How to vote?
2017 is quite the election year in Europe. Holland, Germany, France and maybe Italy, 4 E.U. founder states, are going to cast their ballots to choose their new rulers! As they have done since the end of World War II, citizens (men and women) will go to polling booths to express themselves via the democratic process of voting. Out of several possibilities, they will pick their favorite candidate or none.
I have been voting (because I am not underage anymore, yay!) for 5 years now, and I have to admit that it is the first time I really try to build my own opinion on the matter, and not this weird mix of my parents’, Le Monde and my friends’.
But how? And really what I mean by that “how” is two questions:
- How do I establish my own opinion on each and every single proposition (in the case of a presidential election, each candidate)?
- How, out of all these opinions I would now have at my disposal, do I select one single proposition, one single candidate?
These two questions have, for now, not satisfying answers and the second one exists only because of the voting system we are currently using. What are the different leads to explore in order to enhance the current system ?
I realize that, here, the word “opinion” might not be clear at all, and I will try to define what I mean by that in this article. To have an opinion on a proposition is to attach it two values (not necessarily numbers) on two aspects: its ground and its applicability.
Example: If the proposition is to ban Muslims from entering the United States, the ground is 1/10, very bad, but the applicability is 8/10, reasonably applicable.
Its ground is its ethic, its moral, the reasons that motivate it. Its applicability is whether it is going to cost much, whether it is humanly doable, whether we (as a society) have the needed technical skills or scientific knowledge.
By extension, an opinion on a candidate is the collection of the opinions on all his/her propositions.
How do I establish my opinion?
One easy way to answer that question is the following: “Read the platforms!”. Even if it seems like an easy answer, it’s at least the beginning of it all. But let’s be honest, do you know someone (I am talking non-professional here) who has had the courage to read carefully the platforms of every single candidate for the last election? This is a tedious task. It requires time and a lot of focus to understand sometimes complex reasonings or even sentence structures.
Let’s say that you try to read at least the platforms of the leading candidates (leading in the polls for example). It is still a very difficult exercise to come up with clearly defined opinions on every proposition. 4 factors can contribute to it.
The first is non-expertise. Indeed, you can not be an expert on every topic related to an election. Economy, finance, diplomacy and law, for instance, are central to one’s platform but they are very technical fields, for which everyone can have a global opinion, but hardly a strong one when it becomes a little deeper.
The second is lack of trust. How can you trust the numbers presented in the propositions to justify it? Even when the sources are stated, and you give them credit, how can you be sure that these numbers are the correct ones to understand the big picture of the issue tackled by the proposition (absolute vs dynamic for example)? Being aware of the context and the correct point of view to adopt is, most of the time, out of reach for non-experts.
The third is prejudice. Clearly, your personal history, the media, your relatives will have already influenced you on each candidate’s platform (be it negatively or positively) when you will find time to read it. It is perfectly normal and even sane to have public debates. However, it then becomes very difficult to keep a clear mind when screening platforms widely criticized or acclaimed.
The fourth is bias. The definition I gave of ground is intentionally not very satisfying in that it leaves out the citizen’s subjectivity. For example, if you are rich, you might be tempted to give a high value of ground to a proposition suggesting to lower tax on rich people, because you would benefit from it. At this point, it is important to ask yourself whether you vote for yourself or for the entire society. In other words, how do you calibrate your ground?
It seems, therefore, natural that everyone needs trusted sources to have an opinion on technical propositions. Ideally, you would need to check the validity of these sources, challenge them and see how often they are wrong (and whether they admit it and apologize for it). Obviously, it is extremely difficult to gather information from only trusted sources: even Facebook is struggling to find the correct definition!
All these drawbacks are related to the everlasting problem of the democracy in which the citizens are not correctly informed (“Democracy is the worst form of government except for all the others”, W. Churchill). It is tempting to give in to anti-democratic solutions. A quick reminder of why democracy is so important is explained in this essay by Hippolyte-Adolphe Taine (even if you are not interested in the subject, you might want to read it since it’s sometimes hilarious to see how condescending he gets, remember it’s 1889). Basically, it says that it is immoral to deny a man its right to live or die and therefore he should be able to choose the ones that lead him into a war.
