An illustration of people working on a building and gardening.

Deliberation Nation

How improving civic literacy can reinforce democracy.

By Jill Patton
Illustrations by Eric Comstock
March 2022

Read: Civil Disagreement


Read: Time Out


Read: Isn’t This Stuff Taught in School?


Most days, you can cram a couple hundred people onto bare wooden benches in a stuffy courtroom and behold the state of American democracy.

It’s jury selection.

The judge will spend the day fielding excuses for why almost no one can serve: There are work obligations, travel plans, ailing parents, infants to tend. And then there are all manners of bias that people swear they couldn’t possibly see past to render an honest verdict.

Oh, we think it’s good in theory—about two-thirds of U.S. adults say serving on a jury “is part of what it means to be a good citizen,” according to a 2017 Pew Research Center survey. But in practice . . . well, cue the groan.

So why are we groaning? “A problem with contemporary democracy is that we don’t really realize that the stakes are high,” says Josiah Ober, a Stanford professor of political science and of classics. We’ve got a lot going on, and our civic duties, from serving on a jury to studying the propositions on this year’s ballot, often feel like they get in the way of the rest of life. “We see politics as a game instead of as a highly consequential activity that bears heavily on our flourishing or languishing.” Ober, an expert in democratic theory, sees a failure with civic education. “[Civic education] is, in short, a fairly presented argument for the regime’s legitimacy,” he writes in Demopolis: Democracy Before Liberalism in Theory and Practice, his 2017 defense of democratic principles. Done right, that education overrides people’s desire to check out and focus on other things. He says modern democracies, including the United States, are falling down on the job.

“It’s a collective action problem,” says James Fishkin, a professor of communication and author of the 2018 book Democracy When the People Are Thinking: Revitalizing Our Politics Through Public Deliberation. “There’s a lot of [academic] literature that shows that individual voters have an incentive for rational ignorance.” In other words, why should anyone spend a lot of time learning about political issues when the impact of one person’s vote is so small? Fishkin was stewing about that very problem in the late 1980s when he was a fellow at Stanford’s Center for Advanced Study in the Behavioral Sciences. “I was worried about the upcoming primary season, and public opinion polls being so superficial, and the lack of public will formation in our democracy,” he recalls. Then an idea came into his head: a new kind of public opinion poll that integrates the act of group deliberation in order to yield more informed judgments. More than 30 years later, his technique—called deliberative polling—has been used some 120 times in 32 countries on topics both narrow and national as a way to gauge what the people at large might think if they fully weighed the range of arguments face-to-face in collaboration with others.

In the meantime, new challenges to democracy have emerged, chief among them a lack of coherence in civic knowledge. “News networks and newspapers used to be shared information sources, but we now have a fragmented partisan press and advocacy journalism,” Fishkin says. “The line between opinion and reporting is blurred. In our news feeds, we only listen to the like-minded. The change in technology has increased extreme partisan polarization and further entrenched people in their views.” In response, Fishkin is boosting the reach of deliberative polling and taking the practice of civic deliberation to the masses in hopes that it can help knit together what he calls our “fragmented public sphere.”

Illustration of people cleaning the Statue of Liberty

Democracy’s Foundations

When America’s founders sketched out their plan for a democracy, they relied on the thinking of Aristotle and his idea that citizens of, say, Athens could rule and be ruled on their own terms. In Politics, penned in the 4th century BCE, Aristotle explained that people would need to take on certain civic duties, and they would have to know enough about the government to perform them. So, Ober says, the Athenians developed a formal system of civic education, and the community elected highly respected members to instruct young citizens. Every free adult male, in turn, was required to spend two years in public service, including military training. Those over 30 could also sit for a year on the Council of 500, which handled daily affairs and recommended agenda items to the legislative assembly. Requiring participation meant there was “skin in the game,” Ober says. Your vote to, for example, go to war with Sparta had immediate, obvious and often personal ramifications.

