committee that oversees election reform, said Ney acknowledged the value of the commission's work but wouldn't support reopening election-reform issues while the Help America Vote Act of 2002 a wide-sweeping reform law passed after the 2000 election problems is still being implemented.

Dill said it made little difference whether Congress got involved in the paper-trail issue.

"A federal law would be very expedient, but we're prepared to fight it out state by state," he said.