Rather than this kind of extreme solutions, like technocracies, we should focus on trying to make democracy more user-friendly. This goes through restoring trust in politicians, a better education oriented towards understanding politics and their “tricks” (very difficult to enforce because of political connotations) and firing an interest for public matters. Participatory democracy is addressing these types of demands.
Another solution, considered in this Medium article, is to organize exams to validate everyone’s right to vote. But it is impossible to envision an exam validated by all participating political parties and the logistic would be too complicated.
To fight against fake news and the phenomenon of lack of trust (and somehow non-expertise), we could also imagine some penalties for candidates who spread fake news either via their platform or in speeches. It would be up to an apolitical commission to judge each case. The penalties could be in public television appearance time or fines. Le Monde is trying to put together a tool that might look like what I am advocating for online medias.
How do I compare my opinions?
You now have at your disposal opinions on (ideally) every candidate. Bear in mind that depending on the process that led to establishing an opinion on a proposition, you might enrich it with another value which is credit or trust. Your goal is now, in the current system, to select the candidate who has the best platform according to you. At least, it is your goal if you don’t consider strategic voting, but I will come back to this issue later.
You find yourself facing a very difficult problem. How can you satisfyingly compare two different propositions, each having three distinct values attached to it (for math geeks out there, how do you satisfyingly compare two points in 3D space)? Do you take the best (or the worst) out of the three and compare on that? Do you compare first ground, then applicability, then credit?
Let’s say that you find such a method. It is actually not enough because platforms can not always be opposed proposition by proposition. They generally have many propositions related to a broader topic like for example education. You, therefore, need a way to compare two collections of opinions on a broad topic. How can you do that? Do you take the most representative propositions on each side and compare them?
But anyway, let’s say again that you find a way to compare two candidates on a particular topic. How do you end up comparing them as a whole? A first simple answer might be to say: “Well, you just select all the main topics addressed, you compare them one by one, and the candidate with the most positive comparisons is your choice.” But then again, it is not satisfying. For example, as an individual, you might be more sensitive to immigration-related issues or consider that economy is the most important topic in a platform (even though you are not an expert on economy…). How, then, do you perform this weighted comparison?
Moreover, yet another factor comes into play when it comes to comparing candidates, personality. Indeed, you might be one of those who think that personality and charisma are important when it comes to choosing someone that will not only govern us but also represent us. How do you embed that in the equation?
All of these issues are due to the fact that it is very difficult to aggregate rankings from different sources in a way that may satisfy some reasonable criteria. This impossibility was first unveiled by Condorcet and formalized by Arrow and it is linked with a voting system in which everyone has to select a single candidate.
It is also due to the fact that propositions part of a single topic might interact between each other, and that topics might also interact with each other in a single platform.
My point here was actually not to explain what is, according to me, the best way to vote. You might find that, (again !!) not satisfying, but I proved that it is close to impossible to have a definite methodology.
All in all, I think it is just important to remember the 4 factors that will blur your opinion building in order to avoid suffering from them. Eventually, your decision will be once again a weird mix, but at least a self-conscious weird mix.
I personally will try to read most of the platforms and check a few numbers here and there to see if I should be wary or not (I will be if they are not documented). I will also try to watch different media interventions. This will give me a general opinion and I will then compare some propositions on matters that I hold as important and on which I think I can have a critical thinking. My proposition to proposition comparison will be mainly based on ground because I think that applicability, if not too low in my opinion, can always be worked around, or that we at least can find a compromise.
A nice way to alter the complexity inherent to comparisons might be to change the voting process. It might also be interesting to consider such a change to avoid Condorcet paradox which is the reason behind strategic voting. It states that it is impossible to elaborate a correct process of aggregating rankings.
In this article (in French), are explained different alternatives that might one day soften the problems I discussed here. The main point is that instead of expressing a single vote, you detail your opinions on each candidate by ranking or grading them. This is also suggested by Michel Balinski, a mathematician who explains (mathematically) how even in a one-vs-one election, the current system can turn out to produce a not-optimal result.