This article is brought to you by American Purpose, the magazine and community founded by Francis Fukuyama in 2020, which is proudly part of the Persuasion family.
Democracy is under siege, both in the great global conflict of ideas and in the lived experience of people around the world. We are grappling with a democratic recession—a pull-back in the number of countries that fulfill accepted standards for the practices of party competition-based democracy. And there is the disquieting rise of competitive authoritarianism: regimes that keep many of the formal trappings of democracy but actually practice grossly uncompetitive or rigged elections.
Meanwhile, many countries are seeing the rise of populism, with parties that stir up popular anger and raise questions about whether, if they gain power, they will ever give it up. Hence the basic point of party competition-based democracy is at risk: that “the competitive struggle for the people’s vote,” as Joseph Schumpeter famously put it, should determine who is in power.
At a fundamental level, there are four basic challenges. First, assuming the point of democracy is to make a connection between “the will of the people” and what is actually done, it is very difficult even to assess the public will in our current media and social media environment. So many entities—some anonymous—are making great efforts to propagandize and distort public opinion for policy and electoral purposes, and there are so many efforts to mislead the public with misinformation and disinformation that it becomes very difficult to assess what they genuinely want. Many of us are in our own media enclaves, consuming viewpoints we find congenial with no idea what the arguments might be on the other side of the partisan divide.
Second, this flawed system of political communication fosters extreme partisan polarization. The two parties seem intractable and intense in their divisions. As one recent overview put it, our differences are not just polarized, they are “calcified,” and the result is deadlock. Deadlock leads to a perception that democracies cannot get anything done, further eroding legitimacy.
Third, when it comes to elections, most voting is based on the tribalism of party loyalties. Voters rarely depart from their parties and vote with their convictions on the issues (if they have them). They vote for their team. This makes elections less a contest of ideas and more a contest of mobilization, money and media gamesmanship. Winning an election is less a mandate than it is like the triumph of a sporting event.
Fourth, even if there were some solutions to these three problems, it would seem daunting or impossible to scale them. How could we possibly change the media habits and political norms of an entire society?
The problems confronting democracy are so profound, our disagreements so intense, that it might seem naïve to think deliberation could be the answer. Nevertheless, my answer is that, provided we can muster enough political will, there is promising evidence that the talking cure can work. It would need to be applied systematically, and in some novel ways, but it could eventually transform our republic.
Since 1994, I have led a research program that fosters deliberation with random samples of the public in projects around the world (now with 160 cases). The fundamental idea of deliberation is that people weigh competing reasons for and against policy proposals in a civil and evidence-based environment. While many people do this on occasion in natural settings, the effects can be heightened with an organized design.
We implement the idea with a system called Deliberative Polling, which can be conducted either face-to-face or online. The face-to-face versions, at least on a national basis, require transporting a good sample of people from all over the country to a single place for a weekend of discussion. The online versions have been made more cost-effective and practical through an AI-facilitated moderator for small group, video-based discussions developed with my colleagues at Stanford. It produces results very comparable to face-to-face projects with human moderators. Moderated discussion with diverse others has surprising and lasting effects. These effects, documented in peer-reviewed journals, are the basis for my argument.
In 2019, the first of several “America in One Room” projects—national experiments conducted with Helena, a global problem-solving organization, and with samples recruited by NORC, a nonpartisan research organization at the University of Chicago—demonstrated both dramatic depolarization across party differences and lasting effects on more deliberative voting. For example, before deliberation, 79% of the sample Republicans supported “forcing undocumented immigrants to return to their home countries before applying to live and work in the United States.” After deliberation this support dropped to 40%.
Similarly, before deliberation, only 36% of the Republicans supported Deferred Action for Childhood Arrivals (DACA), “which protects from deportation children brought to the United States by parents who entered illegally.” After deliberation this support rose to 61%. There were similar changes among the Republicans on other immigration questions (e.g. support for refugees). In all these cases, the changes were depolarizing in that the mean positions of the two parties moved closer together and, in particular, those taking the most extreme positions did too.
In the same project, Democrats also depolarized in comparison to Republicans. The movements were most striking on the most ambitious and expensive social programs. For example, Democrats initially supported “a government-funded baby bond for use in education or other purposes.” This bond, awarded at birth, would help with college or other needs when the child reaches adulthood (when we ran this question, it was a prominent proposal from Senator Cory Booker). But with deliberation, support dropped 41 points from 62% to 21%, moving much closer to the Republican position. Similarly, Democrats initially supported a version of “Medicare for All,” but that support also dropped substantially with deliberation from 70% to 56%, much closer to the Republican position. Overall, of 26 proposals that were highly polarized at the start, all 26 showed significant depolarization following deliberation, especially among those who started with the most extreme positions on either side of the partisan divide. These patterns were later replicated with similar projects on climate change and democratic reform. Deliberation depolarizes our most extreme differences.
