Two pieces of technology allow voters to cast their ballots reliably and securely. They are both hundreds of years old and begin with the letter ‘p’
The story goes that during the space race the Americans spent millions of dollars on pens that would work in zero gravity, whilst the Russians gave their astronauts pencils. This is actually an urban legend: both of the rival superpowers started out using pencils before concluding they were a fire hazard and switching to some form of specially designed pen. But true or not the merit that tale has been on my mind because I’ve been mulling the merits of the humble pencil.
The catalyst for this is Jill Stein. Having shown precious little interest in stopping Donald Trump before the election, the Green Party presidential candidate has raised large amounts of money to petition for recounts that aim to show that he did not in fact win a number of swing states. Which if nothing else prompted one of my favourite tweets of the entire election season:
The basis for Stein’s action seems to be the claim by some tech experts and election lawyers that in a number of swing states Clinton performed worse in counties that used electronic voting than those that stuck with physical ballots.
This notion is lent credence by the fact that during the election campaign someone – probably linked to the Russian intelligence services – was carrying out hacks designed to hurt the Clinton campaign. There is no reason to rule out the possibility they would have continued once polling day itself came around.
While all of this is true, it almost certainly doesn’t indicate fraud. When two journalists at FiveThirtyEight, Carl Bialik and Rob Arthur, looked at the gap between Clinton’s vote in counties with different voting methods they found that the disparity was likely the result of demographics rather than manipulation. You would expect the counties that used electronic voting to lean Trump anyway as they were whiter and populated by fewer graduates. In addition, the states in question are all in the midwest and Trump seems to have gained ground across that region and not only in states like that still use paper ballots exclusively. In short, there does not appear to be a mystery for which the manipulation of electronic voting systems might be an explanation.
Whilst it may not have happened this time, the use of electronic voting machines leaves American elections vulnerable to fraud. As this video by the tech journalist Tom Scott illustrates, it is practically impossible to secure such machines.
This is not to say that an election conducted with paper ballots can never be fraudulently swayed. Being involved in British politics gave me direct experience of that. During the first election campaign I ever played a decent sized role in, my side spent a lot of time trying to work out if our Labour opponents were casting fraudulent postal votes. We believed – and the police appeared to agree – this had happened in the previous election. But there was no repeat. In the intervening period, a new law had been introduced requiring voters to include a signature and D.O.B on both their application to vote by post and a form that was posted back along with the ballot. This simple step proved sufficient to mostly end abuse of this voting method. And even at its height postal vote fraud in the UK – to my knowledge – only affected local elections. Even under a system lax enough to draw criticism from international democracy watchdogs, getting hold of enough physical ballot papers to influence a parliamentary election without getting caught proved too difficult for potential fraudsters. Imagine, therefore, how difficult it would be to manipulate an election in American states that have a population in the millions.
Once you dispense with physical ballots and record votes only as data, these logistical impediments to fraud fall away. As Scott says:
once you have electronic voting, it can take as much effort to change a million votes as it does one.
But the problem with electronic voting is not only that it makes fraud possible. It also makes it impossible to disprove fraud. You cannot do a recount because there is nothing to recount. The kind of exercise that Bialik and Arthur conducted depends on their being counties and states that don’t use electronic voting in order to compare with the places that do. And even then, the two authors are forced to admit:
It’s possible nonetheless that the election was hacked, in the sense that anything is possible. (And the best hackers are experts in erasing their tracks.) Maybe hackers knew which control variables we’d look at and manipulated the vote in a way that it would look like it was caused by race, education and population driving different voting preferences. Maybe hackers didn’t manipulate the share of votes in individual counties, but rather the turnout, increasing the number of votes in counties likely to favor one candidate or another.
In this context, there is no justification for the use of electronic voting. It may get you results faster but that’s not worth it if the result at the end may have been tampered with. Even if it hasn’t been then the possibility that it might have been is corrosive of trust in the democratic process and a threat to the peaceful transfer of power.
So please Americans go back to paper and pencils. Not only is that more secure but also more reliable. Maintaining a system based around some of the most common items in the world is easy compared to maintaining machines. If a pencil breaks, you do not need an engineer to fix it. If you run out of them, then someone with no training on their first day at work is quite capable of going to Staples and buying some replacements. And because pencils and paper cost next to nothing they can be replaced as and when needed. By contrast, buying new machines is expensive so many electoral authorities put it off. Do this long enough and you wind up in the situation of Pennsylvania which on November 8th 2016 was still using touch screens:
In addition, there’s no need to teach voters how to use pencil and paper, hence there is less chance of ‘a hanging chad‘ style fiasco.
A number of states have wound up with the weird compromise of having voters cast their ballots using machines that then produce a paper copy of the ballot. Which – to borrow a phrase from Scott – makes the voting machine into nothing more than “the world’s most expensive pencil”.
If they wish to speed up the counting process, election authorities could always bring in machines to count ballots. This is less problematic than using them to record results because there remains a physical ballot paper that can be recounted by hand if there’s a question about the machine’s accuracy.
That exception aside Luddism is the best option when it comes to election. The ubiquity of pencils and paper speaks to their versatility. There is no need to invent any special technology to enable people to vote: a perfectly good tool for the job has been around for at least 500 years.