But Athens wasn’t like early America. About 40,000 of the city’s residents were considered citizens, so when the legislative assembly met—some 40 times a year—roughly 5,000 tunic-clad men showed up at any given time to consider the agenda and vote. Business was conducted out loud, with speeches and debate. It was a direct democracy. The population of the United States in 1776, by comparison, counted 2.5 million, and the country spanned more than 1,000 miles, from present-day Maine to Georgia. The founders opted for a representative democracy: Power lay with the people—or, at least, with the property-owning men eligible to vote at the nation’s founding—but representatives would be the voices in the room.

When James Madison, Alexander Hamilton and John Jay laid out arguments for the government’s formation through the Federalist Papers, their views traveled directly to readers via two New York newspapers. But few publications at the time circulated more than a thousand copies, printing was done laboriously on hand presses, and literacy rates were low. Public schools didn’t gain hold until 1830 (and elementary school wouldn’t become compulsory nationwide until 1918). Most of what citizens knew about government affairs came from politicians themselves. Over time, the gatekeepers to civic knowledge became the media, and it was the press, TV networks and, eventually, online news outlets that determined and reported the facts as they saw them, which provided the basis for voting decisions and civic engagement. “[Those] models for informing the public operated under a couple different constraints,” says Renée DiResta, technical research manager of the Stanford Internet Observatory, a program at the Freeman Spogli Institute for International Studies that researches the abuse of information technology. Editorial gatekeepers controlled how stories were told, and distributors—like broadcast networks—controlled their reach.

These days voting rights are broader, literacy rates are higher, and all the information in the world is at our fingertips. America’s 240 million eligible voters are dazzled with choices—for the school board, the water
district, county judge, a half-cent sales tax funding firefighters, perhaps propositions about development restrictions or education bonds, not to mention various high-profile offices. Amid this dizzying complexity has come a profound shift in role of the gatekeeper—or perhaps the death of it—with the rise of social media platforms. “In the social media era, content creation and content distribution are no longer limited by the older gatekeeping models. Everyone participates in the creation and distribution of content, and online crowds determine what we see,” DiResta says. “Anyone can amass influence.”

As people privilege the sources they find “relatable” over voices with knowledge and expertise, she says, they’ll rely on the statements of a neighbor, movie star or charismatic stranger over those of authorities. And so the American electorate has become increasingly swayed by influencers—including celebrities like Kim Kardashian and Taylor Swift, both of whom have endorsed candidates for state or federal elections. In one 48-hour time period in October 2018, Swift (who had 112 million Instagram followers at the time) inspired 169,000 people to register to vote through Vote.org, according to the organization. About 190,000 voters registered via Vote.org in the entire month of September 2018.

‘Mere exposure to good information that’s been vetted can backfire—people don’t trust it. But we find that with civil discussion with diverse others, people transform.’

It’s not that inspiring people to vote is bad. But when it comes to endorsing candidates or policy positions, things get a little trickier. “In the era of gatekept content, experts and government and the media determined whose views were legit,” DiResta says. Now, “expertise and influence have been decoupled.”

What’s worrisome there, DiResta thinks, is that people are more exposed to manipulation by bad actors. Her research has shown, for example, the ways that Russian military propaganda becomes amplified across the media ecosystem—readers repeat and retweet the narratives that appeal to them without knowing their original source or reliability. And at the same time, she says, algorithms are pushing ever-more homogenous information into people’s social media feeds. That kind of platform design exacerbates partisan animus, a phenomenon explained by political scientists including Stanford’s Shanto Iyengar and Neil Malhotra, MA ’05, PhD ’08, in a 2019 paper about affective polarization—a sense of division not explained by policy positions or political ideology. “Democrats and Republicans both say that the other party’s members are hypocritical, selfish and closed-minded, and they are unwilling to socialize across party lines, or even to partner with opponents in a variety of other activities,” the authors write.

Such conditions make it difficult to answer all the open questions with which we have to grapple as a society. But there’s something worse, Ober will tell you, than failing to find the best solutions. When the people in a representative democracy can’t see eye to eye, they can’t effectively serve as a check on their government. That’s when the biggest enemies to representative democracy—systemic political corruption and autocracy—crop up. And both of those are live issues. Some 63 percent of American participants in the 2017 World Values Survey said they think rich people buy elections fairly often or very often, and 32 percent said they believe voters aren’t often offered a genuine choice in elections. In a democratic system, it’s the voters who have to remedy those problems at the ballot box.

