The Atlantic Council is about as heavy-hitting as thinktanks get – luminaries have included Brent Scrowcroft, Chuck Hagel, Richard Holbrooke, Susan Rice, Eric Shinseki, Jon Huntsman, James L. Jones and Anne-Marie Slaughter – but it has just put out a colossally wrong-headed report on the future of voting technology that goes against everything we’ve learned about computer voting systems over the past fifteen years.
The report, Democracy Rebooted, argues that since computer technology has taken over every other aspect of our lives, we should make greater efforts to adopt it in the electoral arena too. Potential benefits include greater accuracy, faster tabulation of results, and diminished opportunities for fraud. This applies not only to electronic voting machines, but also to voting directly over the Internet – a new frontier in voting technology that has been embraced with particular enthusiasm in Estonia. Without explicitly saying so, the report insinuates that if elections administrators and voters have misgivings about these technologies, it’s because they do not understand what they are dealing with:
Introducing electronic voting incrementally via pilot projects to build trust and familiarity enhances the opportunity for success and acceptance. By including political parties, civil society, and nongovernmental organizations in the decision-making process, governments can build a wide base of support, knowledge, and familiarity for these changes. Launching a robust public education and awareness campaign is another essential component to educating voters on the use of the system, communicating the goals and objectives for implementation, and providing public feedback mechanisms.
This is, in a word, nonsense.
Here in the United States, the introduction of computer voting machines in the wake of the 2000 presidential election is now widely seen as a disaster – because the machines were expensive, poorly programmed, prone to breakdown, and vulnerable to hacking and other forms of manipulation, as countless high-level official studies have shown. Many jurisdictions experienced a severe case of buyer’s remorse and switched to optically scanned paper ballots, generally seen as the most reliable and cost-effective technology because it is easy on the voter and leaves a readily accessible paper trail for a recount in case of doubts about the accuracy of the machine scan.
In other words, knowledge, familiarity and public education led to a resounding rejection of electronic voting, and with good reason.
The Atlantic Council report describes Internet voting as “a viable option” but completely fails to point out the most glaring and most troubling problem: that no computer scientist has yet figured out how to preserve the secrecy of the ballot and at the same time make Internet voting safe from potentially undetectable hack attacks or other forms of manipulation and fraud. A pilot Internet election in Washington D.C. in 2010 was intercepted with ease by a team of computer scientists from the University of Michigan; they were able to change votes undetected and also see who had voted for whom. Similar concerns apply to the Estonian voting system.
The Atlantic Council report suffers from what I identify in my new book as the fallacy of the technological fix – the notion that all it takes to end political corruption, intense inter-party rivalry, and a track record of electoral fraud is a fancy new set of machines. Public servants in the United States have fallen for this fallacy over and over, going back to the late nineteenth century. Whether the technology in question was lever machines, punchcard machines, or touchscreen computers, they have had to learn every time that technology alone does not solve the problems of a corrupted political culture; it merely moves it on to a new platform.
Actually, that’s a generous interpretation of what the report suffers from. A more cynical view would be that the Atlantic Council was swayed by, quote, “the generous support of Smartmatic”, a voting machine company that just happens to promote and sell electronic systems internationally and ran the Internet-based Utah Republican caucus ten days ago.
The Atlantic Council did not help itself, either, with its choice of author. Conny McCormack, the former head of elections in Los Angeles County, has been making the same flawed argument in favor of electronic voting machines (and against a verifiable paper trail for recounts) for years. It’s no mystery why elections administrators like her love computer voting and hate recounts: it’s all about the appearance of efficiency, without the hard work and manpower required to make sure the results in fact are an accurate reflection of the will of the voters.
McCormack’s track record is anything but reassuring: in 1985, while in charge of elections in Dallas, she presided over an electronic voting meltdown that raised significant questions about the outcome of a mayoral election and earned a reprimand from the Texas attorney general’s office, cited here. In Los Angeles, she pushed hard to ditch a functioning $3 million optical scan system in favor of a $100 million electronic replacement and became indignant when the California secretary of state overruled her following a series of machine breakdowns during the 2004 primaries.
In a felicitous piece of timing, The Atlantic Council report was followed almost immediately by a spectacular investigative piece published by Bloomberg Businessweek delving into the exploits of a professional election hacker who has used computer technology to wreak havoc with campaigns and vote outcomes across Latin America.
Andres Sepulveda, now serving a 10-year sentence in his native Colombia, describes in hair-raising detail how he went about his digital version of ratfucking – the Nixon-era bag of dirty tricks designed to mess with election opponents even before the voters get to the polls. He is quoted saying:
“My job was to do actions of dirty war and psychological operations, black propaganda, rumors—the whole dark side of politics that nobody knows exists but everyone can see.”
It’s a side of politics that the Atlantic Council report ignores at its – and our — peril.
Andrew- Thanks for a concise recap of problems that are apparent in the Report. I was a participant in the roundtable on this report last week representing the Silicon Valley based nonprofit election technology research institute: the OSET Foundation (osetfoundation dot org). We’re a group of ex-Apple, ex-Netscape, ex-Mozilla and others working toward publicly owned election technology through open source principles. This, as you aptly point out (that is, technology) alone will NOT repair the challenges of election administration, but we believe our work, rooted in evidenced-based election principles and focused on the VAST mandate (that elections are Verifiable, Accurate, Secure, and Transparent in process) can go a long way to shifting things from black box voting to glass box voting and shed light on the real problems. However, when I asked Ms. McCormack (about 1:45:15 into the 2-hour meeting) why “public ownership” and “open source” were not even considered in the innovation research portion of the study, we were informed that it “ended up on the cutting room floor for want to space and scope of the work product.” Whatever. But to your point, our approach is to recognize that the elections ecosystem is comprised of the “5 P’s” … People, Process, Platform, Policy and Politics. Our work trains on the “Platform” guided by the reality that all the technology in the world cannot address social engineering (People) or Politics. We believe our work may impact Process and we’d like to think our work might influence Policy (on our belief that code causes change). So, while focusing on the Platform alone will not solve the real problems you point out, we hope to shift the klieg lights off of the technology on to the real issues by making the black box a glass box. But it sure as hec won’t happen because we forego evidence-based election technology or increase the attack surface by rushing to pajama voting. It would be great to talk with you about this more. It was an interesting 2-hour round table, and there were some great points raised notwithstanding the omni-presence (hec, actual presence) of Smartmatic 😉 Thanks for your article.
Is Los Angeles County going to change to non-verifiable electronic voting? If so, when will the change take place? What can we do about it?