> I believe that simplicity is a must for any system that invites participation from all of society and just voting against someone is a lot simpler than ranking, or quadratic votes or whatever.
It's a pity there isn't an example of what a ballot paper would look like under the proposed system, but I'm imagining that next to each candidate there would be three boxes representing: approve, neutral, disapprove.
I suppose that's easier than a ranking system, because there might be lots of candidates that most voters haven't heard of (or don't have a strong opinion of) and it can be hard to put them into an exhaustive ranking.
The temptation, though, will always be to approve of the one candidate you like, and disapprove of all the rest (and candidates will encourage their supporters to do this), so the system will quickly devolve into plurality-style voting anyway, except the votes will be harder to count because there will be 3^n different ways of marking the ballot, rather than just n (where n is the number of candidates).
In any case, this looks a lot like Score Voting, with possible scores of 0, 1 or 2 (which I think is mathematically the same as -1, 0, and 1). Again this has the problem that you can't just count the number of votes in each sorted pile, and instead have to have n different running tallies, with no checksum at the end to make sure the tallies all total to the number of votes cast, for example.
This isn't a problem for a computer algorithm, but anything that encourages people to outsource their democracy to a piece of firmware, commissioned and maintained by one ruling party, is just asking for trouble. Even if the firmware works as claimed, you still introduce the problem that FUD can be used to attack a legitimate result.
Don't get me wrong, though, I'm not against voting reform, and I'm definitely not a fan of FPTP. I think in some jurisdictions RCV is an improvement (if it's the only reform that has familiarity and momentum behind it), and in others it makes more sense to push for Approval Voting, which is simpler to count than RCV. In fact, Approval Voting is a lot like Score Voting with only approve and disapprove options.
You're spot on in mathematics and CS on this. But the important take-away to me is that there is no way to express disapproval in the current western democratic voting system. Even non-voting is not disapproval. I'd welcome very much if we could agree that we have a problem there, and shift focus from criticizing new ideas to improving them.
To me, we definitely need democracy 2.0. If possible without resorting to meritocracy, but some sound foundation where it's possible to voice disapproval.
And yet, when I talk to people about why they vote the way they do, or when I listen to campaign advertisements, I get the sense that actually the majority of votes cast are "protest votes" whereby one is voting for candidate X because they wish to express their hatred/disapproval/disdain of candidate Y.
> I'd welcome very much if we could agree that we have a problem there, and shift focus from criticizing new ideas to improving them.
I'm happy to think more about this problem and see if we can find some common ground (which is why I mentioned other voting reforms as ways of capturing more information about the preferences of the public), but I think I'm still a little unsure about why "expressing disapproval using voting systems" should be a goal.
Expressing disapproval in parties is, I think, an important part of a healthy democracy, but that can be done using free speech, protests, and opinion polls. Is that not enough? Maybe it's best for societies to follow your advice and, instead of criticizing existing parties, work to improve them (from within) or positively support alternative parties offering a different vision.
I think there's a real danger when trying to add "express disapproval" as an output of a voting system, since (as the article itself noted) it's already very difficult to create a voting system which fulfils other reasonable requirements. If voters have a choice between expressing disapproval of a party, and increasing the chances of their preferred party winning, they will opt for the latter. The existence of tactical voting suggests that it is very hard to make a voting system reward honesty, so we shouldn't try to overload it and make it do too many things.
4 comments
[ 3.6 ms ] story [ 25.5 ms ] threadIt's a pity there isn't an example of what a ballot paper would look like under the proposed system, but I'm imagining that next to each candidate there would be three boxes representing: approve, neutral, disapprove.
I suppose that's easier than a ranking system, because there might be lots of candidates that most voters haven't heard of (or don't have a strong opinion of) and it can be hard to put them into an exhaustive ranking.
The temptation, though, will always be to approve of the one candidate you like, and disapprove of all the rest (and candidates will encourage their supporters to do this), so the system will quickly devolve into plurality-style voting anyway, except the votes will be harder to count because there will be 3^n different ways of marking the ballot, rather than just n (where n is the number of candidates).
In any case, this looks a lot like Score Voting, with possible scores of 0, 1 or 2 (which I think is mathematically the same as -1, 0, and 1). Again this has the problem that you can't just count the number of votes in each sorted pile, and instead have to have n different running tallies, with no checksum at the end to make sure the tallies all total to the number of votes cast, for example.
This isn't a problem for a computer algorithm, but anything that encourages people to outsource their democracy to a piece of firmware, commissioned and maintained by one ruling party, is just asking for trouble. Even if the firmware works as claimed, you still introduce the problem that FUD can be used to attack a legitimate result.
Don't get me wrong, though, I'm not against voting reform, and I'm definitely not a fan of FPTP. I think in some jurisdictions RCV is an improvement (if it's the only reform that has familiarity and momentum behind it), and in others it makes more sense to push for Approval Voting, which is simpler to count than RCV. In fact, Approval Voting is a lot like Score Voting with only approve and disapprove options.
To me, we definitely need democracy 2.0. If possible without resorting to meritocracy, but some sound foundation where it's possible to voice disapproval.
I'm happy to think more about this problem and see if we can find some common ground (which is why I mentioned other voting reforms as ways of capturing more information about the preferences of the public), but I think I'm still a little unsure about why "expressing disapproval using voting systems" should be a goal.
Expressing disapproval in parties is, I think, an important part of a healthy democracy, but that can be done using free speech, protests, and opinion polls. Is that not enough? Maybe it's best for societies to follow your advice and, instead of criticizing existing parties, work to improve them (from within) or positively support alternative parties offering a different vision.
I think there's a real danger when trying to add "express disapproval" as an output of a voting system, since (as the article itself noted) it's already very difficult to create a voting system which fulfils other reasonable requirements. If voters have a choice between expressing disapproval of a party, and increasing the chances of their preferred party winning, they will opt for the latter. The existence of tactical voting suggests that it is very hard to make a voting system reward honesty, so we shouldn't try to overload it and make it do too many things.