This article was first published by WhoWhatWhy
Americans may have a reason not to trust New Hampshire as the barometer it has become for the viability of presidential candidates. The reason? Doubts about the security of the state’s voting system.
Historically, the New Hampshire primary has been extraordinarily significant in the selection of presidential candidates. A good performance here can catapult contenders to the front of the field while a bad one has brought the campaigns of one-time frontrunners to a screeching halt.
The people of New Hampshire are aware of the important role they play. Taking into account the Granite State’s size and population, no other group of Americans gets to know the presidential candidates better. In this election cycle, Republican candidates visited New Hampshire more than 300 times and spent nearly 600 days there.
But in spite of the state’s significance, it seems that New Hampshire’s government is not doing all it can to ensure the integrity of their “first-in-the-nation” primary.
The state principally uses AccuVote optical scanners, which means that voters fill out a paper ballot that is then scanned and counted by a computer. However, a few jurisdictions – mostly smaller ones – still count ballots by hand. In 2008, there were discrepancies in both parties’ primaries. Tallies for Hillary Clinton on the Democratic side and Mitt Romney on the Republican side were consistently higher when machines did the counting, and lower when humans did. In 2012, Romney again got a bump in machine-counted districts.
These discrepancies may have perfectly legitimate explanations. In 2008, a partial recount on the Democratic side uncovered nothing suspicious. On the Republican side, Romney and the big business policies he promoted might be expected to do better in cities than in rural districts.
Without Checks, Foul Play Cannot Be Ruled Out
Yet without further investigation foul play cannot be ruled out. After all, New Hampshire often holds the keys to the kingdom for nominees, and voting machines have notoriously bad security. With billions of dollars now being spent on presidential elections — and with the ultimate power in the land at stake — would it be surprising if a campaign or a Super PAC or other interested party tried to gain an advantage through tampering with voting machines?
And that’s why concerned citizens of the Granite State are petitioning their government to make some simple fixes that could ensure this important primary is won by the candidate who actually receives the most votes. However, the state government does not appear to be responsive to the concerns of its citizens. Why?
“We have a duty to our state, the country and even the world to get the count right in New Hampshire,” Deborah Sumner, who has been an activist on this issue for nearly a decade, told WhoWhatWhy. “Our system lacks the checks and balances that the people of New Hampshire deserve as well as the candidates.”
A Simple Solution
The solution, Sumner and others claim, is simple. If all jurisdictions with voting machines would do a parallel hand count, then there would be no doubt about the outcome.
Wally Fries, a former election moderator for the city of Danville, pioneered this type of verification method. He explained to WhoWhatWhy that the parallel hand count simply consists of cross-checking the results in a few select races. If the hand count for this sample matches the machine result, then election officials can be virtually certain that no foul play is involved.
“All machines are subject to error,” said Fries, who managed engineers for a living and therefore has a lot of expertise in the area. That is why, during his 25 years as election moderator, he set out to create a verification protocol that is virtually foolproof.
“I wanted to create a mechanism so that voters could have complete confidence,” he said.
In fact, all election moderators in New Hampshire have the right to order this type of verification. Or at least they used to, according to a 2010 directive that gave moderators broad discretion to initiate procedures they deemed necessary to ensure a fair count.
Recently, Sumner and others charge, state officials have discouraged some jurisdictions from using the parallel hand count.
Last year, residents of the city of Keene even sent a letter to Stephen LaBonte, the assistant attorney general in charge of election law, asking if the 2010 directive was still valid. LaBonte never responded. He also did not reply to an inquiry from WhoWhatWhy.
This unresponsiveness is one of the many reasons Sumner lacks faith in the state’s voting system. She feels too many officials value the convenience that voting machines provide over the accuracy of a hand count verification. She does have faith in the moderators, however.
A Firewall against Tampering
Elected by their local jurisdiction, New Hampshire moderators play a crucial role in the primary process, which represents an important stream of revenue for the state. Moderators put in long hours during election season and get paid next to nothing, Peter Webb, a moderator and attorney, told WhoWhatWhy.
Prior to the election, they test the voting machines and count the ballots received from the state. “Election Day is at a minimum a 6:30 AM to 9:30 PM shift, without break, for the volunteer election workers,” Webb said. Among their responsibilities are ensuring that all laws are adhered to, that the seals of the voting machines haven’t been broken, that only registered voters with the right type of ID cast ballots, that write-in votes are counted and, at the end of the day, that the numbers add up and that the ballots are boxed, sealed and safeguarded.
“We have conducted random arbitrary hand counts in the past and in each case determined that our machine count was accurate,” said Webb. “I don’t believe that [after their long day] the election officials have either the physical energy or remaining cognitive capacity to then hand count all the ballots to verify the results. An army of fresh volunteers might… do so, but practical realities such as the availability of volunteers, time constraints, human error, purity of the process, the need for finality can make that impractical.”
Fries, however, noted that even in his large jurisdiction, the parallel hand count never took more than an hour and it allowed all involved to go home knowing that the election had been properly conducted.
Everyone WhoWhatWhy spoke to for this article praised the many Granite State election volunteers. Sumner pointed out that they all swear an oath of office on Election Day. “The computer has not taken an oath and the people programming it have not,” she added.
Without the cross-checking process of the parallel hand count, Sumner believes the New Hampshire vote can be manipulated.
“It’s easy to exploit a New Hampshire election,” she told WhoWhatWhy. For example, memory cards in the machines could be switched out or software could be written in a way that allows an AccuVote machine to function perfectly during the pre-election test but then to skew the results of the actual voting.
Sumner likened this to the Volkswagen software that allowed cars to perform one way during emissions tests and another when on the road.
She would also want ballots to be made available to the public before the actual voting but Fries does not agree. In the smaller jurisdictions in New Hampshire, pre-balloting would make it too easy for a large employer to instruct employees to vote a certain way and mark their ballots so that their boss would know they had done as told.
Barring a parallel hand count, Sumner says she cannot be certain that the vote will be clean. To do her part, she has alerted the campaigns of Hillary Clinton, Bernie Sanders, Donald Trump and Rand Paul of potential problems. Only Clinton’s people got back to her. Still, Sumner hopes that the others are putting pressure on New Hampshire officials through back channels.
When asked what would be a red flag for her that the voting had been rigged, she said: “If Bush pulls an upset.”