Black Box Voting

Questions mount over New Hampshire's primary

By Lynn Landes
Online Journal Contributing Writer

Download a .pdf file for printing.
Adobe Acrobat Reader required.
Click here to download a free copy.

February 11, 2004�It's been all downhill for Howard Dean since he lost the New Hampshire primary by a significant margin. But, now questions are being raised about the security of New Hampshire's voting system in the wake of a recent analysis of the election results. It could add up to nothing, but it does underscore how easily technology can be used to sabotage the voting process.

Only one company, Massachusetts-based LHS Associates, Inc., programs and services all of New Hampshire's optical (ballot) scanners. Only two manufacturers, GOP-friendly Diebold and ES&S, provide all of the state's scanning equipment. And only Microsoft's Excel software tallies the results of all of those machines. It looks like New Hampshire has put all its eggs in one basket.

However, New Hampshire is unusual. Unlike many states that allow ballot-less voting, in 1995 the New Hampshire legislature passed a law that requires paper ballots in all elections. Has this law made New Hampshire�s voting system any more secure? Twenty percent of the ballots are hand-counted, but 80 percent are optically scanned�a technology that has a long history of being highly vulnerable to election fraud, which is documented in various reports as well as in the book, VoteScam: The Stealing of Democracy.

Recently some people have been asking if the 2004 New Hampshire primary was rigged.

Martin Bento published online an interesting analysis of New Hampshire's election results based on the voting systems used.* It's been getting a lot of attention. According to Bento, Howard Dean lost to John Kerry by only 1.6 percent when the ballots were hand-counted, 9.7 percent when ES&S optical scanners were used, and 14.7 percent on Diebold scanners.

That doesn't look good. On the other hand, pre-election and exit polls do seem to match the election results. Of course, polling organizations can be as partisan as think tanks, so their accuracy should always be suspect. The exit poll for the major news networks was done by (Warren) Mitofsky International and Edison Media Research (on whose board Mitofsky, the "father" of exit polling, sits). It's really a reconstituted version of the highly secretive and controversial Voter News Service. Curiously, Mitofsky's exit polls do not disclose their over-all results. Everything is broken down into subcategories. The people at Edison told me to ask Kathy Levine of ABC's World News Tonight for that information. Levine told me that I could "buy it" from Edison "like the major networks did." As if I could afford it. Well, there always seems to be something fishy going on with the networks and their election polls, something that's also documented in VoteScam.

Moving on. Others point out that geography was the real factor in how the New Hampshire vote count went down, that the more populated areas where optical scanners are more often used are closer to Massachusetts (Kerry Country). Whereas, the rural areas where hand-counts rule are nearer to Vermont (Dean Country). The problem with that analysis is that it's a 'guess' at best. The only way to really know for sure is to check the ballots. But, there seems to be little interest in doing that.

"We don't audit our elections," says New Hampshire's Assistant Secretary of State Anthony Stevens. Frankly, I don't know a state that does, although it sounds like a good idea. Stevens contends that the numerous hand recounts that the state has conducted over the years for contested elections, serve as a deterrent to machine tampering and ensure that the machines work properly. But, Stevens admits that he wants to examine the exit polling data to see if there's anything to Bento's analysis.

That begs the question, why not just check the ballots? Isn't that what they're there for? Reliance on questionable polling data is no substitute for examining the hard evidence of how people actually voted. State election officials say that they have never overturned an election due to a machine malfunction. But, that doesn't mean that it can't happen. And there are trillions of dollars at stake in this particular election.

Although New Hampshire law does not provide for audits, "it doesn't prohibit them either," says Assistant Attorney General Bud Fitch. Considering the fact that a handful of corporations control the optical scanners and vote tabulation system in the state, one would think that election officials would be more vigilant and less trusting.

Voters can't count on any certification process or pre-election testing to prevent vote fraud. Although touchscreen voting machines (DREs) have received the bulk of criticism lately, computer security experts are quick to point out that optical scanners can also be easily rigged to manipulate votes and remain undetected. Even a specially marked ballot can reprogram software as it's being scanned. Or, Microsoft's Access or Excel program could tally the results incorrectly. Off-the-shelf programs are exempt from any federal guidelines (such as they are) for voting equipment. And it doesn't help that all the software is proprietary (i.e., a corporate trade secret), although "open source" software is highly vulnerable to tampering, as well.

It doesn't take a vast conspiracy to rig an election. Just one person from LHS or Diebold or ES&S or Microsoft or an election official or some rogue programmer could rig a large part if not an entire election in New Hampshire.

And if that news isn't bad enough for the security of New Hampshire's current voting system . . . the future is in Internet voting, if Republican Governor Craig Benson has his way. It's the most insecure voting system in the world. Even the Pentagon canceled their online voting project for this year due to a scathing report from a panel of computer scientists. But, New Hampshire's governor is undeterred. The state has joined the National Student/Parent Mock Election project. Together with America Online (AOL), the nation's largest Internet provider, New Hampshire has joined other states across the country "to make it possible for participants to cast votes online from around the world if they wished, foreshadowing the way Americans will vote in the future," according to the project's website.

I've always viewed the people in New Hampshire as conservative folks who like to hold government accountable. Yet, from what I can tell, they're practicing the same faith-based voting that's going on in the rest of this crazy country. Oh well. There's one thing we can all count on. Questions will continue to plague American elections as long as voting machines are part of the process.

* See Martin Bento's analysis, Methodology and Code of New Hampshire Analysis.

Lynn Landes is one of the nation's leading journalists on voting technology and democracy issues. She is also the publisher of EcoTalk.org. Lynn is a former news reporter for DUTV and commentator for the British Broadcasting Corporation (BBC). Contact her at lynnlandes@earthlink.net or (215) 629-3553.

 The views expressed herein are the writers' own and do not necessarily reflect those of Online Journal.
 Email
editor@onlinejournal.com
 Copyright © 1998-2005 Online Journal. All rights reserved.