Beyond Elections: Why Democracy Depends on Trust, Restraint, and Collective Action

Does democracy work? What can cause it to fail? And why do people who support democracy so often hesitate to offer a defense when it actually matters?
These and similar questions inspired two days of discussion at the Institution for Social and Policy Studies among leading experts in democratic backsliding.
“For a long time, the proposition that a place like the United States would flirt with authoritarianism was unthinkable from most political scientists,” said Milan Svolik, Institution for Social and Policy Studies faculty fellow and Elizabeth S. & A. Varick Stout Professor of Political Science. “But our current politics is challenging us — sometimes on a daily basis — to examine our confidence in democratic stability in many countries around the world, including the United States.”
For the second year, Svolik gathered Americanists and comparativists — once studying seemingly different political worlds — to rethink what political science truly knows about democratic stability. Participants often acknowledged the fading of American exceptionalism, as patterns once studied abroad increasingly illuminate domestic politics.
The event was sponsored by Democratic Innovations, an ISPS program designed to identify and test new ideas for improving the quality of democratic representation and governance.
“This conference advances the vision of the Democratic Innovations program,” said Alan Gerber, director of ISPS and Sterling Professor of Political Science. “This impressive group of scholars are demonstrating how studying American politics through a comparative lens provides essential context for understanding our present moment and the challenges ahead.”
Carlo Prato, associate professor of political science at Columbia University, and Gretchen Helmke, Thomas H. Jackson Distinguished University Professor at the University of Rochester, discussed how people often want to defend democracy but will not act unless they believe others will act too.
Across cases, speakers returned to a shared worry that people hesitate to defend democracy when they fear retaliation if power later changes hands.
“Resistance to backsliding is a coordination problem,” Prato said, noting how strong party discipline among elected officials can make resistance more likely — not less — when punishment depends on leadership survival. “It’s not just whether leaders can punish — it’s what happens to punishment if resistance succeeds.”
Speakers at the conference showed how laws fail when actors doubt collective follow-through. And how institutions fail not when rules disappear, but when people stop expecting those rules to be enforced fairly.
In addition, speakers demonstrated how elites and voters often tolerate or enable democratic violations because they benefit. In this way, democratic erosion arises less out of confusion or ignorance than calculated tradeoffs.
“Democratic backsliding is not usually about democracy itself,” said Thomas Pepinsky, Ph.D., ’07, the Walter F. LaFeber Professor of Government and Public Policy at Cornell University “It’s about using political system change to remake the social order.”
Pepinsky argued that democratic backsliding is motivated less by nihilism or the simple desire to win elections unfairly, and more as a struggle over who counts as a political subject, how society is organized, and what the state is for.
He concluded democracy is endangered not only when elections are manipulated, but when pluralism itself becomes politically contested.
Several speakers emphasized that public opinion is not merely a background condition but an active force shaping whether elites resist or accommodate democratic erosion.
Helmke discussed how restraint and “pulling punches” can protect democracy or entrench corruption and impunity. And she showed that countries prosecute their leaders only when they are weak and political retaliation becomes unlikely, not when justice demands it most.
“Impunity is not the absence of politics,” Helmke said. “It’s a perverse form of inter‑party cooperation.”
Sean Westwood, associate professor of government at Dartmouth College, discussed how proponents of electoral reforms often over-promise results from changes to election mechanics. He showcased a large systematic review of ranked-choice voting, showing weak, null, or negative effects on its ability to deliver multiparty democracy, moderation, or trust among voters.
“Ranked choice voting is being marketed based on benefits that it cannot plausibly deliver,” Westwood said. “American democracy is far too fragile to ‘just try things’ and hope for the best.”
Gregory Huber, acting director of ISPS and Forst Family Professor of Political Science, emphasized the danger of political scientists offering policy advice without solid evidence. Rather than offering a single cure, conference attendees repeatedly emphasized humility: democratic decline rarely has simple institutional fixes.
“The stakes of giving bad advice are high,” Huber said. “We want the world to view political scientists as serious scholars who evaluate evidence in an unbiased way, not as advocates with credentials.”
Westwood and others stressed how democratic procedures that work in high-trust societies can backfire when trust has already broken. Legitimacy, he said, depends on perceptions and not just correctness.

Svolik expressed enthusiasm about the field’s progress.
“What we have now is higher‑quality work that’s really trying to confront the doubts we have about the stability of democratic institutions and the mechanisms behind that,” Svolik said.
And while he recognized that political science cannot keep up with the fast-moving nature of political actions and breaking news, he felt confident in its role.
“What we can do is translate current events into more fundamental questions about politics that we don’t yet understand,” he said, proposing examples such as which institutions are best for democratic stability and what voters really want. “Those are questions that will still be here a hundred years from now.”