But does it have any lasting effects? One indicator of lasting effects is how the deliberators vote up to a year later. We found that those who deliberated, as contrasted with a control group who did not deliberate, voted overwhelmingly according to their post-deliberation policy preferences in the 2020 presidential election. We replicated this effect with the 2021 national project America in One Room: Climate and Energy, which affected the voting by deliberators in the mid-term elections a year later.
Deliberation appears to foster increased civic engagement in which citizens acquire a greater sense of political efficacy (they believe they have opinions worth listening to), continue to become more informed in the following months (as measured by knowledge questions), and vote in a way that implements their judgements on the issues.
This effect was particularly striking because the climate deliberators—a national sample of nearly a thousand—discussed the issues with our AI-assisted online video platform in about 100 small groups of 10. The online platform moderates the discussion and in principle could be used to reach much larger populations cost-effectively. In any case, these projects support the solution to the third problem: how can we foster more deliberative voting?
If deliberation has these effects, how can it be scaled? How can we create a more deliberative society? The answer, in my view, goes back to ancient Athens, which employed random samples deliberating on important public decisions, before, during and after the debates in the Assembly. Aristotle famously described the rotation method whereby citizens would each “rule and be ruled in turn.” This picture has usually been dismissed as an historical curiosity applicable only to very small states.
But there is now a “deliberative wave” around the world of random samples discussing complex and often contested policy questions, often on behalf of governments. Our projects have addressed many questions in different contexts. They have brought wind power to Texas, helped Japan and South Korea make decisions about nuclear power, helped desegregate the Roma-only schools in Bulgaria, helped formulate a ballot proposition in California, and assisted a major technology company in consulting its customers about the metaverse and about the future of AI. It has spread in schools as a form of civic education in the United States and become part of the process of constitutional change in Mongolia, producing two successful amendments—most recently an amendment adding proportional representation to the electoral system.
This is but a partial list of applications that have actually been tried. There are many other possibilities. Imagine if deliberation were employed with a national sample as a start to the U.S. primary season, representing the whole country under conditions where citizens could grapple with the issues, and giving birth to momentum in the subsequent primaries. Or imagine a national “Deliberation Day” in which all voters had an opportunity to deliberate before a national election. If deliberation were widely deployed to deal with difficult and contentious issues, the opportunities for each citizen to deliberate would multiply to the point that the rotation method would work at the large scale. Most of these applications could be done cost-effectively on our platform or other platforms with similar technology for deliberative discussion. As we have seen, intensive deliberation leaves its mark on the habits of citizenship and voting. These applications are each justifiable. But together they could add up to something more—a more deliberative society in which public opinion would be more meaningful, our most extreme divisions would moderate, and our democracy would be peopled by more deliberative voters. It is not an impossible fairy tale but a matter of collective political will. Over the long term we can energize a more substantive form of democracy.
James Fishkin is the Janet M. Peck Chair in International Communication, Senior Fellow at the Freeman Spogli Institute for International Studies and Director of the Deliberative Democracy Lab at Stanford University. He is the author of Can Deliberation Cure the Ills of Democracy?
Follow Persuasion on X, Instagram, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.
And, to receive pieces like this in your inbox and support our work, subscribe below:
 | A guest post by
|
I’d suggest that Mr Fishkin’s point is in agreement with the challenge posed to us by our Founders. Flawed as both they and the governmental blueprint they bequeathed to us were, at the core of it was the chance to see if ‘We the People’, for all our differences in outlook could together find just enough the courage, the honesty, the compassion, the tolerance, the understanding, the humility, the humor, the wisdom, the hope, and the sheer common sense to rule outselcves from the bottom up with as much justice and equity as is humanly possible. At any one point, the levels of those traits in our political discourse is the measure of our success or lack thereof.
We are both the inheritors of and the participants in the most extraordinary, the most crucial, the riskiest, and the most complex ongoing experiment in human society and government ever attempted. IM(not so humble)O, our basic problem is that far too many of us simply don’t understand either what that means or the responsibility it places on American citizenship.