It was an exhausting yet amazing experience to serve as an Election Officer today. I have gained a profound respect for the folks who operate and manage the election process with integrity and good intention. Yet in my experience today I also gathered some anecdotes to support my general skepticism about the robustness of the electoral process. Don’t get me wrong – our crew did a great job today, votes were cast, and the process went smoothly. But I nonetheless left my polling station this evening with a less than perfect faith in our voting process. Why? There is the slight concern about the adequacy of our process for verifying a voter’s identification, and there’s the lack of clarity about what we Election Officers were actually swearing/affirming (the oath/affirmation I signed last week was different than the one I was given this morning), but the most disconcerting element of this election process is the electronic voting system.
You see, a voter makes selections for various races and ultimately casts a vote on the touchscreen of a computer. The computer records the vote tally for each race on a small USB drive, much like the one many of us wear around our necks or on a keychain. The problem with this system is that there is no paper trail recording each voter’s ballot. There is no way to go back and recreate a particular ballot at a particular machine for review purposes. Furthermore, there is no way to verify that if a voter selects "A" the computer actually records "A" in its memory. All we have, at the end of the night, is a total count of votes for each race. We can’t go back and actually look at the ballots of individual voters.
But with a system that creates a physical voting record – a paper ballot or a computer system that generates a paper trail, for example – we can go back and look at each ballot, set them out on a table, and count them by hand. We can verify that X number of people voted for SMITH, and Y number of people voted for JACKSON, because we can sort the ballots that the voters themselves completed. On a computer, I can’t do that. I have to trust that the computer’s count is correct. I cannot verify the numbers the computer gives me with any external count, for there is nothing physical, nothing tangible for me to count.
What I can do is verify that the number of people who checked-in at the registration books with the Election Officers matches the number of votes cast on the computers. At our polling station, the number of people who signed-in matched the number of votes cast on the computers. We know that the raw numbers match up – no ballot was lost – but the main problem remains: voters cannot verify that their selections are recorded correctly on the computer, since no hard copy, no physical specimen is created. I think this is a problem. I’ve had enough odd computer and ATM errors that make me doubt the perfect integrity of a computer system and a silicon microchip.
Again – this was a great experience and I am grateful for the people who make this process go so well. But in our quest for faster election results for our 24-hour cable news cycle, I fear that we might be creating an opening for technological failures – or worse, technological tampering – to do significant harm to our democracy.