Who among us are the most willing to jettison democratic elections? Which voters not only detest their political adversaries but also long for their destruction?
These questions are now at the heart of political science.
Five scholars have capitalized on new measurement techniques to identify partisan sectarian voters, a category that they said “does indeed predict antidemocratic tendencies.”
In their recent paper “Partisan Antipathy and the Erosion of Democratic Norms” Eli Finkel and Jamie Druckman of Northwestern, Alexander Landry of Stanford, Jay Van Bavel of N.Y.U. and Rick H. Hoyle of Duke made the case that earlier studies of partisan hostility used ratings of the two parties on a scale of 0 (cold) to 100 (very warm) but that that measure failed to show a linkage between such hostility and antidemocratic views.
In fact, the five scholars wrote, “Partisan antipathy is indeed to blame, but the guilty party is political sectarianism,” not the thermometer rating system:
Insofar as people experience othering, aversion, and moralization toward opposing partisans, they are more likely to support using undemocratic tactics to pass partisan policies: gerrymandering congressional districts, reducing the number of polling stations in locations that support the opposing party, ignoring unfavorable court rulings by opposition-appointed judges, failing to accept the results of elections that one loses, and using violence and intimidation toward opposing partisans.
Who, then, falls into this subset of partisan sectarians?
The authors cited nine polling questions that ask voters to assess their feelings toward members of the opposition on a scale of 1 to 6, with 6 the most hostile.
The first set of questions measured what the authors called othering. The most extreme answers were:
I felt as if they and I are on separate planets.
I am as different from them as can be.
It’s impossible for me to see the world the way they do.
The second set of questions measured aversion:
My feelings toward them are overwhelmingly negative.
I have a fierce hatred for them.
They have every negative trait in the book.
The third set of questions measured moralization:
They are completely immoral.
They are completely evil in every way.
They lack any shred of integrity.
How, then, to identify voters high in antidemocratic views? Prototypical questions here were: “Democratic/Republican governors should ignore unfavorable court rulings by Republican/Democratic-appointed judges” and “Democrats/Republicans should not accept election results if they lose.”
The Finkel et al. analysis linking partisan sectarianism to antidemocratic views received strong support but not a wholesale endorsement from Nicolas Campos and Christopher Federico, political scientists at the University of Minnesota, who modified the Finkel approach.
“Following Finkel et al.,” Campos and Federico wrote in their recent paper “A New Measure of Affective Polarization,” partisan hostility consists of “more than an undifferentiated tendency to feel more negatively about out-partisans than in-partisans, and we believe that the broad concepts of othering, aversion and moralization provide a good starting point for identifying the multiple components.”
In an email elaborating on arguments in their paper, Federico wrote that while he and Campos used the same othering, aversion and moralization model, they changed the focus, most especially in the case of moralization:
Finkel et al. conceptualize moralization as seeing the out-party as evil or morally corrupt. Instead, we focus on the belief that one’s own partisan identity is rooted in fundamental moral values, beliefs about right and wrong, and so on. Research on “moral conviction” in psychology strongly suggests that intolerance and inflexibility toward those who disagree (and a willingness to reject democratic outcomes that differ from one’s own preferences) is rooted in a tendency to moralize one’s preferences.
In his email, Federico argued that the third dimension of partisan animosity, “aversion — a tendency to dislike and want to distance the out-party from the in-party — is the strongest predictor of antidemocratic attitudes and support for violence.”
Do partisan sectarians demonstrate unique demographic and other characteristics?
Campos replied to that question by email: “There are no consistent patterns nor large effects in terms of race, ethnicity, age, income or education. The only consistent pattern we find is that Democrats are slightly higher in antidemocratic views in our samples.”
Shanto Iyengar, a Stanford political scientist whose 2012 paper, “Affect, Not Ideology: A Social Identity Perspective on Polarization” (written with Gaurav Sood and Yphtach Lelkes), is a seminal work in the study of partisan hostility, stressed in an email responding to my queries that one of the most significant dangers it poses is
the weakened ability of partisans to hold their leaders accountable. The crux of the problem is that partisans have come to view the opposing party in such harsh terms that they are unwilling to sanction leaders of their own party who engage in unethical or illegal activity.
In the most recent ANES survey, we examined partisans’ willingness to support candidates with questionable credentials. The pilot study included four questions asking respondents whether a set of unethical or illegal actions would “keep you from voting for a candidate for public office.” The actions in question included conviction on a felony charge, acceptance of a bribe from a foreign government, mishandling of classified documents and facing accusations of sexual harassment.
