House Election Law Committee
GP Room 158
33 North State Street
Concord, NH 03301
Mr. Chairman and Members of the Committee,
On behalf of Verified Voting, I write to express our measured opposition to HB 1821, which would require that ballot images and cast vote records be released on a central website within 48 hours of each election. Verified Voting is a non-profit, nonpartisan national organization founded in 2004 by computer scientists to strengthen democracy for all voters by promoting the responsible use of technology in elections. We applaud HB 1821’s intention to increase transparency in New Hampshire’s elections. However, publishing these digital records poses dangers to voter anonymity and public confidence that, in our view, outweigh any likely benefits.
To briefly review the key terms: “ballot images” are digital images of voters’ ballots as recorded by tabulating scanners; “cast vote records” are digital records, one per ballot, of how tabulation software interpreted the votes on that ballot. Neither ballot images nor cast vote records can be assumed to accurately reflect the actual ballots, although in practice they usually do. A traditional New Hampshire hand count produces nothing like ballot images or cast vote records. Publishing these records is a sociopolitical experiment, not an established open-government practice.
Publishing ballot images and cast vote records can facilitate vote selling or coercion by undermining ballot anonymity: these digital artifacts can be used to attest who voted for a certain candidate. Voters can write their names or other identifying marks on their ballots to be captured in the images. Voters can write in unique candidate names to distinguish their ballots and artifacts. Voters even can vote some contests in a distinctive pattern, a sort of digital fingerprint. Of course, as with many other attacks, vote selling or coercion relying on digital artifacts would be hard to execute on a large scale—we are not predicting disaster—but the possibility is inherent.
While it is possible that publishing digital artifacts could enhance public confidence, it may be more likely to backfire. Very few people are equipped to evaluate large troves of raw election-related data or claims about the data. Given this limitation, and the non-authoritative nature of ballot images and cast vote records, publishing them seems unlikely to boost confidence. Contrariwise, recent years have seen many wild claims, from across the political spectrum, of election fraud, often supposedly supported by highfalutin technical data analyses. Time after time, we have found these analyses to be rooted in misunderstandings and in some cases perhaps bad faith. Providing more grist for election fraud enthusiasts may not actually be dangerous, but it does not seem helpful.
After an election, state and local officials do wide-ranging work to help ensure the accuracy and trustworthiness of the results. We believe that despite its good intentions, HB 1821 tends to distract, and potentially even to detract, from this important work. We recommend finding this bill inexpedient to legislate. Thank you for your careful consideration
Sincerely,
Mark Lindeman
Policy & Strategy Director
Verified Voting