American democracy is badly broken, but there’s no shortage of ways to improve it. One dramatic example will be on display in Dallas from Sept. 19 to 22, when a project called "America in One Room” will bring together a representative sample of more than 500 registered voters to engage in a process known as “deliberative polling” focused on five big issues: health care, immigration, the environment, the economy and foreign policy.
Participants will be polled initially, then presented with a common set of briefing papers (or video presentations) and then “randomly divided into smaller discussion groups, each led by a moderator trained to encourage fair and civil conversation,” as described in an recent New York Times op-ed by James Fishkin and Larry Diamond. They will then return to a plenary discussion after which they'll be polled again to get the results of the shared deliberative process. They will also interact with presidential candidates.
It’s an idea that Fishkin, director of the Center for Deliberative Democracy at Stanford University, first unveiled in the Atlantic in 1988, and it’s been used more than 100 times in different venues around the world. “Every place it has been applied … it has had constructive effects in clarifying the public’s considered judgments,” Fishkin and Diamond wrote in the Times. “Why not employ this at the start of our presidential race, when the public is clearly hungering for some way of understanding the vexing issues we face?"
Speaking to Salon, Fishkin explained that the process they devised is “meant to answer this type of hypothetical: What would the public think under good conditions? Or under the best practical conditions that we can provide on the specific issue?"
There are at least two other ideas that could build on this, and help make the 2020 race more reflective of what the American people really want — as opposed to what political operatives or media pundits think they want. Perhaps now that our democracy is imperiled like never before we’ll be willing to give these ideas the attention they deserve.
The general idea of representative mini-publics dates back at least to Athenian democracy, and has taken on different forms over time. But Fishkin’s deliberative polling model has unique strengths, as noted in a 2005 Hewlett Foundation report, "Mapping Public Deliberation," which covered everything from New England town meetings to closed stakeholder processes:
Because its organizers strive for the rigor and reproducibility of experimental social science, the deliberative poll is the most methodologically consistent and systematized of the venues discussed here. Given its pre-, post-, and control group polling methodology, deliberative polling provides some of the most robust evidence that citizens learn through deliberation and that this learning changes their policy stances.
The report observed, however, that “there is little evidence that officials take the results of deliberative polls into account when shaping policy decisions.” This isn’t strictly true, as we’ll see shortly. But it is a challenge, and a reason to look for complementary ideas that can help amplify and build on the results of deliberative polling.
One of those other ideas is that using a “citizen's agenda” as the guiding framework for campaign coverage. This is a model pioneered by the Charlotte Observer in 1992, and championed repeatedly since by NYU journalism professor Jay Rosen. If conventional “who’s winning?” coverage approaches politics like a sports beat — a horse race — the citizen’s agenda approach is more like a restaurant critic, helping people to find what they’re looking for.
Rosen’s 1999 book, “What Are Journalists For?” began with a discussion of the 1996 National Issues Convention that helped establish the model, and Rosen thanked Fishkin in his acknowledgments, saying, “he gave me much to think about with his own work on deliberative democracy." So the connection here is nothing new. But it’s nuanced, as we’ll see below: Their purposes are complimentary, not identical.
The second related idea is “public interest polling,” pioneered by Alan Kay, which I’ve written about before, most extensively in 2015 here. It has a superficially similar look, in that it presents balanced information vetted by experts, and uses a sequential process to get refined results. But there’s no interpersonal deliberation, and the aim is primarily to discover a latent consensus that may already exist under the surface. It’s no substitute for deliberative polling, but it can potentially complement it — and any pollster who wants to do so can adopt its methods. It can also help journalists pursue a citizen’s agenda.
The 1996 National Issues Convention came after a similar process in Britain, and introduced the idea to the U.S., according to Fishkin. “It was meant to be a deliberative evaluation of the candidates and the issues,” he told me. “But the advisory group was worried that the candidates might not come if we evaluated the candidates, so we ended up just doing the issues. That was disappointing. I told everybody that if they didn't have the punch line about the candidates, it would not have any direct political effect. And that's what happened.”