“[A]ny modern democracy that cannot resort to decision-making by the people if and when necessary remains vulnerable to capture by a political elite,” Ober writes in Demopolis. He also makes the point that such dangers aren’t purely theoretical. “Tyranny has often been a successor to democracy, in antiquity and modernity.”

The Big Fix

To improve civic literacy and the quality of public decision-making, Fishkin and Alice Siu, ’03, MA ’07, PhD ’08, director and associate director, respectively, of Stanford’s Center for Deliberative Democracy, are trying to give voters more voice and agency with what they view to be the “magic sauce” of democracy: deliberation.

“We have a highly polarized environment,” Fishkin says, “and a lot of people are getting their information from social media. Mere exposure to good information that’s been vetted can backfire—people don’t trust it. But we find that with civil discussion with diverse others, people transform.”

A key strategy toward that end is reinventing public opinion research via the deliberative poll. Organizers of a deliberative poll gather a representative sample of voters, meant to be a microcosm of the population at large, to consider policy proposals. The sample includes “standard categories” of representation, Fishkin says, such as class, gender, education, income and ethnicity, as well as “attitudinal representativeness,” meaning the mindsets of the participants reflect those of the broader population. The sample size must also be large enough that changes in opinion will be statistically significant. The poll organizers—Center for Deliberative Democracy staff members typically team up with civil society organizations, governments, media companies, universities or nonpartisan groups—survey the participants on a targeted set of issues. Next, the voters gather together for a period of time (say, a weekend) to discuss those issues. They mull over briefing materials vetted by an ideologically diverse advisory group, and they have small-group, moderated discussions with ground rules. There’s also a plenary session in which participants pose questions directly to relevant experts. Afterward, participants are polled again. The new opinions represent the views the broader public might hold if people became informed enough to make considered judgments. The results are then shared with the media and community leaders for use in reporting and decision-making.

In 2019, the Center spearheaded America in One Room, a deliberative poll that brought 526 people—a representative sample of American voters—to a hotel outside Dallas for a long weekend. Relying on their personal views and experiences as well as briefing materials created by a committee of experts, the attendees broke into small groups and debated the major topics of the 2020 election: the economy, the environment, foreign policy, health care and immigration. They considered and discussed policy proposals.

At the end of the weekend, many participants interviewed by the New York Times said they didn’t believe they’d changed their mind on anything. But, they said, they did have more empathy for folks who felt differently, and they had gotten practice debating in a civil manner.

In actuality, some of them had changed their minds. Participants moved toward the center on several issues, researchers from NORC at the University of Chicago who conducted before-and-after surveys found. For example, Democrats lessened their support for a $15 minimum wage, from 83 percent to 59 percent, while Republicans increased their support for rejoining the Paris climate agreement, from 28 percent to 44 percent.

‘The public is very smart if you can tap into that wisdom. People only appear stupid because they aren’t putting a lot of cognitive effort into having an opinion.’

“What I hear during exit interviews from deliberative polls is, ‘Well, I’ll try to listen to the other side,’ ” Siu says. “People say, ‘I can’t replicate deliberative polling, but I can talk to someone who thinks differently.’ ”

The basic concept behind deliberation, Fishkin writes in Democracy When the People Are Thinking, is weighing. “People should weigh the arguments, the competing reasons, offered by their fellow citizens under good conditions for expressing and listening to them and considering them on the merits.” From a psychological perspective, it’s that mental wrestling that helps us overcome our impulse to seek out facts that support what we already think. And when we’re with others in person, we have the opportunity to express empathy.

“The public is very smart if you can tap into that wisdom,” Fishkin says. “People only appear stupid because they aren’t putting a lot of cognitive effort into having an opinion.”

