We Can’t Even Agree on What Is Tearing Us Apart
Today, even scholars of polarization are polarized.This was not always the case.In 1964, Philip Converse of the University of Michigan wrote a groundbreaking paper, “The Nature of Belief Systems in Mass Publics,” that attempted to determine the size of the share of the electorate that could reasonably be described as having a consistent set of political convictions.Converse, a political scientist, was interested in figuring out how many people have what he called “a belief system, a configuration of ideas and attitudes in which the elements are bound together by some form of constraint or functional interdependence.”For Converse, such a system suggested that if “an individual holds a specified attitude, he holds certain further ideas and attitudes.” For example, “If a person is opposed to the expansion of Social Security, he is probably a conservative and is probably opposed as well to any nationalization of private industries, federal aid to education, sharply progressive income taxation, and so forth.” Converse called voters who fit this description “ideologues.”At the time he was writing, Converse noted, only 3.5 percent of voters could be described as “ideologues,” 12 percent as “near ideologues,” and the remaining 84.5 percent cast their ballots on the basis of whether their group would benefit, the state of the economy or, in Converse’s words, “no issue content.”Now, nearly six decades later, the issue is not the lack of an ideological and partisan electorate but the dominance of polarized elected officials and voters, some driven by conviction, others by a visceral dislike of the opposition, and still others by both.This turbulence has proved to be a gold mine for scholars seeking to find order in the disorder.Take two papers, both published in August 2021: “Constrained Citizens? Ideological Structure and Conflict Extension in the U.S. Electorate, 1980-2016,” by Christopher Hare, a political scientist at the University of California, Davis, and “Moderates” by six political scientists, Anthony Fowler, Seth Hill, Jeff Lewis, Chris Tausanovitch, Lynn Vavreck and Christopher Warshaw.Hare, in his paper, argues that polarization is present throughout the electorate:As among elites, the left-right dimension has come to encompass a wide range of policy, partisan, and value divides in the mass public. Further, these trends hold for voters at all levels of political sophistication. Widespread conflict extension appears to be a defining feature of mass polarization in contemporary U.S. politics.The level of conflict, Hare continued,speaks to a fundamental trade-off in a pluralistic democracy. On the one hand, the presence of reinforcing policy cleavages — particularly those involving emotional, “high heat” social and cultural matters — changes the tenor of democratic deliberation. Political competition becomes Manichaean: less transactional and more messianic. This environment fosters partisan-ideological antipathy, resistance to compromise, and other manifestations of affective polarization.Citizens with low and moderate levels of political sophistication, Hare argues, “are catching up to their highly sophisticated counterparts in terms of ideological constraint,” and thisconflict extension tends to reinforce itself through a positive feedback loop. Overlapping conflicts provide multiple pathways to the same partisan-ideological configuration. Better-sorted voters, in turn, further clarify party differences. Each process feeds into the other, and partisan competition engulfs both a wider set of conflicts and a larger segment of the electorate. Consequently, when the same political opponents repeatedly clash over emotional issues and fundamental values with messianic zeal, it becomes easier to view the other side as evil rather than merely incorrect.In an email, Hare argued that the current trends in partisanship “all seem to contribute to a feedback loop.” The sorting of voters into two camps, Democrats and liberals versus Republicans and conservatives, is notan innocuous form of polarization, especially when the attitudes and attributes on which voters are being sorted are such deep wellsprings of dispute (religion, race, urban vs. rural, core values such as moral traditionalism and egalitarianism, etc.). As these divisions overlap with each other and an increasing number of diverse policy issues (including value-laden issues like abortion, immigration, guns, etc. that also expose cultural fault lines and are much less amenable to compromise and standard political horse trading), it’s little surprise that politics becomes more combustible.Fowler and his co-authors, on the other hand, contest the view that voters are deeply polarized:We find that a large proportion of the American public is neither consistently liberal nor consistently conservative but that this inconsistency is not because their views are simply random or incoherent. Instead, we estimate that many of those who give a mix of liberal and conservative responses hold genuine views in the middle of the same dimension of policy ideology that explains the views of consistent liberals and consistent conservatives.There are, Fowler and his collaborators point out,many genuine moderates in the American electorate. Nearly three in four survey respondents’ issue positions are well-described by a single left-right dimension, and most of those individuals have centrist views. Furthermore, these genuine moderates are a politically important group. They are highly responsive to the ideologies and qualities of political candidates.In an email responding to my inquires, Fowler wrote:Elites are highly polarized, but members of the general public are not. Of course, there is sorting in the sense that liberals are more likely to identify as Democrats and conservatives are more likely to identify as Republican. So there are differences between the opinions of the average Democrat and the average Republican, but unlike for members of Congress, the distribution of ideology in the American public is unimodal, with most people close to the middle.There is a second crucial area of disagreement among scholars over what is known as “affective polarization,” the idea that there has been a steady increase in the level of animosity among voters toward members of the opposite party.Lilliana Mason, a