Jef writes a column in his local paper. Called "Pacifica Moods", it appears in the Pacifica Tribune. Recently, the citizens were asked to vote on a minor issue, but the method was flawed. The week before the election, Jef wrote this article. While letters to the editors about it all were supportive, the election went ahead as planned, with the kind of outcome that Jef had predicted.

The article has been slightly modified to explain the circumstances that local citizens already knew but that a more general readership would not.

VOTING FOR A PORTAL

Recently, the citizens of my town were being given a chance to indicate our preferences as a guide to what the portals to a long-awaited bypass tunnel should look like. The tunnel bypasses a dangerous section of coastal road that, while scenic, often falls into the ocean, cutting off our only access directly to the south; the tunnel itself has nearly universal acceptance. The differences we were being asked about were purely cosmetic, so there were no engineering but only aesthetic concerns involved. We were asked to vote for the one we liked the best.

The obvious way to determine which is the favorite is to choose the portal that gets the most votes. That is the obvious way, but it is also the wrong way, and is likely to choose a portal that the majority dislikes. Let me explain why.

Let's take a simpler example, where there are only four choices. One looks like a dragon's mouth, the second looks like natural rock, the third looks like gray brickwork, and the last is a simple, smooth concrete arch. To make the numbers simple, we'll pretend that there are only 100 voters. Say that 28 of them want the dragon, and the other 72 hate the dragon, and want a gray opening that blends in with the landscape. It is clear that the townsfolk don't want the dragon.

But among those 72 who don't want the dragon, 26 like the rock, 24 like the brick, and 22 like the concrete. When you tally these votes, the dragon wins, and 72% of the town is unhappy, as all would have preferred ANY of the other portals better.

In any voting situation where you can only vote for one and there are more than two choices, there is a very good chance that the majority will is being thwarted.

A much better system is where everybody ranks their choices, in this case, from 1st to 4th. The voting is now finished. But the counting is done like this: The choice with the least first place votes is eliminated. In the portal example the concrete would be eliminated as a possible winner, and everybody who voted for concrete now have their second choice rated as number 1. Say that their second choices were evenly split between the rock and the brick, so now we have dragon 28, rock 37, and brick 35. The dragon is eliminated, and its votes are distributed to either rock or brick (whichever scored higher on each ballot). Either rock or brick will win, and if the dragon voters were evenly split, rock wins. This is an outcome that 72% of the people can be comfortable with.

In all of our elections where there are more than two candidates or options, the present plurality voting system often gives us winners that do not represent what the citizens want. The system I describe, called the "instant runoff" method, is far superior. It is not a perfect system, but (as Kenneth Arrow proved) there is no perfect voting system. There are other methods that are also superior to the plurality system we use today, in terms of reflecting what the people want.

Our founding fathers did not have the benefit of Arrow's work (published in 1952) or of other work on voting theory. But they were enlightened men, and had they the information we now do, would surely have chosen voting methods that would make our country more beholden to the citizens' wishes.