A man votes at City Hall in San Francisco on Nov. 8. (AP Photo/Jeff Chiu)

On Tuesday, New York Times columnist David Brooks described the U.S. political party system as having devolved into an unsustainable “us versus them” mentality in which “the ends justify the means.” Unless something changes, he wrote, decent people will say enough is enough and “we could end up with a European-style multiparty system.”

In some ways, we’re already there. Our research, soon to be published in Public Opinion Quarterly, finds that many Americans are indeed fed up with politics on both sides of the aisle. While it’s common to describe the public as “polarized,” we think that exaggerates how much “us versus them” really characterizes the way American voters see the two parties. Many partisans dislike the other side. But many also don’t think much of their own party.

The problem for Brooks’s vision of a popular overthrow of the two-party system, however, is not just the formidable structural barriers — it’s also that the very people most disaffected with the two parties are the least likely to be politically active. As a result, dissatisfaction with the Republicans and Democrats may not be enough to spur a fundamental change to the party system.

How we did our research

Our study draws on two sets of data. The first comes from an experiment we conducted in 2016 in a nationally representative survey of 2,136 American adults conducted by the survey firm GfK Knowledge Networks. The study was funded by Time-sharing Experiments for the Social Sciences.

In our experiment, we asked people how they would feel if their child marries someone from a particular political party. This is similar to a question from a 2012 article by Shanto Iyengar and colleagues that is widely cited as evidence of polarization. In that study, about 49 percent of Republicans and 33 percent of Democrats reported that they would be somewhat or very “unhappy” about the prospect of their child marrying someone from the other party.

We asked about one-third of our respondents the standard question: How would they feel if their child married someone across party lines. But the rest of the sample was given more context. They were told either that the hypothetical child-in-law talks about politics “rarely” or “frequently.” We also asked respondents how they would feel about their child marrying someone from their own party who talked about politics with the same frequency.

This gives us a better sense of how polarized people really are. We consider a survey respondent as “polarized” if they are unhappy with their child marrying someone from the other party and happy with their child marrying someone from their party.

Are Americans polarized, or do they just not want to talk about politics?

Among respondents who were asked the standard question, we find that at most 25 percent of the public is polarized; the majority are strong partisans. Among those participants who identify as strong partisans, about 48 percent are polarized. But about two-thirds of Americans have weaker loyalties; among that group, we find that only about 14 percent are polarized.

Among the respondents given information about how often their hypothetical son- or daughter-in-law talks about politics, we find the level of polarization drops even further.

When the child-in-law is described as “rarely” talking about politics, then people don’t care as much that they are member of the other party. On the other hand, respondents are less happy with a child-in-law who talks about politics frequently — even if that person shares their party affiliation.

That suggests what many people want in a new family member is someone who doesn’t talk politics.

Some Americans have grown cool toward their own party

Our second analysis is from the 2016 American National Election Study, considered the gold standard of nationally representative election surveys. Over the last several decades, researchers have asked people how “warm” they feel toward the two parties on a scale from zero to 100. Since the 1970s, partisans have grown increasingly “cool” toward the other party.

But our research finds that some Americans’ feelings about their own party are cooling, as well. In 2004, fewer than 13 percent of ANES respondents rated their party below 50 — that is, not warmly. But by 2016, that cooler rating had climbed by eight percentage points.

That change is concentrated among people who aren’t strong partisans. Among this group, the average rating of their own party fell from 67.7 to 58.7 between 2004 and 2016. If people are growing to dislike even their own party, then something is changing. But the change isn’t simply increasing polarization.

Another way to see this is by looking at the percentage of partisans who give their own party a rating of more than 70 (what the ANES describes as “a fairly warm or favorable feeling”) and the other party less than 30 (“a fairly cold or unfavorable feeling”). These, of course, aren’t the most polarized Americans, but this would seem to fit a minimal definition of polarization.

With this measure, we find that only about 38 percent of the public is polarized, which means that 62 percent is not. As in our experiment, there is a large divide between strong partisans and everyone else. About 65 percent of strong Democrats and 54 percent of strong Republicans are polarized. But among all other respondents, only 19 percent are.

Mobilizing the disaffected is hard when they don’t care much for politics at all

In a sense, Brooks is ahead of a lot of pundits in observing that the “us versus them” politics of the Democratic and Republican parties has led people to desire something new. To be sure, the parties are not hemorrhaging voters, though there has been a small decline in partisan identification over the last decade. In the 2016 ANES, 57 percent of respondents said they wanted a third party.

The problem, however, is that a new party would have to mobilize these Americans. And what defines these citizens — the ones who express the most disenchantment with the two-party system — is that they’re not politically engaged. As we saw, many don’t even want to talk about politics even when they agree with the other person.

And this is Brooks’s problem. To overcome the structural barriers preventing new parties would require people who really dislike both Democrats and Republicans. But these people – our research suggests — might just dislike politics.

John Barry Ryan (@ryanbq) and Yanna Krupnikov (@ykrupnikov) are associate professors of political science at Stony Brook University. Samara Klar (@samaraklar) is assistant professor of political science at the University of Arizona.