political scientist at Johns Hopkins, described affective polarization in a 2016 paper, “A Cross-Cutting Calm: How Social Sorting Drives Affective Polarization”:In recent decades, a particular type of partisan sorting has been occurring in the American electorate. American partisan identities have grown increasingly linked with a number of other specific social identities. These include religious, racial, and other political group identities, such as the Tea Party.For those whose social identities have become deeply entwined with their partisan identities as Democrats or Republicans, political defeat can produce intense anger. “This anger,” Mason writes, “is driven not simply by dissatisfaction with potential policy consequences, but by a much deeper, more primal psychological reaction to group threat. Partisans are angered by a party loss because it makes them, as individuals, feel like losers too.”In a separate 2018 paper, “Ideologues Without Issues: The Polarizing Consequences of Ideological Identities,” Mason argued that “American identities are better than American opinions at explaining conflict,” noting:Identity-based ideology can drive affective ideological polarization even when individuals are naïve about policy. The passion and prejudice with which we approach politics is driven not only by what we think, but also powerfully by who we think we are.Two political scientists, Lilla V. Orr at Stanford and Gregory Huber at Yale, dispute Mason’s portrayal of affective polarization in their 2019 paper, “The Policy Basis of Measured Partisan Animosity in the United States.” They argue that policy and ideology play a strong role in the level of dislike or hatred between members of opposing parties and that policy agreement can moderate much of the hostility:In a series of survey experiments asking participants to rate the warmth they feel toward people described in vignettes, we find that issue positions influence interpersonal evaluations more than partisanship. Issue effects remain strong even when partisanship is also presented (they decrease by about 12 percent), but the effect of partisanship decreases substantially in this setting (by about 60 percent).Perhaps most striking, Orr and Huber found that “respondents more favorably evaluate out-partisans who hold a shared policy position than co-partisans who disagree on policy.”In an email, Huber wrote that there is “a very simple alternative” to explain growing partisan animosity:People pick which party to be in on the basis of issues, (and) animosity is based in policy disagreement. We know people disagree about stuff, and of course that is the basis of politics. If we all agreed, we wouldn’t need a political system to manage our disagreements. But, disagreement needn’t be elite driven or linked to being in a tribe; it might just be that we have preferences.Huber elaborated:Because the parties are clearly defined ideologically, we tend on average to (correctly) assume that the mass public’s views are also linked to their partisanship. So if I want to understand why people are affectively polarized, it is because they are correct that on average Republicans and Democrats disagree about issues, and not just random issues, but the issues that lead us to pick which party we belong to. I wouldn’t call that tribal — that sounds like issue-based politics.Orr contended in an email:Several experiments have successfully manipulated feelings toward people from the opposing party and found no effects on anti-democratic attitudes or other predicted consequences of affective polarization. These results imply that affective polarization might not be as inherently dangerous as many researchers previously assumed. I don’t mean to downplay the harm caused by partisan-motivated violence or dismiss efforts to combat blind partisan hostility, but a widespread failure to be upset by some of the things our fellow U.S. residents are trying to accomplish through politics can also put people in physical danger.In an effort to clarify the relationship between ideological and affective polarization, I queried a number of political scientists and received some thoughtful responses.Sean Westwood of Dartmouth stressed the importance of “sorting”:Sorting argues that citizens and elites proactively move themselves to parties that best capture their views. By consequence this should reduce the prevalence of misaligned senators like Joe Manchin. Sorting makes the parties more cohesive and moves the party toward more extreme positions on average, increasing ideological polarization. Evidence for sorting among elites is very strong. The best way to think of this, I think, is that sorting causes polarization.What, then, about affective polarization?Affective polarization is more widespread than principled ideological polarization among voters. Even voters entirely ignorant of the policy positions of their party can develop an emotional attachment to co-partisans and a negative view of the opposition.Unfortunately, Westwood continued,We don’t really know where affective polarization comes from or why it is surging. Some argue that it comes from ideological sorting, but this isn’t very satisfying because many people have affective preferences and are simultaneously unable to correctly identify the policy positions of the parties.How do sorting, ideological polarization and affective polarization interact?Sorting among elites makes it less likely for more conservative voices to exist within the Democratic Party and for more liberal voices to exist in the Republican Party. This makes the parties more ideologically cohesive and more likely to adopt more extreme policy positions, increasing ideological polarization. This is magnified by the lack of centrists running for and winning office. Polarization is the consequence of the complementary phenomenon of growing ideological cohesion among the parties and fewer moderates to temper the movement of parties to ideological extremes.At the same time, Westwood argues:Voters are becoming more affectively polarized. The connection between elite ideological polarization and citizen affective polarization seems credible on face value, but evidence to support such a relationship is very hard to come by. This is what makes research on the causes of affective polarization so important and our current understanding so frustrating. I would