Why do people on the other side seem so unreasonable?

By: Andrew Gelman/Monkey Cage Washington Post

George W. Bush famously described himself as “a uniter, not a divider.” A few years into his presidency there was a survey that reported that 49 percent of Americans thought he was a “uniter” and 49 percent thought him a “divider” — a poignant reminder that Americans are so polarized, they’re even polarized about polarization.

And of course the same goes for Obama who, like Bush, spent two terms as president trying, but failing, to be a bipartisan figure.

A characteristic feature of polarization seems to be the impression that one’s own side is reasonable and that all the polarizing comes from the other side of the political aisle.

I came across an amusing example of this today, ironically from a political scientist, Gerard Alexander, who, in an op-ed entitled, “Jon Stewart, Patron Saint of Liberal Smugness,” writes

Many liberals, but not conservatives, believe there is an important asymmetry in American politics. These liberals believe that people on opposite sides of the ideological spectrum are fundamentally different.

This is just too perfect. It’s a beautiful paradox. If Alexander is right, then there is an important asymmetry in American politics, which is the thing that he’s saying conservatives don’t believe! If he’s wrong, then there is no asymmetry, which is what he’s saying conservatives believe in the first place. It’s like something out of Lewis Carroll.

I’m reminded of economist Emily Oster’s quote that economists are different from almost everyone else in society in that they assume everybody is fundamentally alike.

I also think it’s a bit odd for Alexander to describe liberals as “fixated” on Sarah Palin: it’s not liberals who nominated Palin for vice-president, or who put her on TV. Unless you want to describe William Kristol and Roger Ailes as liberals.

Also this, from Alexander:

My strongest memory of Mr. Stewart, like that of many other conservatives, is probably going to be his 2010 interview with the Berkeley law professor John Yoo. Mr. Yoo had served in Mr. Bush’s Justice Department and had drafted memos laying out what techniques could and couldn’t be used to interrogate Al Qaeda detainees. Mr. Stewart seemed to go into the interview expecting a menacing Clint Eastwood type, who was fully prepared to zap the genitals of some terrorist if that’s what it took to protect America’s women and children.

Mr. Stewart was caught unaware by the quiet, reasonable Mr. Yoo, who explained that he had been asked to determine what legally constituted torture so the government could safely stay on this side of the line. The issue, in other words, wasn’t whether torture was justified but what constituted it and what didn’t.

First off, let me say that this is a horrible thing to say about Clint Eastwood, who was never involved in this:

On December 1, 2005, Yoo appeared in a debate in Chicago with Doug Cassel, a law professor from the University of Notre Dame. During the debate, Cassel asked Yoo,

‘If the President deems that he’s got to torture somebody, including by crushing the testicles of the person’s child, there is no law that can stop him?’, to which Yoo replied ‘No treaty.’ Cassel followed up with ‘Also no law by Congress — that is what you wrote in the August 2002 memo’, to which Yoo replied ‘I think it depends on why the President thinks he needs to do that.’

It’s interesting to see what Alexander did there. First, he changed the question from “crushing the testicles of the person’s child” (which Yoo refused to say the government couldn’t do) “zap[ping] the genitals of some terrorist” (which is, presumably, also a treaty violation but doesn’t sound so bad). Second, Alexander didn’t even say “suspected terrorist,” he said “terrorist.” But of course one of the concerns with torture is that it’s not only done on terrorists, it’s also done on people who get picked up on suspicion of terrorism, or just people who somebody thinks might have some information. Alexander gets to characterize Yoo as “reasonable” only by shifting the ground of the debate. He also describes Yoo as “quiet,” which seems pretty irrelevant to me. Personally, I’d prefer someone who yells but opposes torture to someone who is soft-spoken and supports it. Then again, I don’t personally know anyone who’s been killed by a terrorist; maybe if I did, I’d feel differently.

Anyway, the point is that Alexander demonstrates political polarization in his column, in the very way that he is castigating liberals for unreasonable for seeing themselves as more reasonable than conservatives. And, if you’d like, you could characterize this post as even more polarization, in that of all the things I could write about today, I’ve chosen this.

My point, though, as a political scientist, is that along with our very reasonable concerns about difficulties of communication between the left and the right — an often disturbing lack of national unity — comes a corresponding split in perceptions. Alexander is correct to see a tendency among many liberals to think of themselves as more reasonable than conservatives. But his column, ironically, demonstrates the symmetric tendency of many conservatives to see themselves as the reasonable ones. So, instead of mere disagreements on policy, we get disputes about the legitimacy of each side’s concerns, as well as ahistorical views such as an attribution to liberals of Sarah Palin’s fame.

None of this is new, but it has been getting worse in the past generation, and this particular column reminded me of it.

Finally, it’s data time. This is the Monkey Cage, after all. Here are some graphs from my book Red State Blue State on partisan polarization during the Korean, Vietnam and Iraq wars.

Andrew Gelman is a professor of statistics and political science at Columbia University. His books include Bayesian Data Analysis; Teaching Statistics: A Bag of Tricks; and Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way They Do.