Deliberative polling, though, is tailored and time-consuming—a huge amount of work goes into preparing the briefing materials alone. Siu says that’s done under the supervision of large, multifaceted advisory panels comprising a full spectrum of viewpoints, from policy advocates to academics to representatives of private industry. The panelists all contribute to the process of identifying the policy proposals, selecting the background reading, and editing and reviewing the final briefing packets and videos. “The document, how it’s composed, is quite arduous,” Siu says. They start with a long list of proposals and whittle it down to what can be discussed in a weekend’s time—and to ideas that are actually feasible to implement. “No pie-in-the-sky ideas,” Siu adds. Then you have to select the participants and produce the event itself. Even if deliberative polling is the solution to democracy’s woes, it’s going to have to become less bespoke.

Turns out, scaling it up is in the works. Four years ago, Fishkin and Siu teamed up with Ashish Goel, MS ’98, PhD ’99, a professor of management science and engineering, to create the Stanford Online Deliberation Platform, an automated moderator for group deliberations that’s programmed to give set speaking times to participants and to prompt questions and debate. It requests arguments for and against proposals, prods the people who aren’t speaking, and gives participants the opportunity to weigh in on whether all sides of the issue have been fully explored. It solicits questions from the plenary, and everyone votes on which ones will be discussed.

Just in the past year, the platform has been used twice at Stanford to help university administrators consider the design of the new school on climate and sustainability. The first poll engaged 203 faculty over a weekend in January 2021, and the second involved 184 undergraduate and graduate students in May.

More broadly, the online platform was used for America in One Room: Climate and Energy, a 1,000-person mass poll on climate and energy policy that the nonpartisan organizers described as the “largest controlled experiment with in-depth deliberation ever held in the United States.” The event, held virtually in September, included intensive, multiday discussions among more than 100 small break-out groups. More than 30 expert panelists were on hand to advise participants and ranged from the head of government relations at Shell Oil Co. to the director of policy and external affairs at the Nature Conservancy. “On 66 of the 72 issue propositions in the survey, participants changed significantly over the course of the deliberation toward wanting to do more to combat climate change,” researchers reported.

In a November event funded by the American Academy of Arts and Sciences, a national deliberative poll of people aged 18 to 29 used the platform to consider policies related to social media and democracy, civic responsibility and representation. Participants dove into quandaries— Should social media companies be regulated like the news media?—and weighed policy proposals, such as making federal election day a holiday, instituting compulsory public service and using ranked-choice voting. The final report was coauthored by 15 Stanford students taking Communication 138/238, a deliberative democracy practicum co-taught by Siu and Haas Center for Public Service executive director Tom Schnaubelt. They concluded that one of the benefits of participating in the poll was that it enhanced people’s civic knowledge.

Benefiting the public is central to the Center for Deliberative Democracy’s plans going forward. Whereas deliberative polling relies on representative sampling to produce an accurate microcosm of considered views, a new concept—what Fishkin calls deliberative scaling—ratchets up the number of participants beyond what’s needed to create a representative sample with the aim of jumpstarting civic engagement more broadly. In principle, Fishkin says, there’s no limit to the number of participants in a mass event.

“Because the platform works so well, we’re intending to scale the deliberation to much larger numbers in partnerships with universities and civic groups around the country,” Fishkin says. “When people engage in this process, they are more motivated to participate in public affairs. They become more civically engaged. We activate them as citizens, and we think that’s a useful contribution.”

Jill Patton, ’03, MA ’04, is the senior editor at Stanford. Email her at jillpatton@stanford.edu.

Civil Disagreement

By Evan Peng

“ ‘The Declaration [of Independence] is just an ordinary memo,’ ” says political science lecturer Brian Coyne, PhD ’14, quoting from Our Declaration by Harvard political scientist Danielle Allen. “What do you think she means by that?” he probes, opening up discussion to his section of COLLEGE 102: Citizenship in the 21st Century.

Citizenship is defined as membership and participation in a large, self-governing group. Or, at least, that’s the working definition that Coyne and his students are using this quarter.

But what does it mean to be a good citizen—of communities as small as a 10-person club and of the global community as a whole? This is the central question that students ponder in COLLEGE 102, the winter-quarter component of Stanford’s new first-year core curriculum, Civic, Liberal and Global Education, or COLLEGE.

