a permalink

news and thoughts on and around the development of the iCite net
by Jay Fienberg

What is a decision in the system of votes

posted: May 14, 2004 8:18:05 PM

I don't write or blog about political systems or their people, so I've been hesitant about posting this. But, I felt like I wanted to put this out there. This is an observation / theory about how the system of votes work in the big U.S. elections.

So, there is a common idea that voting is about making a choice about who will win an election for, say, president of the U.S. And, in general, elections are considered in terms of who will win. Your vote helps select the winner.

What I think is misunderstood is how much voting is actually about selecting the loser(s) and how that type of selection process is somewhat different. I think we miss this nuance because, emotionally, we get caught up in the idea of the winner.

So, we tend to think: who will my vote help win, i.e., my vote for Candidate A will help Candidate A win. But, what we actually should be thinking is: who will my vote help lose, i.e., my vote for Candidate A will help Candidates B and C both lose.

So, I think the question worth asking is not who one is voting for, but who one is voting against. The more essential decision we are making by voting is selecting whom we don't want to win the election. I think that is actually the primary responsibility we have in collectively making decisions through voting in the existing system.

And, actually, I think the political people's campaigns are designed around this understanding (even if they don't admit it): campaigns are designed to help the other candidates lose.

Now, with the big U.S. elections, I think one must draw a basically flat line between the so-called left and right ends of the political spectrum (whatever the heck that is!). I don't really get what's left or right about it (those labels don't guide my wayfinding), but somehow a lot of people seem to agree that Candidate A is left of Candidate B, and, vice versa, Candidate B is right of Candidate A.

So, we end up with this flat line, like:

Candidate A - Candidate B - Candidate C - Candidate D

And, what I think is that, in this voting-against system, one's vote for a candidate functions as if in a weighted manner against the other candidates, working back towards the center and to the other side (and, then around again). So, a vote for Candidate A functions as if it's mostly a vote against Candidate B, and then less so against Candidates C and D.

So, the two main candidates in an election end up positioned in the center (Candidates B and C, above), and more fringe candidates end up further out to the left or right. What I am suggesting is that a vote for a fringe candidate is effectively more a vote against one of the central candidates on the same side of left / right than against any candidates on the opposite side of the line.

My idea about this, in part comes from at least a couple of the presidential elections in my lifetime. Votes "for" political candidates like Nader and Perot were more specifically votes against Gore and Bush (the father) respectively.

In the upcoming U.S. presidential election, it seems like people on the right are generally happy about further-left candidate Nader's candidacy. Why? Because votes for Nader are more effectively votes against the more center-left candidate Kerry than against the candidate Bush (the son) on the right.

I don't have a proof of this theory, but I think one factor is that the more fringe the point of view of the candidate, the more minority his or her voting constituency (i.e., less people to vote for that candidate). I think this might be a factor at least at the level of the big state and national elections where the voter base is averaged across the whole country or the whole of a (at least, large) state.

So, I decided to post this because I think it marginally fits in with one theme of this blog: recognizing where systems exist, observing people's interaction with them, noting the results, and considering what is uninspected. In this case, I am just saying that I think there is something worth observing about votes (in systems like the U.S. voting system) in their relation to election losers rather than just in relation to election winners.

permalink | comments {4} · trackbacks {0}

also available as: rss · rss2 · rdf · atom

Comments and Tracbacks

Comment by: David Weinberger · http://www.hyperorg.com/blogger
posted: May 15, 2004 8:54:12 AM

Jay, you write: "The more essential decision we are making by voting is selecting whom we don't want to win the election." Can you say more about what makes it more essential? Likewise, you suggest "a vote for a fringe candidate is effectively more a vote against one of the central candidates...,etc." but I'm not sure where the "more" is coming from.

There are times when I've voted for a candidate and times when I've voted against his opponents. I think your observation operates at a different level since you seem to be applying it to all elections, but I'm not sure the level at which you intend it to apply.

It's a really interesting observation. Thanks.

