August 02, 2004
Because Of The Not-So-Wonderful Things It Does
Geekgirl2 comments at Musings & Ephemera:
Well it does mean that everyone votes (or most people) BUT does it mean their vote counts for something? I'm not so sure about that!
wrt Tim's post (We're off to see the wizard, supra) about compulsory voting down under:
Well, I dunno. I appreciate that voter participation is a key part of democracy, but punishing folks for not voting seems a little harsh. Of course, I'm not sure that bribing voters is the right approach, either. Is a "compulsory" voting process a "good thing" for democracy?
My answer to Tim's question is an unequivocal "no", and my answer to Geekgirl2 is a highly equivocal "maybe".
First, it strikes me that being compelled to vote, even if you have the choice to spoil your ballot (or, say, create a Hudemonic ballot stir-fry), smacks of banana republicanism and/or totalitarian régime with 95% voter turnout and 110% in favour of whoever The Man happens to be. (After all, whoever you vote for, [the] government wins.)
Second, it says rather scary things about the state of democracy if people have to be forced to turn out at the polls. Not only are you admitting that your demos don't care, but you're trying to fake legitimacy. See the previous point.
Third, on a strictly "what gets us the best decision" basis, it's by no means clear that having people who haven't thought enough about the issues that they actually want to express an opinion go and express one is a good thing.
Think of it this way: you've got people who have an opinion, and each of them is expressing a signal. The purpose of your electoral system is to try to aggregate these signals in some manner (and I'll be among the first to argue that first-past-the-post is an utterly stupid way of doing this, and yes, yes, I know all about Arrow's theorem thank-you-very-much) and arrive at some sort of a decision representing the consensus, majority, or plurality view.
You've then got people who don't care what the outcome is: in a system where they're not compelled to turn out, they'll generally stay at home and thus not express any preference (well, until the bums, whoever they are, take office, and promptly get slagged at the pub, but whatever); in a system where they have to turn out or get punished, the best you can hope for is that they're going to transmit a random noise which will cancel out. If you're unlucky (and I would venture to guess that this is more likely than not), they're going to align their signals in some way [say, by who has the flashiest campaign ads, or who's top on the ballot, or whatever] and swamp out the signal that would otherwise be transmitted by the people who care.
In short, forcing indifferent people to turn out means that you drastically increase the amount of noise in your electoral system, reducing the quality of the results you get. You could, if you were inclined, try to model this as a transmitter/receiver/noise problem, and Tim might even be so inclined — that is, before he thwaps me upside the head for completely misunderstanding communication theory.
Am I an elitist bastard running-dog lackey etc for believing that only people who actually care should make the decisions? Sure, and proud of it, thank you very much. And failing a fool-proof way to determine who actually cares, the best approximation we have is by seeing who shows up on polling day to cast their ballot, without coercion.
Now, on to the question of whether the vote counts for something. As the number of people voting rises, the chance that your vote will be the decisive one falls (you can, of course, subdivide the voting populace, much in the way that the US Electoral College does or the 308 ridings in the Canadian House do). If you force more people to show up and cast a ballot, your vote will count for proportionately less and less.
Archive URL for this entry: http://www.pointsofinformation.ca/archives/individual/2004/08/02/cdjones_because_of_the_notsowonderful_things_it_does_240.html
TrackBack URL for this entry: http://www.pointsofinformation.ca/poi-ping.cgi/211
