Polls and Public Opinion

Affective polarization — “a poisonous cocktail of othering, aversion and moralization” — has prompted an explosion of research as the threat to democratic norms and procedures mount.Intensely felt divisions over race, ethnicity and culture have become more deeply entrenched in the American political system, reflected in part in the election denialism found in roughly a third of the electorate and in state legislative initiatives giving politicians the power to overturn election results.Many researchers have begun to focus on this question: Is there a causal relationship between the intensification of hostility between Democrats and Republicans and the deterioration of support for democratic standards?“Growing affective polarization and negative partisanship,” Jennifer McCoy and Murat Somer, political scientists at Georgia State University and Koç University-Istanbul, write in a 2019 essay, “Toward a Theory of Pernicious Polarization and How It Harms Democracies: Comparative Evidence and Possible Remedies,”contribute to a perception among citizens that the opposing party and their policies pose a threat to the nation or an individual’s way of life. Most dangerously for democracy, these perceptions of threat open the door to undemocratic behavior by an incumbent and his/her supporters to stay in power, or by opponents to remove the incumbent from power.What is affective polarization? In 2016, Lilliana Mason, a political scientist at Johns Hopkins, wrote that when a voter’s “partisan social identity” merges with his or her racial, religious, sexual and cultural identities, “these various identities work together to drive an emotional type of polarization that cannot be explained by parties or issues alone.”Mason argues that “threats to a party’s status tend to drive anger, while reassurances drive enthusiasm” so thata party loss generates very negative, particularly angry, emotional reactions. This anger 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.One optimistic proposal to reduce partisan animosity is to focus public attention on the commonality of Democratic and Republican voters in their shared identity as Americans. Matthew Levendusky, a political scientist at the University of Pennsylvania, has written extensively on this subject, including in his 2018 paper “Americans, Not Partisans: Can Priming American National Identity Reduce Affective Polarization?” and in his soon-to-be-published book, “Our Common Bonds: Using What Americans Share to Help Bridge the Partisan Divide.”“I show,” Levendusky contends in his 2018 paper, “that when subjects’ sense of American national identity is heightened, they come to see members of the opposing party as fellow Americans rather than rival partisans. As a result, they like the opposing party more, thereby reducing affective polarization.”There are serious problems, however, with a depolarization strategy based on American identity, problems that go to the heart of the relentless power of issues of race, ethnicity and immigration­ to splinter the electorate.In their December 2022 paper, “ ‘American’ Is the Eye of the Beholder: American Identity, Racial Sorting, and Affective Polarization among White Americans,” Ryan Dawkins and Abigail Hanson write thatWhite Democrats and White Republicans have systematically different ideas about what attributes are essential to being a member of the national community. Second, the association between partisanship and these competing conceptions of American identity among White Americans has gotten stronger during the Trump era, largely because of Democrats adopting a more racially inclusive conception of American identity. Lastly, appeals to American identity only dampen out-partisan animosity when the demographic composition of the opposing party matches their racialized conception of American identity. When there is a mismatch between people’s racialized conception of American identity and the composition of the opposition party, American identity is associated with higher levels of partisan hostility.Dawkins and Hanson acknowledge that “national identity is perhaps the only superordinate identity that holds the promise of uniting partisans and closing the social distance between White Democrats and White Republicans,” but, they continue,If conceptions of national identity itself become the subject of the very sorting process that is driving affective polarization, then it can no longer serve as a unifying identity that binds the entire country together. In fact, frames that highlight the association of American identity to historic norms of whiteness can ultimately divide the country further, especially as the United States transitions into a majority-minority country. Indeed, continued demographic change will likely make the schism between White Democrats and White Republicans wider before things have any hope to improve.I asked Levendusky about the Dawkins-Hanson paper. He replied by email that he was now “convinced that there is no simple path from animosity (or affective polarization) to far downstream outcomes (albeit important ones)” — adding that “there’s a long way from ‘I dislike members of the other party’ to ‘I will vote for a candidate who broke democratic norms rather than a candidate from the other party’ and the process is likely complex and subtle.”In an August 2022 paper, “Does Affective Polarization Undermine Democratic Norms or Accountability? Maybe Not,” David E. Broockman, a political scientist at Berkeley, Joshua L. Kalla, a political scientist at Yale, and Sean J. Westwood, a political scientist at Dartmouth, pointedly reject the claim made by a number of scholars “that if citizens were less affectively polarized, they would be less likely to endorse norm violations, overlook copartisan politicians’ shortcomings, oppose compromise, adopt their party’s views, or misperceive economic conditions. A large, influential literature speculates as such.”Instead, Broockman, Kalla and Westwood contend, their own studies “find no evidence that these changes in affective polarization influence a broad range of political behaviors — only interpersonal attitudes. Our results suggest caution about the widespread assumption that reducing affective polarization would meaningfully bolster democratic norms or accountability.”Broockman and his co-authors measured the effect of reducing affective polarization on five domains: “electoral accountability, adopting one’s party’s policy positions, support for legislative bipartisanship, support for democratic norms, and perceptions of objective conditions.”