Three of the behaviors in question arguably are associated with Donald Trump, while only one can potentially be linked to Joe Biden. We found a huge partisan divide in responses to these questions with Republicans proving much more likely to ignore unethical/illegal behavior — 62 percent were prepared to vote for a candidate facing allegations of sexual harassment, and more than 40 percent would vote for a convicted felon and candidate who compromised national security.
Overall, Iyengar wrote, the data “would seem to bear out Trump’s now infamous claim that he could stand in the middle of Fifth Avenue and shoot somebody without losing any voters.”
In a separate study, “Unsorted Partisanship and Antidemocratic Orientation in the American Public,” Ariel Malka of Yeshiva University, Thomas Costello of M.I.T. and Federico found that certain types of Democrats and Republicans are most drawn to antidemocratic views:
Cultural conservatism and out-party (pro-Republican) favorability are reliably associated with antidemocratic orientation among Democrats, with effect sizes exceeding those of key co-variates such as education. Among Republicans, left-leaning economic attitudes are reliably associated with antidemocratic orientation.
In other words, those whose views conflict with those of their party are most critical of democratic norms. I asked Federico if he could explain this, and he emailed back that he found no clear answer in the data but was willing to suggest two possibilities.
First:
Citizens who deviate from their own party’s position on a set of issues tend to be less politically engaged. Less politically engaged individuals also tend to be less supportive of democratic norms. So part of this may simply be that economically liberal Republicans and socially conservative Democrats are less engaged and thus less likely to have absorbed democratic norms.
Second:
Populist beliefs — i.e., a combination of cultural conservatism and economic liberalism — also tend to be associated with lower support for democratic norms. Democrats with culturally conservative attitudes and Republicans with economically liberal attitudes both fall into the populist belief pattern, so what we see in these two groups may simply reflect their greater populist bent.
While most voters voice support for democracy and fair elections, there are nuances to this comforting view.
In their 2023 paper “Professed Democracy Support and Openness to Politically Congenial Authoritarian Actions Within the American Public,” Malka and Costello explored a fundamental contradiction in American politics.
“Professed opposition to democracy was relatively rare and most common among citizens who felt disengaged from politics,” Malka and Costello wrote, “but a different pattern of findings emerged for attitudes toward flagrant, politically congenial authoritarian policy action and election subversion framed with a pro-democracy justification.”
In these cases, the authors found that “antidemocratic attitudes were relatively common, related to cultural conservatism among both Republicans and Democrats, and — consistent with an involved-but-ignorant hypothesis — highest among those who combined strong political interest with low political sophistication.”
Malka and Costello based their work on a 6,010-person survey conducted five weeks before the 2020 elections and determined that while “Republican identification was associated with more antidemocratic positions,” the effects of party identification were much smaller than those of cultural conservatism:
Cultural conservatism was a relatively strong positive predictor of support for authoritarian actions and election subversion among both Republicans and Democrats. Willingness to flagrantly degrade democracy for political ends seems to be less a matter of partisanship than a matter of outlook with respect to traditional versus progressive cultural conflict.
Perhaps most significant:
The Americans most likely to support such democracy degradation were those who combined low political knowledge with high subjective political involvement, a finding that was consistent across Republicans and Democrats. Strong involvement with politics may be favorable for giving lip service to democracy but may also energize support for politically congenial antidemocratic behavior among those who are unsophisticated.
Other scholars have reached different conclusions.
In their new book “Partisan Hostility and American Democracy” Druckman, Samara Klar of the University of Arizona, Yanna Krupnikov of the University of Michigan, Matthew Levendusky of the University of Pennsylvania and John Barry Ryan of the University of Michigan — argued that “animosity toward the other party works to undermine political accountability, since the actual behaviors become secondary to partisan signals.” In addition, “animosity consistently matters; as animosity increases, so does the belief that the other party should compromise and, even more so, that their party should not. At the highest levels of animus, partisans reject the possibility of their party compromising from their ideal point.”
At the same time, Druckman and his co-authors found that demography did not play a strong role, writing: “The most striking finding is the lack of clear relationships between demographics and animus. For education and income, there was basically no relationship with animus for either party.”
They did conclude that “respondents with higher animus were more likely to hold extreme positions and to identify as extremely liberal or conservative.”
But, they continued, “these effects were modest, not massive, and a minority of respondents at even the highest levels of animus held such beliefs. Apparently, high-animus partisans are not necessarily those who consistently hold the most extreme positions.”
