People make decisions every day, often without even much thought. And governments too must make decisions, but democratic governments generally rely on formal mechanisms, at least for
important decisions. An extreme view holds that the details of deciding matters less than that there is some well-documented procedure that is followed. Whether the decisions themselves are good or bad matters seems less of a concern and anyway that difference is too hard to judge. The legal doctrine of stare decisis seems to lean toward this view, insisting that legal decisions made in the past, no matter how wrong they might now seem, should be the guide for decisions made today and in the future. Perhaps such attachment to precedent is why we cling to plurality voting for elections.
We rely on elections not only for decisions, but also for assessing public opinion. And especially in that second role, it is hard to imagine plurality elections performing well. In essence, voters are asked to choose between two candidates. Not a great deal of nuance can be expressed in either response to such a question.
While we can hope that adopting a better voting system might lead to making better decisions, surely not just any other system would provide a big improvement. My own belief is that BAV would be a particularly good choice, however, and perhaps I have been able to convince others of that as well. BAV can be understood as an improvement over approval voting; and I would argue, a significant improvement. This article will again explore that improvement.
My motivation for this article came from voting in an approval voting election. You may recall that approval voters are asked to specify, for each candidate, either support or non-support; the candidate with the greatest number of support votes then wins the election. In this particular election, the candidates were works of art, mostly photographs but in some instances paintings or drawings. Not surprisingly, my thoughts turned to how the election might be improved by using balanced approval voting and my initial impression was a surprise. It seemed that BAV would not improve the election at all.
The difference between approval voting and BAV is that BAV asks voters to further refine their non-support opinion so as to distinguish indifference from opposition. But why, I wondered, would someone ever oppose a work of art? Surely it is possible to not like a piece of art, but the art causes no harm and some people probably would enjoy it. But in other respects, this photography contest seemed exactly like an election, so how can BAV not be an improvement over approval voting? I needed to take some time to think about this.
But I was distracted by an altogether different observation. This approval election was unfair, and not at all in a subtle way. The glaring unfairness came from the excessive number of candidates. Nearly 700 photos had been entered.
When I voted, there seemed to be no alternative to examining one photo at a time and choosing, for each image, whether to cast a vote for it or abstain. I could stop at any time, but otherwise I had no alternative to considering each images in the very same order as every other voter.
I leisurely examined the first few images, but quickly realized I had to speed things up. I clicked ever faster through the next fifty, stopping briefly to give a thumbs-up to one or two that happened to catch my eye during the few milliseconds the image was on my computer screen. But I persevered through perhaps another hundred photos before conceding that I lacked the patience to finish the task. There remained more than 500 photos that I never even looked at, even very briefly. It seems likely that other voters did much the same, though I suspect that many abandoned the project sooner than I had. I wondered how many people had ever bothered to vote in more than one of these contests.
What makes this voting especially pernicious is that each would review the photos in exactly the same order. It seems doubtful that more than a handful of the last 500 images were viewed by even a single voter. And it is no consolation that nearly every voter examined the first two or three pictures. Winning such a contest would seem to depend much more on a photo's position in the queue than on being an exceptional photo.
I pointed out this problem on a discussion board and roughly twenty people responded (in a discussion which has now been declared private). My own comment included the suggestion that for each voter, the order of viewing the photos should simply be scrambled so that each photo would have an equal opportunity to be evaluated by at least a few of the voters. Because all voting is through the internet, this proposal would really not be at all difficult to implement.
The responses to my comment suggest to me that the normal reaction to a problem is not to find a solution or way to avoid the problem, but only to find a way to avoid being inconvenienced by it. One commenter urged voters to find and vote only for only their own photos. Many suggested ways a voter might bypass great blocks of photos. But not even one comment (aside from my own) showed any concern for the fairness of the contest. In fact, quite the opposite was true; several comments pointed out that the contest was just for fun, so why worry. That may be true, but the unfairness makes these contests less meaningful and, at least to me, less interesting.
This example clearly illustrates that an excess of candidates can be a problem for the fairness of an election. This is the case no matter what voting system is adopted. Scrambling the order of candidates on the ballot seems to avoid this problem, and with computerized voting it can be easily accomplished. The need for paper ballots does make this more difficult, however. There are more complex voting systems that work around this problem while remaining both balanced and evaluative. Alternatively, it would not be hard to accomplish this scrambling with a network of computers (say, one or more at each polling station) to cooperate in issuing a unique customized ballot for each voter; a serious downside to this scheme is that such ballots would make tallying the votes more difficult and probably more prone to error.
Better would be to introduce a preliminary election, possibly one conducted over the internet, either with or without a more complex voting system, to reduce the field of candidates in the more important, final election to a reasonable number, say around a dozen. An election using BAV could remain quite manageable with even a bit more than a dozen candidates. And it would not take even a dozen candidates to provide voters with healthy variety of candidates to choose from. With even three different but viable political parties, successful Gerrymandering would become much more difficult; with five or more it might be impossible.
Elections with an excessive number of candidates do present a problem, but we have seen there are ways to manage this problem. So, let us set this issue aside and now consider the question of whether these photo contest are in some way so fundamentally different from political elections. Recall that the question came up in the context of pondering why anyone might oppose a work of art. But a vote of opposition need not be a rejection of it as a work of art, it might only express a judgement that the piece is unsuited for a particular photo contest.
Each contest at Fine Art America (and probably photo contests anywhere else) has rules and at least some mechanism for enforcing those rules. There is often a rule to limit the subject matter, but often there are other, often more technical restrictions. Each photo that is entered in a contest faces the possibility of being rejected for a failure to comply with one of these rules. Quite often that will be a matter of interpretation so someone acts as a judge to make these decisions.
But by using BAV rather than approval voting, the voters gain an opportunity to weigh in on this issue. And this is not fundamentally different than what would happen in a political election. Using BAV, a voter can decide to cast an opposition vote when, in the view of that voter, the candidate is unqualified for the office in question. With other voting systems, that decision is usually left entirely up to the discretion of politicians and courts.
So BAV provides ordinary voters with an extra measure of power and influence. If democracy is the objective, that surely is a step in the right direction.
Post Article Comment and Rate This Article
