Don’t Touch

When you want to scream in print, you use all capital letters. I’ve never wanted to scream any message more than this one: ARIZONA MUST NOT CONVERT ITS VOTING SYSTEM TO TOUCH-SCREEN MACHINES, BECAUSE THEY CAN’T BE TRUSTED.

I wish I felt better getting that out, but I’m afraid it’s going to take a lot more than screams from a dedicated American voter who has always thought her vote counted. I am dismayed to confront the very real possibility that my vote could be rigged or hacked or changed by a machine with “severe security flaws.”

Recent lawsuits and investigations are singing louder than a canary in a gas-infused mine. They signal that Arizona may be on its way to an election crisis, and the bottom line is very clear and ominous: If Americans can’t have faith that their votes do indeed count, the very foundation of our nation crumbles. Phony or rigged elections might simply destroy our democracy.

Arizona isn’t the first place in the nation to confront this issue head on a half-dozen other states are also on alert but it is one of the important places. And when I tell you how bad this problem is, you’re going to wonder what in the hell Secretary of State Jan Brewer is thinking.

Since the 2000 federal election introduced us to “hanging chads” and mispunched paper ballots, the nation has been alert to the notion that our voting system might be obsolete. In 2000, Arizona still used punch cards in 10 counties, and Brewer initiated a program to update them. Commendable. Ultimately, each Arizona county transitioned to “optical scan” voting the kind Maricopa County voters have used for years. Wonderful.

But then there was a significant wrinkle: Congress worried that impaired voters blind or handicapped couldn’t vote without assistance. So, in 2002, it passed the Help America Vote Act. Among other things, the act demanded that by the 2006 mid-term election, handicapped voters had at least one machine in every precinct that they could use without outside assistance.

As the Wall Street Journal has put it, state and county officials “stampeded” to spend the $3 billion allocated by the act and hurriedly started buying new voting machines.

In Arizona, Brewer worked with county election officials and finally decided that for impaired voters, Arizona’s paper ballot system would be replaced with touch-screen machines that record votes on a disk. The difference between the two systems is striking: Paper ballots leave a physical trail, a way votes can be recounted or verified, while touch-screen machines do not.

“The introduction of these machines into Arizona’s electoral process will not only create serious concerns, but also real risks about the accuracy of election results,” attorneys for Voter Action told the Legislature. “These machines have a track record of inaccuracy and unreliability, with such problems as unrecorded votes, switched votes and phantom ‘over-votes.’”

Earlier this year, Voter Action filed suit in state court to prevent the use of touch-screen machines for impaired voters in this fall’s election. It was the ninth lawsuit of its kind in the nation. And it put Arizona on a list of concerned states that includes California, New York, Maryland, New Mexico, Illinois, Pennsylvania and Florida.

Chuck Blanchard, a former state legislator who is an attorney for Voter Action, told me they fear that using these machines for impaired voters this year will be “the first step to using these machines for everyone” down the line.

That, of course, raises the stakes considerably. I’ll argue that every single vote is extremely important. Therefore, if even one machine in a precinct is hinky, that is one machine too many.

Ironically, Blanchard adds, the touch-screen machines that Brewer so loves “don’t do what they’re supposed to give independent access to disabled voters.” As proof, he submitted an affidavit from a blind voter with more than 30 years of professional experience as a computer scientist and election engineer who found the touch-screen system inadequate.

“It does not accomplish its goal, it has huge security problems and it’s hack-able,” Blanchard said. All three problems were stressed in his lawsuit against Arizona.

The suit received national attention, as have the ongoing problems of touch-screen machines across the country.

Brewer told the Wall Street Journal that the lawsuit’s allegations were “unsubstantiated.” And she screamed to high heaven in the local press that the suit was an attempt to limit voting by handicapped citizens, which is so absurd on its face that it’s hard to believe she would stoop so low.

On July 24, Superior Court Judge Barry Schneider dismissed Voter Action’s suit against the secretary of state, denying the group’s motion for preliminary injunction. Presuming Voter Action does not pursue further litigation, the state may use touch-screen machines for handicapped voters beginning this November.
Meanwhile, Jan Brewer seems to be one of the few election officials in the country who doesn’t seem to grasp the problems with the touch-screen machines.