The idea of citizenship has been emphasized since Stanford’s inception; the university’s Founding Grant states that its purpose is “to promote the public welfare by exercising an influence in behalf of humanity and civilization . . . and inculcating love and reverence for the great principles of government as derived from the inalienable rights of man to life, liberty and the pursuit of happiness.” Words inspired, of course, by the Declaration of Independence.

That said, “I think many of us, both among the students, among the faculty, but also among our alumni, have felt that in recent years, not only has this purpose of the university slipped, but also it has become more pressing than ever,” says Dan Edelstein, professor of French and director of Stanford Introductory Studies. COLLEGE was developed in part in response to this urgency and as a recommitment to Stanford’s founding values.

The course draws its instructors from across the university’s schools; faculty backgrounds range from religious studies to computer science and everything in between. “I really believe in the mission of helping our students prepare for lives where they will encounter disagreements on fundamental questions, to give them some tools and inspire them to commit themselves to work through these disagreements and to participate,” says philosophy professor Debra Satz, dean of the School of Humanities and Sciences, on her motivations for being an instructor in COLLEGE 102.

This model of emphasizing constructive disagreement is why class sections are kept at around 12 students each, with classes largely based on discussion. “We didn’t want to herd 1,700 people into some giant auditorium and have them listen to huge lectures,” says Coyne, “because part of the goal is to get them to work together.”

As for Allen’s assertion, for the next 80 minutes, that “ordinary memo,” viewed as a rough draft rather than a monument set in stone, would provide the basis for discussions ranging from whether other documents of citizenship qualified as memos to how to grapple with ideas of citizenship inherited from unjust times. The Declaration of Independence itself? It is a memo in that it establishes the terms of American citizenship, the students said—an understanding they arrived at together.

Evan Peng, ’22, is an editorial intern at Stanford. Email him at stanford.magazine@stanford.edu.

Photos of students participating in Democracy Day on Stanford campus
Photos: David Gonzales, ’93

Time Out

By Valerie Trapp

Sean Casey got an emaiL saying the palm tree in the Democracy Day graphic was too happy. “You could take it in a direction that is much more negative, and rightly so,” says Casey, ’23, chair of the Democracy Day Coordinating Committee. But the proposal he and Jonathan Lipman, ’21, put before Stanford’s Faculty Senate last summer called for a noninstructional day of special programming to be held campuswide on Election Day each year. It would signal the university’s commitment to democratic principles and encourage student voting, civic engagement, dialogue and public service. Its symbol, according to Casey, needed to be optimistic.

Illustration of a palm tree running
Illustration: Jennifer Lee, ’22

In its inaugural year, Democracy Day fell on November 2, 2021, a cloudy Tuesday. On Wilbur Field was Donuts & Democracy, a deliberative polling event centered on discussions about social media and technology. Armed with input from experts, participants debated the role of social media in society, along with proposals to create a publicly funded or nonprofit social media platform, in small groups moderated by undergraduates who were enrolled in a deliberative democracy practicum. Later in the day, programming included an interdisciplinary panel of faculty, “Is Democracy in Danger?” on Meyer Green; a virtual session with Sen. Jeff Merkley, ’79, who emphasized the importance of active civic engagement and advised students on how to shape the nation’s future; and an outdoor screening of the documentary Boys State, which depicts a thousand teenage boys forming a mock representative government. “I came here because of my lack of information,” said first-year student Francisco Ortiz. “I was introduced to topics that I never even considered were big problems today.”

Virginia Bock, the Haas Center for Public Service communications manager and the day’s staff leader, hopes future Democracy Days will encompass more hands-on student activities—from voting to community service projects to events designed by individual student groups. When Casey steps back to think, he is quick to invoke perspective (“I’d be surprised if our little day is changing the hearts and minds of legislators in Washington”). Still, he is already working with other universities to implement Election Day commemorations on their campuses. “I’m optimistic that progress is near,” he says. That would make a certain cartoon palm tree very happy indeed.

Valerie Trapp, ’22, is former editorial intern at STANFORD. Email her at stanford.magazine@stanford.edu.

Illustration of people holding up letters that spell out civics

Isn’t This Stuff Taught in School?