Comment by: Jay Fienberg · http://icite.net
posted: May 15, 2004 3:12:49 PM

Thanks for your comment David.

"Make your vote count", we're told. I always equated that with "vote for the person you want to win".

But, if there are, say, four candidates in an election for a single position (without instant run-off, etc.), your vote could be thought of as counting *for* one candidate and *against* three.

This is maybe a clearer example of how (I think) one's vote counts more against than for. One's vote acts as a statement against three candidates, but for only one.

In this system then (when there are more than two candidates for a single position), maybe we should say that one's voting decision is essentially a statement against all of the candidates except one, rather than an endorsement of one of them.

So, having said that, I don't know if "essential" is such a good word. Part of what I am getting at is that the "statement" our votes end-up representing is not just a simple affirmation / endorsement of one candidate.

Then, when we look at the large elections in the U.S., I think the number of votes tends to normalize into a contest between left and right, and which contest tends to define itself as a normalized balance of many issues (i.e., there tends to be an even match between left and right because so many issues get expressed in terms of the left's and the right's position on them).

So, although there are often more than two candidates, they get projected down into flat, binary (yes/no and left/right) space in which our votes go to decide the election.

So, I don't know if this is a correct model and/or if there is a good way to figure out how correct it is, but imagine multiple candidates are being mapped into a single "role" (say, a left candidate). Say, two candidates on the left are basically going after what ends up being "votes for the left". Then, the competition between them can more effectively weaken the left side then it can the right side (i.e., X votes divided between two left candidates and X-1 votes for one right candidate results in both the left candidates losing).

So, I think some game theory could illustrate that (using this example) the way the left could beat the right would require the left candidates to cooperate and try to get all the votes directed to only one of them: competing between them weakens themselves more than it weakens the opponents on the other side of the left-right spectrum. (And, thus, "divide and conquer".)

Anyway, I was just thinking that something like what I've described, i.e., how to affect who loses, is a more significant factor in how one's vote counts than in looking at how to affect who wins one of these big elections.

Also, this all suggests how a democracy might benefit from a different way to count votes altogether. (I'm guessing people might appreciate being able to express more nuance through their votes). As an example, I just found The Mathematics of Voting, by Florin Diacu, which offers some mathematical insight on what might be beneficial.

Comment by: Tim Macalpine · http://www.debt-solution-tips.com
posted: May 22, 2004 5:43:22 PM

Jay, I'm in my 30's. So, I've participated in several Presidential elections.

But, I've never seen, read or heard the theory you've just explained. That was one of those posts that made me sit back and think. (Admittedly, I had to read a couple of parts twice as my noodle isn't used to processing such lofty ideas :-).

Frankly, I'm still wading through your comment...

"one's vote for a candidate functions as if in a weighted manner against the other candidates, working back towards the center and to the other side (and, then around again). So, a vote for Candidate A functions as if it's mostly a vote against Candidate B, and then less so against Candidates C and D."

Whoa. That's gonna take a few minutes for me to chew on.

Anyway, I further admit that I've typically had the perception that I was voting for "The Winner" rather than voting against "the loser." It's what I've allowed my mind to adapt to. But, as I think about your post, you're right... Our votes are cast in favor of one and against another. At times, though not usually the intent, those votes serve both purposes. Not always, but sometimes.

The Perot vote is a good example. Some voted for Perot because they liked him. That is a vote in favor of him. Of course, that vote in favor of Perot also served the dual purpose of a vote against his opponents.

But, some folks voted for Perot simply because they disliked the "other guys."
In this case, they used their vote to vote against those "other guys."

Well, I'm rambling, but to sum it up, thanks for such a thought-provoking post.

Tim

Comment awaiting approval
posted: Mar 21, 2006 4:08:49 PM

Note: All comments and trackbacks are moderated. Spam is deleted. Other comments are approved as promptly as possible.

Note: Older posts no longer accept new comments or trackbacks.

« prev post
Protocol, format, medium

» next post
Just back from Technoratania