see ideological polarization and affective polarization as two parallel and related phenomena, but with different antecedents and different effects on behavior.Yphtach Lelkes, a professor of communications at the University of Pennsylvania, addressed my queries from a different angle, emphasizing ideological constraint:With the risk of sounding overly academic, researchers sometimes say polarization can be thought of as a loss of dimensionality in the issue space. That is, if I knew your position on abortion, I didn’t necessarily know your position on health care. Among elites, the dimensionality of the issue space has completely collapsed. If I know where a senator stands on abortion, I know where that senator stands on health care, gun rights, immigration, etc.Political scientists, he continued,have long argued that a sign of the disconnect between the mass public and elites is that the mass public’s attitudes are multidimensional (knowing where someone stands on abortion doesn’t necessarily tell me where they stand on health care), while elites’ attitudes are unidimensional.That is changing, Lelkes argued, citing the work of Chris Hare in the paper I already mentioned: “Hare (who knows a ton about this stuff) has recently shown that public opinion is now collapsing onto one dimension.”Affective polarization can have substantial real-world consequences.James Druckman, Samara Klar, Yanna Krupnikov, Matthew Levendusky and John Barry Ryan, of Northwestern, the University of Arizona, Stony Brook, the University of Pennsylvania and Stony Brook, studied partisan responses to the Covid-19 pandemic in their 2021 paper, “Affective polarization, local contexts and public opinion in America,” and found:For worry about Covid-19 and support for Covid-19 policies, the marginal effect of animus is significant and negative for Republicans in counties with few cases; the confidence intervals for the other marginal effects overlap with zero. Increases in animus are statistically significant only for Republicans in counties with low cases, suggesting that, for worry and support, partisan gaps are largely a function of Republicans with considerable animus towards Democrats.Gary Jacobson, of the University of California, San Diego, summarized the evolution of polarization in his emailed reply:Over the past four decades, largely in response to the more sharply differentiated alternatives presented by the national parties and their leaders, voters have sorted themselves into increasingly distinct and discordant political camps. As partisan identities, ideological leanings, and policy preferences have moved into closer alignment, individuals’ political attitudes have become more internally consistent and more distinct from those of partisans on the other side, leaving ordinary Republicans and Democrats in disagreement on a growing range of issues.Political cleavages that once split the public in diverse ways, Jacobson continued,now tend to coincide. Affective reactions to parties and candidates have diverged, largely because of partisans’ growing antipathy toward the other party’s adherents and leaders. Widening demographic differences between the parties’ identifiers — differences in race, age, sex, religiosity, education, community and marital status — have also contributed to partisan “tribalism,” as has the increasing partisan homogeneity of the states and districts.Sorting, in Jacobson’s view,is a primary source of affective polarization. If people on the other side are consistently “wrong” on all the important political questions, you’ll find it harder to like or respect them. This is especially the case if some of the issues have a large moral component (abortion, civil rights), where the other side is viewed as not just misguided but immoral, even evil.Demographic sorting by race, education and geography “contributes to affective polarization by clarifying who is ‘us’ and who is ‘them,’” Jacobson wrote: “The more the parties appear to be differentiated by morality and identity, the more intense the partisan conflict and the greater the likelihood of violence.”Jacobson was not sympathetic to the findings in Fowler’s “Moderates” paper:My analysis is orthogonal to the findings of “Moderates.” Fowler et al. find that electorate is largely moderate in ideology, but they have little to say about how their findings relate to the data (like Hare’s) that regularly find that Americans have become increasingly divided along party lines on ideology, issues, and party affect, and so forth.Jacobson stressed that the authors of “Moderates”are all first-rate scholars and their methodological skills far surpass my own, so I take the empirical work seriously. But they haven’t yet related it to the wider literature or explained why so many centrist voters seem unable to elect centrists, or why it is when there is a national tide running against a party, it’s mostly moderates who lose.As a sign of the schisms in the American electorate, Jacobson recalled a defeated congressman “complaining that voters showed their desire for moderating national politics by voting out all the moderates.”There is another key factor underpinning growing polarization and the absence of moderate politicians.“Most legislative polarization is already baking into the set of people who run for office,” Andrew Hall, a political scientist at Stanford, wrote in his book, “Who Wants to Run: How the Devaluing of Political Office Drives Polarization”: “Indeed, when we look at the ideological positions of who runs for the House, we see the set of all candidates — not just incumbents — has polarized markedly since 1980.”This trend results from the fact that since “the winning candidate gets to influence ideological policies” in increasingly polarized legislatures and the Congress, “the ideological payoffs of running for office are not equal across the ideological spectrum.” As a result, “when costs of running for office are high or benefits of holding office are low, more moderate candidates are disproportionately less likely to run.”In other words, polarization has created its own vicious circle, weeding out moderates, fostering extremists and constraining government action even in times of crisis.The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.Follow The New York Times Opinion section on Facebook, Twitter (@NYTopinion) and Instagram. More