Uncivil Agreement: How Politics Became Our Identity

Political scientist Lilliana Mason argues that the growing hostility between American Democrats and Republicans is driven less by genuine policy disagreements than by the alignment of social identities, including race, religion, and ideology, behind the two political parties. She calls this process "social sorting" and contends that it has transformed American politics into a team-based competition in which victory matters more than governing.

Mason opens by recounting the 1954 Robbers Cave experiment, in which social psychologist Muzafer Sherif divided 22 demographically similar fifth-grade boys into two teams at a state park in Oklahoma. Kept separate during the first week, the boys formed group identities from scratch. When the teams met and competed, the mere fact of group membership drove them to name-calling, cabin raids, and fistfights. In a subsequent test, every boy overestimated the number of beans collected by his own teammates, even though the experimenters displayed the same quantity each time. Mason draws a direct parallel to contemporary partisans, arguing that Americans increasingly distort objective reality along partisan lines. She cites research showing that partisan stereotyping increased by 50 percent between 1960 and 2010, and frames the 2016 election of Donald Trump as the culmination of social sorting.

Mason establishes a conceptual framework separating social-identity-based elements of political concepts from their issue-based counterparts. She distinguishes "social polarization," the increasing social distance between partisans through bias, emotion, and activism, from "issue-based polarization," the divergence of actual policy positions. She makes a parallel distinction between "identity-based ideology," the sense of belonging to groups called liberal or conservative, and "issue-based ideology," the policy attitudes a person holds. In her own 2011 survey, conservative identification correlated with conservative policy positions at only r = 0.24, demonstrating that ideological identity and policy positions are only weakly linked. This framework is foundational: Americans can feel deeply divided even when their policy preferences overlap substantially. Mason argues that when multiple identities converge behind one party, they form a "mega-identity" with amplified psychological and behavioral effects.

Mason traces social sorting from the 1950s through 2012, noting that political scientists long argued American democratic stability depended on party loyalties not aligning with social divisions. In the 1950s, the parties sent mixed social signals: Conservative southern Democrats coexisted with liberal northern Democrats. This began to change after the Civil Rights Act of 1964, when conservative white southerners gradually migrated to the Republican Party. By 2013, racial identity alone predicted party identity. A religious divide followed in the late 1980s when the Christian Coalition allied with Republicans, and its Contract with the American Family, a conservative family-policy agenda, was incorporated into the Republican platform by 2000. Using American National Election Studies (ANES) data, Mason shows that old partisan divisions along Protestant/Catholic and union/nonunion lines largely disappeared, replaced by ideological, religious, and racial divides. By 2012, Republicans were on average 30 percentage points more white than Democrats. The increase in partisan-ideological-identity differences was more than twice as large as the increase in partisan policy differences. Three non-policy forces facilitated sorting: declining civic engagement, clearer partisan cues from leaders, and partisan media growth.

Mason next examines partisan prejudice, arguing that party identity alone drives significant bias independently of policy disagreement. ANES feeling-thermometer data, which measure how warmly respondents rate each party on a 100-degree scale, show that partisan preference for one's own party increased by 10 percentage points between 1984 and 2012, while policy extremity did not rise comparably. Even Democrats who hold consistently conservative positions rate the Democratic Party warmly, while Republicans who agree with Democrats on most issues rate it coolly. Using a social-distance scale measuring willingness for social closeness, friendship, or marriage across party lines, Mason finds that Americans are 36 percentage points less willing to marry an outgroup partisan. Regression analysis confirms that a strong partisan with moderate issue positions is still about 29 percentage points more willing to spend time with ingroup than outgroup partisans.

Mason then demonstrates that social sorting magnifies prejudice beyond what partisanship alone produces. Drawing on social psychologist Marilynn Brewer's research, which found that people with aligned identities are less tolerant than those with cross-cutting identities, Mason shows that among strong partisans, those with well-sorted identities rate the two parties 56 degrees apart on feeling thermometers, compared to 37 degrees for those with cross-cutting identities. Using a richer sorting scale from her 2011 YouGov survey incorporating partisan, ideological, racial, religious, and Tea Party identities (the Tea Party being a conservative movement within Republican politics), well-sorted strong partisans rate the parties 78 degrees apart versus 58 for cross-pressured partisans, meaning those whose identities pull toward both parties. Moving from the most cross-cutting to the most well-sorted identities increases partisan bias by 64 degrees, while moving from the most moderate to the most extreme policy positions increases it by only 24 degrees, making sorting nearly three times more powerful than issue extremity.

Turning to emotional reactivity, Mason argues that social sorting drives increasing anger independently of policy concerns. ANES data show that partisan anger at the outgroup presidential candidate rose from about 40 percent in 1980 to 60 percent by 2012. Using a panel study, she finds that individuals whose sorting increased reported a 28 percentage point increase in anger, compared to 17 points for those who became less sorted, while issue intensity changed by only about one percentage point. A critical finding is that people with the most cross-cutting identities respond to threatening political messages with essentially no anger at all, suggesting the electorate grows angrier as the calming influence of cross-cutting identities disappears.

Mason extends these findings to political activism, showing that sorting drives increased participation motivated by identity defense rather than policy goals. In her YouGov data, moving from the most cross-cutting to the most well-sorted identities increases prior activism by about 30 percentage points and intended future activism by about 43 percentage points. Drawing on the argument that cross-pressured voters provide necessary democratic flexibility, Mason contends that the shift to a well-sorted electorate reduces responsiveness to changing conditions, as the most active citizens are motivated by team loyalty and anger rather than thoughtful policy evaluation.

In her final chapter, Mason explores potential remedies while acknowledging barriers. Contact theory, rooted in social psychologist Gordon Allport's 1954 work, suggests intergroup friendships can reduce prejudice, but partisans increasingly occupy separate social arenas. Social norms set by party leaders could help, but leaders are incentivized to maintain conflict. Superordinate goals, which resolved the Robbers Cave experiment when the boys cooperated on shared problems, might unite partisans, but declining trust undermines this possibility. Mason also notes that projected demographic shifts toward a majority-minority population by 2052 may initially intensify sorting, since research shows that informing white respondents about these changes increases their preference for spending time with other whites. She raises the possibility of an "unsorting" through a rift in one major party, pointing to the Tea Party and the Freedom Caucus, a hard-right bloc of House Republicans, as potential catalysts.

Mason concludes that social sorting has been harmful to American democracy. While it has simplified electoral choices and increased engagement, it has produced an electorate that is more biased, angrier, and more active on behalf of prejudice than policy disagreements warrant, and that is increasingly unresponsive to changing information and real national problems.