She has told The Arizona Republic she has no concerns. “We have consulted with so many people, I feel very, very confident… that everything is secure,” she has said.

But the issue has plagued her, with protesters of the new machines even disrupting her announcement that she was seeking re-election this November.

Brewer, Arizona’s highest elected Republican official, called the protesters “anarchists” and told the Republic, “I have instituted procedures and rules and security measures that in effect, if followed, will not allow any miscounting of votes…. If they follow my procedure manuals, there will not be an issue.”

I count three major equivocations in that sentence. When was the last time you wanted your vote equivocated? Never? I’m with you.

The state of Maryland is with you, too. There, the Republican governor has stopped the use of touch-screen machines. So has the state of New Mexico, where the Democratic governor persuaded the legislature to require verifiable ballots. So has the state of Illinois, which is rethinking the use of touch-screen machines after using some of them in last March’s primary election. There were so many technical and human errors with their Sequoia Voting Systems equipment, that it took a week to tabulate results. North Carolina also is unhappy with its Election Systems & Software Inc. machines, whose software malfunctioned, leading to uncounted votes in the 2002 general election. And, if you’re a concerned voter, Arizona’s plan will give you the heebie-jeebies.

I almost had a heart attack. The implications of this are pretty astounding.” That quote comes from one of the nation’s most acclaimed and respected computer scientists, Professor Aviel Rubin from Johns Hopkins University, about new flaws recently uncovered in the very machine Brewer is purchasing for Arizona the Diebold Election Systems touch-screen machine.

In 2003, Dr. Rubin completed the first in-depth analysis of security flaws in the code for that machine, made by the Ohio-based company that is one of the nation’s three voting machine manufacturers.

Unlike Arizona’s current voting machines, where voters mark a ballot that can be read by a computer and leave a paper trail that can be double-checked, the touch-screen machines leave no paper trail and only record the vote electronically. There is a flimsy “paper” roll in the machine much like credit card receipt paper that has all kinds of problems, including the fact that it disintegrates when touched. So, for all practical purposes, nobody in the business considers it a legitimate paper trail.

That means there is no way to double-check that the machine is correctly counting the votes, neither does it detect any inherent flaw that would permit rigging the vote.

Dr. Rubin’s test of the Diebold machine found “serious security vulnerabilities.” Specifically: “Voters could trivially cast multiple ballots with no built-in traceability, administrative functions could be performed by regular voters, and the threats posed by insiders such as poll workers, software developers and janitors were even greater.” His concerns have since been substantiated by five additional major tests.

Yet, he notes: “Vendors, and until very recently most election officials, have continued to insist that the machines are perfectly secure. I do not know of a single computer security expert who would testify that these machines are secure. I personally know dozens of computer security experts who would testify that they are not.”

Those words should ring loud and clear in Arizona. Dr. Rubin is only one of the experts who filed affidavits in Blanchard’s lawsuit to bar the Diebold machines from the upcoming election.

I am certainly no computer whiz, but three things in Dr. Rubin’s statement particularly struck me:

• I wondered where he found the Diebold “code” to test in the first place, because that would presumably be secret, proprietary information. He reports that he discovered the code “on an unprotected internet site.”
• He also learned that the same “pin” number was being used nationwide by Diebold supervisors, meaning that “an unauthorized person could use it to gain access to supervisor functions on the voting terminal.”
• Furthermore, he notes that studies show it doesn’t even take a pin number to access the system “to change ballot definition files and election results.” All it takes is a “widely available Microsoft Access database program.”

Nobody has to be a computer expert to realize that any one of those problems should be enough to say or scream STOP.

And then there’s what computer security investigator Harri Hursti did to the Diebold system. Last year, he was asked by an elected official in Florida’s Leon County to try to hack into it.

He did. Hursti demonstrated that the machines were vulnerable to tampering, even to the point of installing different software. Diebold officials told The New York Times that there was “theoretical security vulnerability,” but added, “The probability for exploiting this vulnerability to install unauthorized software that could affect an election is considered low.”