By Rebecca Beyer

When it comes to K-12 civic education in the United States, by almost any measure, the system is failing. A 2018 Education Week survey found that only eight states require a yearlong civics course in high school—and 15 states don’t require one at all.

It’s not because students are gaining that knowledge earlier. On the most recent national assessment, only 24 percent of eighth graders were proficient in the subject. One obvious problem: Half of the students tested hadn’t taken a civics-focused class.

Civics is the study of how our government was formed, how it functions, and what roles individuals play in that process—and currently, its teaching is a hodgepodge. With federal legislation that would authorize billions in grants for civics education stalled, nonprofit organizations and academic researchers are trying to fill the gap. And their efforts are gaining traction.

“On my good days, I’m very optimistic,” says Louise Dubé, executive director of iCivics, a nonprofit civic education provider founded in 2009 by former U.S. Supreme Court Justice Sandra Day O’Connor, ’50, JD ’52. “This is a unifying idea for all Americans to get behind.”

David Davenport, ’72, a research fellow at the Hoover Institution, agrees. In 2020, Davenport authored a report for the Orrin G. Hatch Foundation called “Commonsense Solutions to Our Civics Crisis.” In addition to more funding, testing and teacher training, Davenport recommends the so-called layer cake approach to civics, which begins introducing children to age-appropriate ideas in elementary school. “If you wait for a single, one-semester course in high school, kids don’t have any context,” Davenport says. “They’ll show up at class with nothing.”

One resource for middle and high school teachers is iCivics, which aims to cultivate an appreciation for civic engagement among young people. The organization provides hundreds of free curricular resources—including 14 nonpartisan educational video games, such as Argument Wars (in which players argue real Supreme Court cases), Counties Work (which asks players to manage a county and get reelected) and Do I Have a Right? (which tests players’ knowledge of the Constitution). According to assessments that iCivics embedded in two of its election-related games in 2021, students’ knowledge of civic content improved by 26 percent after playing the games. Perhaps as notable: There was a 38 percent jump in student interest in learning about topics like the Electoral College or participating in voting.

Anyone can access iCivics content on their own at any time—and more than 145,000 teachers and 9 million students in all 50 states do each year. But iCivics has also worked through the Educating for American Democracy initiative to design a roadmap for effective civics education, a project funded by the National Endowment for the Humanities and the U.S. Department of Education. “The American democratic system is not an intuitive system—it needs to be taught,” Dubé says. “Our goal is to rebuild a healthy American democracy and to reimagine civic education to do that.”

For Stanford Graduate School of Education professor Sam Wineburg, PhD ’89, that work begins where American democracy has faced the greatest challenge in recent years: the internet.

In 2014, the Stanford History Education Group that Wineburg leads started the Civic Online Reasoning (COR) program, which provides free lessons designed to help students evaluate information they find online. “We are in an incredibly polarized time, and what’s feeding that is the spread of misinformation,” he says. “If we want to be informed citizens, the way we do that in the 21st century is we go online. We don’t go to the public library to learn about the efficacy of a soda tax or whether we should ban private prisons; we google it.”

In fact, Google is one of COR’s key partners. Last year, the search engine—based in part on research by Wineburg and Michael Caulfield, a research scientist at the University of Washington’s Center for an Informed Public—began including an icon designed to help users in the United States assess the credibility of search results. The feature, which appears as a three-dot menu next to your search results, encourages COR skills such as lateral reading (checking what reputable websites say about a source) and click restraint (intentionally skipping over the first search results, which are often advertisements). Wineburg says the feature is a “small nudge to see if we can make things a little better.”

The primary focus of COR, like iCivics, is to make materials teachers can use in the classroom. Its curriculum—based on the work of professional fact-checkers—includes nearly 30 lesson plans that cover using Wikipedia, evaluating claims on social media and identifying trustworthy evidence, among other topics. “You can preach to teachers until you’re blue in the face,” Wineburg says. “We need materials. We need concrete things. That’s where my efforts are.”

Rebecca Beyer is a Boston-based journalist. Email her at stanford.magazine@stanford.edu.

© Stanford University. Stanford, California 94305.