Gottlieb on Voting

// Published May 13, 2011

Posted by Hendrik Hertzberg JULY 20, 2010
The New Yorker

In this week’s magazine, Anthony Gottlieb reviews “Numbers Rule: The Vexing Mathematics of Democracy, from Plato to the Present,” a new book by George Szpiro, a journalist and mathematician. The book—and Gottlieb’s essay—are about a subject that got its teeth around my ankle years ago and refuses to let go: voting systems.

For me, Gottlieb’s piece has many pleasures, including a brisk demolition job on “first past the post,” the system used for the mother of parliaments—you know, the old lady with that nice House on the Thames, which she shares with her husband, Big Ben—and her colonial offspring in the United States, Canada, and India. As Gottlieb points out, the rest of the democratic world has opted for more up-to-date political technologies, so much so that

it’s clear that no country would pick first-past-the-post voting today. Of democracies with no significant British past, only Nepal now elects its national assembly this way.

Even some democracies with quite significant British pasts have dumped first-past-the-post. New Zealand switched to “mixed-member” proportional representation, the kind they use in Germany, in 1993. Australia switched to instant-runoff voting—where you rank your choices—way back in 1918. And the country with the most significant British past of all, Britain, has decided to decide whether it wants to do likewise.

If I have a complaint about Gottlieb’s piece (and I do), it’s that he’s far too hard on instant-runoff voting. Like all voting systems, especially those for single-seat offices where there can be only one winner, I.R.V. has flaws. The economist Kenneth Arrow, Gottlieb writes, “examined a set of requirements that you’d think any reasonable voting system could satisfy, and proved that nothing can meet them all when there are more than two candidates”—and won the 1972 Nobel Prize for this discouraging discovery. But according to Gottlieb, “the quirks of instant-runoff voting are an extreme case.”

I must disagree. From a strictly mathematical point of view, the quirks of I.R.V. are about average. They’re hardly more troublesome than (and, as Gottlieb does not mention, often identical with) those of the non-F.P.T.P. alternatives, “approval voting,” where you put an X next to the candidates you find acceptable, but without ranking them, and “range voting,” where you give points to the candidates, like stars in a movie or restaurant review, and the candidate with the most points wins.

Of the author he is reviewing, Gottlieb writes: “Szpiro, who is more interested in math than in politics, says relatively little about how voting systems have played out in the real world.” The real world, where voting actually occurs, is where I.R.V. has proved its mettle. In that world, approval and range voting have never managed to achieve anything approaching liftoff. I.R.V. is all over the place, from Sydney to San Francisco to (soon, I hope) London, and for good reasons—reasons that have more to do with political behavior and human intuition than with mathematical purity.

One problem with approval voting, for example, is that voters are instinctively reluctant to give equal weight to a candidate they really like and candidates they can merely tolerate. The result is often “bullet voting,” which range voting also encourages. With both, the winner can be a candidate who would finish dead last in a simple plurality election. That’s an impossibility under I.R.V.

I don’t want to wander too much further into the tall grass, but if you’re looking for the Full Wonky, Rob Richie, the executive director of FairVote (I’m on the board of that admirable little outfit, by the way), has what you’re looking for. What Churchill said of democracy in general can be said of I.R.V. in particular: it’s the worst system—except for all the others.

That complaint aside, Gottlieb’s essay is a delight. (Who knew that the Venice of the Doges had an electoral college that makes ours look like a neighborhood preschool? I didn’t.) Anyway, he acknowledges that “almost any alternative voting scheme now on offer is likely to be better than first past the post.” For that alone, he’s got my vote.


NEWS FLASH: The Massachusetts House today voted final approval for the National Popular Vote bill, 112-31. (The original vote for passage was 113 to 35.) Now it’s on to the state senate and the governor.


Originally posted on