One Diebold official tried to erase the concerns by suggesting nobody would ever do such a thing. “For there to be a problem here, you’re basically assuming a premise where you have some evil and nefarious election officials who would sneak in and introduce a piece of software,” said David Bear, a company spokesman. “I don’t believe these evil elections people exist.”

I wish I didn’t believe it. But I’m apt to side with the co-chairmen of the bipartisan Commission on Federal Election Reform, former Secretary of State James A. Baker III and former President Jimmy Carter, whose 2005 report warned: “Software can be modified maliciously before being installed into individual voting machines. There is no reason to trust insiders in the election industry any more than in other industries.”

We have good reason to be concerned, because a lot of people have been suspicious of our voting system since the 2000 election fiasco in Florida.

In addition, there is a growing body of evidence coming not from political parties, but from independent scientists that George W. Bush’s 2004 re-election victory may have been rigged.

A new book by a University of Pennsylvania professor who admits that he “despise[s] the Democrats” is titled Was the 2004 Presidential Election Stolen?. Author Steven F. Freeman told Rolling Stone magazine that it is “impossible” that exit polls in three battleground states showed John Kerry winning the election, while the actual votes went to George W. Bush. He decided to examine the raw polling data because he “was mystified about how the exit polls could have been so wrong.”

“When you look at the numbers, there is a tremendous amount of data that supports the supposition of election fraud,” he concluded.

Dr. Freeman isn’t alone. Last January, a team of mathematicians from the National Election Data Archive took a second look at the actual vote tallies in the 49 Ohio precincts where exit polls were conducted. In nearly half those precincts, there were widespread discrepancies, all of which favored Bush. In one precinct, exit polls suggested Kerry received 67 percent of the vote, while the certified tally gave him only 38 percent. The statistical odds of such an extreme variance are about one in 3 billion. In its report, The Gun is Smoking: 2004 Ohio Precinct-level Exit Poll Data Show Virtually Irrefutable Evidence of Vote Miscount, NEDA concluded the results were “completely consistent with election fraud.”

Ohio is significant because it is the state that settled the election, giving George W. Bush the victory. Ironically, Ohio is also the home state for Diebold.

Ask one of Arizona’s most conservative legislators, Mesa Republican Karen Johnson, why she’s so insistent that we don’t abandon a paper trail in our elections, and she will tell you that “Florida really had a big mess” and that “Ohio had problems.”

Johnson, one of the Legislature’s extreme philosophical right-wingers, is the key sponsor of a bill mandating that Arizona always have a verifiable way to recount an election. She has gathered an entire bevy of lawmakers Republican and Democrat to sign on to a bill that will ensure an accurate vote count.

“We have just got to make sure we have integrity in our system voter confidence is imperative,” Johnson says. But while that’s as obvious as the day is long to her, she has stumbled upon plenty of roadblocks. “I have seven children, and this is harder than birthing a baby,” she says.

First, she worked with election officials from all 15 Arizona counties, getting some on board while others had concerns. Ultimately, she persuaded everyone to agree.

“I’ve never worked on a bill with so many people across the board, so many working for the right reasons,” she says with pride.

Just when she thought the bill was a go, more problems popped up, and Johnson says they came from the secretary of state. Brewer’s office, however, contends that they “haven’t weighed in on this bill.”

Ultimately, the Legislature passed a version of Johnson’s bill that requires a durable paper record of votes, which would allow the state to manually audit an election to assure accuracy. She expects that it will be signed into law.

My talk with Karen Johnson was one of the few really encouraging moments as I explored this subject. Although I could not reach Brewer, her deputy, Kevin Tyne, answered my call. He is a very nice man who I am convinced sincerely believes what he is saying. And he says that there’s nothing to worry about.

“A lot of the scare about touch-screen machines is that they can be hacked, but that’s only true if you [don’t] have all the security safeguards in place, and we have the security safeguards in place.” He says Brewer has been working on this for two years and has instituted “tons of security measures we’re very proud of.”

I know I sound like the canary in the mine again, but if you have to develop “tons” of security measures, doesn’t that suggest that you have a security problem?

So, let me say it again, this time without raising my voice: Arizona should not jeopardize its elections by converting to a touch-screen voting machine not for anyone. The sanctity of the ballot is too precious to throw away.