“Our results,” they write, “run contrary to the literature’s widespread speculation: in these political domains, our estimates of the causal effects of reducing affective polarization are consistently null.”In an email, Westwood argued that the whole endeavor “to fix anti-democratic attitudes by changing levels of partisan animosity sounds promising, but it is like trying to heal a broken bone in a gangrenous leg when the real problem is the car accident that caused both injuries in the first place.”Westwood’s point is well-taken. In a country marked by battles over sex, race, religion, gender, regional disparities in economic growth, traditionalist-vs-postmaterialist values and, broadly, inequality, it is difficult to see how relatively short, survey based experiments could produce a significant, long-term dent in partisan hostility.Jan G. Voelkel, a sociologist at Stanford, and eight of his colleagues, report similar results in their October 2022 article “Interventions Reducing Affective Polarization Do Not Necessarily Improve Anti-democratic Attitudes.” “Scholars and practitioners alike,” they write, “have invested great effort in developing depolarization interventions that reduce affective polarization. Critically, however, it remains unclear whether these interventions reduce anti-democratic attitudes, or only change sentiments toward outpartisans.”Why?Because much prior work has focused on treating affective polarization itself, and assumed that these interventions would in turn improve downstream outcomes that pose consequential threats to democracy. Although this assumption may seem reasonable, there is little evidence evaluating its implications for the benefits of depolarization interventions.In “Megastudy Identifying Successful Interventions to Strengthen Americans’ Democratic Attitudes,” a separate analysis of 32,059 American voters “testing 25 interventions designed to reduce anti-democratic attitudes and partisan animosity,” however, Voelkel and many of his co-authors, Michael N. Stagnaro, James Chu, Sophia Pink, Joseph S. Mernyk, Chrystal Redekopp, Matthew Cashman, James N. Druckman, David G. Rand and Robb Willer significantly amended their earlier findings.In an email, Willer explained what was going on:One of the key findings of this new study is that we found some overlap between the interventions that reduced affective polarization and the interventions that reduced one specific anti-democratic attitude: support for undemocratic candidates. Specifically, we found that several of the interventions that were most effective in reducing American partisans’ dislike of rival partisans also made them more likely to say that they would not vote for a candidate from their party who engaged in one of several anti-democratic actions, such as not acknowledging the results of a lost election or removing polling stations from areas that benefit the rival party.Voelkel and his co-authors found that two interventions were the most effective.The first is known as the “Braley intervention” for Alia Braley, a political scientist at Berkeley and the lead author of “The Subversion Dilemma: Why Voters Who Cherish Democracy Participate in Democratic Backsliding.” In the Braley intervention, participants are “asked what people from the other party believe when it comes to actions that undermine how democracy works (e.g., using violence to block laws, reducing the number of polling stations to help the other party, or not accepting the results of elections if they lose).” They are then given “the correct answer” and “the answers make clear the other party does not support actions that undermine democracy.”The second “top-performing intervention” was to give participants “a video showing vivid imagery of societal instability and violence following democratic collapse in several countries, before concluding with imagery of the Jan. 6 U.S. Capitol attack.”“To our knowledge,” Willer wrote in his email, “this is the first evidence that the same stimuli could both reduce affective polarization and improve some aspect of Americans’ democratic attitudes, and it suggests these two factors may be causally linked, more than prior work — including our own — would suggest.”Kalla disputed the conclusions Willer drew from the megastudy:The most successful interventions in the megastudy for reducing anti-democratic views were interventions that directly targeted those anti-democratic views. For example, Braley et al.’s successful intervention was able to reduce anti-democratic views by correcting misperceptions about the other party’s willingness to subvert democracy.This intervention, Kalla continued,was not about affective polarization. What this suggests is that for practitioners interested in reducing anti-democratic attitudes, they should use interventions that directly speak to and target those anti-democratic views. As our work finds and Voelkel et al. replicates, obliquely attempting to reduce anti-democratic views through the causal pathway of affective polarization does not appear to be a successful strategy.I sent Kalla’s critique to Willer, who replied:I agree with Josh’s point that the most effective interventions for reducing support for undemocratic practices and candidates were interventions that were pretty clearly crafted with the primary goal in mind of targeting democratic attitudes. And while we find some relationships here that suggest there is a path to reducing support for undemocratic candidates via reducing affective polarization, the larger point that most interventions reducing affective polarization do not affect anti-democratic attitudes still stands, and our evidence continues to contradict the widespread popular assumption that affective polarization and anti-democratic attitudes are closely linked. We continue to find evidence in this newest study against that idea.One scholar, Herbert P. Kitschelt, a political scientist at Duke, contended that too much of the debate over affective polarization and democratic backsliding has been restricted to the analysis of competing psychological pressures, when in fact the scope in much larger. “The United States,” Kitschelt wrote in an email,has experienced a “black swan” confluence, interaction and mutual reinforcement of general factors that affect all advanced knowledge societies with specific historical and institutional factors unique to the U.S. that have created a poisonous concoction threatening U.S. democracy more so than that of any other Western society. Taken together, these conditions have created the scenario in which affective polarization thrives.Like most of the developed world, the United States is undergoing three disruptive transformations compounded by three additional historical factors specific to the United States, Kitschelt suggests. These transformations, he wrote, are:“The postindustrial change of the occupational structure expanding higher education and the income and status educational dividend, together with a transformation of gender and family relations, dismantling the paternalist family and improving the bargaining power of women, making less educated people — and especially males — the more likely socio-economic and cultural losers of the process.”