Those with high levels of partisan animosity are both politically engaged and knowledgeable.
They “had more animosity, discussed politics much more frequently, posted about politics on social media much more frequently, expressed greater interest in politics by stating they were extremely or very interested and possessed substantially more political knowledge as captured by being in the top quartile of a scale of factual knowledge items.”
One question that goes beyond issues of partisanship: Are technological advances eating away at the foundations of the democratic state?
Two members of the psychology department at N.Y.U., Jay Van Bavel and Claire Robertson, have begun to explore these uncharted waters. They published two papers together this year, “Morality in the Anthropocene: The Perversion of Compassion and Punishment in the Online World” (with Azim Shariff) and “Inside the Funhouse Mirror Factory: How Social Media Distorts Perceptions of Norms” (with Kareena del Rosario).
“We argue,” Robertson wrote in an email elaborating on her work with Van Bavel and her colleagues, “that the unnatural or ‘supernormal’ scale and distance of the internet distorts basic moral processes like the instincts for compassion and punishment. It’s less that the normal responses aren’t available but more that our normal responses have unintended consequences.”
For example, Robertson wrote,
the extreme scale of social networks online are wildly different from the small, tribal groups humans evolved in. We are now able to connect with thousands of people per day, far more than our evolutionary ancestors would have interacted with in their entire lifetimes.
Even though punishing moral transgressors was adaptive in small groups by signaling to others that bad behavior is not acceptable and deterring future cheaters, the scale at which people are punished online becomes out of proportion for their crime. A poster child for this phenomenon was Justine Sacco, who tweeted a racist joke on Twitter in 2013. Historically a transgression like this might be handled with social sanctions from close others, but in this case, it resulted in over one million people publicly shaming her.
In “Inside the Funhouse Mirror Factory” Robertson explained, “We’re arguing that the structure of the internet makes more moderate opinions practically invisible online. Pew research found that about 97 percent of all political tweets were posted by the most active 10 percent of users.”
While the “people who post about politics online tend to be more ideologically extreme and are highly identified with their political party, Robertson continued, “in real life, extreme partisanship is actually rather rare.”
The distorted reality of the internet, Robertson argued, feeds into what she called “ensemble encoding,” an instinctive drive to detect and then abide by social norms: “Ensemble coding is beneficial because it is cognitively efficient, allowing people to encode a single average representation of a series of stimuli, rather than requiring encoding and memory for every item in a set.”
While an efficient method of detecting social norms in day-to-day life, ensemble coding
can become distorted online due to the structure of the normative information. False norms emerge, in part, because social media is dominated by a small number of extreme people who post only their most extreme opinions and do so at a very high volume, while more moderate or neutral opinions are practically invisible online.
The result?
“When people use this skewed online information to form a model of the average, normative opinion on an issue, their average opinion is also skewed toward the extreme.”
In a separate email, Van Bavel wrote: “We are trying to figure out a broad set of principles about how social media drives conflict and distorts our perceptions of norms.”
Humans “evolved in small, interdependent groups,” he said. “Our sense of morality is designed to navigate and regulate social life in small communities, and this simply doesn’t scale very easily to the large, synthetic world of social media.”
Normal responses are not available, Van Bavel contended, “because the internet is inherently not an in-person form of communication and, more importantly, the structure and incentives of the internet are to exaggerate moral violations and to focus on them as a dominant aspect of life.”
The human brain, Van Bavel noted,
did not evolve for online interactions but instead for the real world dynamics of interpersonal interactions. We are skilled at reading body language, understanding the social context and cooperating with others. The online environment can take some of our social instincts and exploit or distort them. This can lead to online public shaming or ineffective collective action, rather than to the type of third-party punishment that is often necessary for fostering cooperation.
Where does that leave us?
When the massive scale of the internet interacts with people’s instinct to feel compassion for victims of moral transgressions, it can result in compassion fatigue.
The processes Robertson and Van Bavel described have entered our politics. How else to explain the habituation of millions of voters, perhaps even a majority, to the ever-growing list of Trump’s transgressions?
“Nearly five billion people are on social media,” Van Bavel wrote:
The average user is online for three hours a day and scrolls through roughly 300 feet of news content — the height of the Statue of Liberty.
We are just starting to understand how this impacts our relationships and sense of identity, let alone society and democracy. Although the effects might be small on a daily or weekly basis, there is reason to think it might matter a great deal on the scale of years or decades due to the large user base.
The post Political Scientists Want to Know Why We Hate One Another This Much appeared first on New York Times.