“The expansion of education goes together with a secularization of society that has undercut the ideological foundations of paternalism, but created fierce resistance in certain quarters.”“The sociocultural and economic divisions furthermore correlate with residential patterns in which the growing higher educated, younger, secular and more gender-egalitarian share of the population lives in metropolitan and suburban areas, while the declining, less educated, older, more religious and more paternalists share of the population lives in exurbia or the countryside.”The three factors unique to this country, in his view, are:“The legacy of enslavement and racial oppression in the United States in which — following W.E.B. DuBois — the white lower class of less skilled laborers derived a ‘quasi-wage’ satisfaction from racist subordination of the minority, the satisfaction of enjoying a higher rank in society than African Americans.”“The vibrancy of evangelical ‘born again’ Christianity, sharply separated from the old European moderate, cerebral mainline Protestantism. The former attracts support over-proportionally among less educated people, and strictly segregates churches by race, thereby making it possible to convert white Evangelical churches into platforms of white racism. They have become political transmission belts of right-wing populism in the United States, with 80 percent of those whites who consider themselves ‘born again’ voting for the Trump presidential candidacy.”“The institutional particularities of the U.S. voting system that tends to divide populations into two rival parties, the first-past-the-post electoral system for the U.S. legislature and the directly elected presidency. While received wisdom has claimed that it moderates divisions, under conditions of mutually reinforcing economic, social, and cultural divides, it is likely to have the opposite effect. The most important additional upshot of this system is the overrepresentation of the countryside (i.e. the areas where the social, economic, and cultural losers of knowledge society tend to be located) in the legislative process and presidential elections/Electoral College.”Kitschelt argues that in order to understand affective polarization it is necessary to go “beyond the myopic and US-centric narrow vision field of American political psychologists.” The incentives “for politicians to prime this polarization and stoke the divides, including fanning the flames of affective polarization, can be understood only against the backdrop of these underlying socio-economic and cultural legacies and processes.”Kitschelt is not alone in this view. He pointed to a 2020 book, “American Affective Polarization in Comparative Perspective,” by Noam Gidron, James Adams and Will Horne, political scientists at Harvard, the University of California-Davis and Georgia State University, in which they make a case thatAmericans’ dislike of partisan opponents has increased more rapidly than in most other Western publics. We show that affective polarization is more intense when unemployment and inequality are high, when political elites clash over cultural issues such as immigration and national identity and in countries with majoritarian electoral institutions.Writing just before the 2020 election, Gidron, Adams and Horne point out that theissue of cultural disagreements appears highly pertinent in light of the ongoing nationwide protests in support of racial justice and the Black Lives Matter movement which has sparked a wider cultural debate over questions relating to race, police funding and broader questions over interpretations of America’s history. In a July 4th speech delivered at Mt. Rushmore, President Trump starkly framed these types of “culture war” debates as a defining political and social divide in America, asserting “our nation is witnessing a merciless campaign to wipe out our history, defame our heroes, erase our values and indoctrinate our children.”The study of affective polarization sheds light on how vicious American politics has become, and on how this viciousness has enabled Trump and those Republicans who have followed his lead, while hurting Democrats whose policy and legislative initiatives have been obstructed as much as they have succeeded.Richard Pildes, a professor of constitutional law at N.Y.U., addressed this point when he delivered the following remarks from his paper “Political Fragmentation in Democracies of the West” in 2021 at a legal colloquium in New York:There is little question that recent decades have seen a dramatic decline in the effectiveness of government, whether measured in the number of important bills Congress is able to enact, the proportion of all issues people identity as most important that Congress manages to address, or the number of enacted bills that update old policies enacted many decades earlier. Social scientists now write books with titles like Can America Govern Itself? Longitudinal data confirm the obvious, which is the more polarized Congress is, the less it enacts significant legislation; in the ten most polarized congressional terms, a bit more than 10.6 significant laws were enacted, while in the ten least polarized terms, that number goes up 60 percent, to around 16 significant enactments per term. The inability of democratic governments to deliver on the issues their populations care most about poses serious risks.What are the chances of reversing this trend?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