This project attracted a lot of attention. “I have a stack of about 600 newspaper articles," Fishkin said. The first significant result came just two years later, via a series of eight deliberative polls in Texas that played a key role in transforming energy policy toward wind. The Texas Public Utilities Commission had a requirement to consult with the public, but had no idea how to meaningfully do so — until Fishkin’s idea came along.
So the commission did eight polls, covering all eight service area territories in the state. The average willingness to pay “at least a dollar more per month on their utility bills went from 50% to 84%, approximately,” Fishkin said. “In other words, a massive change of opinion.” In fact, the utilities only ended up charging about 25 cents a month, but the end result “moved Texas from last to first in the amount of wind power," Fishkin said. "Texas surpassed California in 2007, and when we started it was number 50.”
Deliberative polling has “now spread to 28 countries, and 108 cases,” Fishkin said, including six national deliberative polls in Denmark, with a seventh just commissioned. South Korea had two nuclear reactors half-constructed when President Moon Jae-in came into office in 2017. Rather than make a decision on his own, Moon called for a process "on the model of a national deliberative poll,” Fishkin explained, Although the South Korean public is generally opposed to nuclear power, they actually “decided to continue the construction," Fishkin said, because "they didn't want to import fossil fuels to make up the difference.”
That kind of decision — on an issue where the public may be ambivalent — is an especially attractive one for deliberative polling, since it can facilitate people owning that ambivalence rather than stifling it and projecting anger and frustration at political elites. A more serious commitment comes with making deliberative polling a normal part of the political process — a step taken by Mongolia in 2017, when it passed a law integrating deliberative polling into the process of amending its constitution, arguably the most important law-making function there is. The law was passed after an initial, more modest experience.
Mongolia’s first encounter with a deliberative poll in December 2015 had a fairly narrow scope on a very concrete issue: prioritizing 14 infrastructure projects in the capital city, Ulaanbaatar. As described in the press release following the deliberation, “Eight of the fourteen projects changed significantly. Three went up, five went down. After all the discussions, the people clearly had three top concerns. First was an environmental focus …. A second major concern was to realize the vision in Master Plan to decentralize industry and employment and move them from the city center …. Thirdly, they were concerned about debt and taxes.”
Note that these broad thematic concerns didn’t have to arise simply to answer the question of prioritizing spending. They reflected value added by the participants themselves, not individually but in the collective face-to-face deliberative process. The press release went on to say:
The participants greatly increased their sense of efficacy and trust. They increased their agreement that “public officials care about a lot what people like me think” and they increased their “trust that [municipal] government would do what is right.” The participants also evaluated every component of the process highly. 95 percent or more thought each part of the process was “valuable.”
The positive nature of this experience was just one factor that led to the 2017 law. Fishkin co-authored a follow-up report, noting that support for two of the most ambitious proposals "dropped dramatically with deliberation.” Support for adding a second chamber to parliament dropped from 61 to 30 percent, and support for switching to indirect election of the president for a single six-year term dropped from 61.5 to 41 percent. Each of these had been supported by one of Mongolia’s two major parties, but both were dropped as a result of the deliberative process.
But the results weren't all negative by any means. “Nine of the top ten proposals at the end of the deliberation concerned the transparency, accountability and meritocratic operation of government through protections for the civil service and the judiciary (the tenth was granting authority to the prime minister to appoint and fire members of the cabinet).”
Mongolia’s willingness to embrace deliberative polling may reflect how young its democracy is. Unlike Americans, citizens of a country that was ruled by one party from 1921 to 1996 don’t take their democracy for granted. Perhaps the realization that we can’t take democracy for granted either will help us to take it more seriously now.
Perhaps the most ambitious American effort to date took place in California in 2011 (description and PBS documentary here). Deliberation focused on four areas — the statewide initiative process, legislative representation, local government, and tax and fiscal policy. It resulted in state legislation leading to a ballot initiative — which, however, was defeated the following year.
A postmortem analysis Fishkin co-authored noted that “the proposals actually deliberated on by the people might well have passed if not encumbered by additional elements not deliberated on by the public that drew opposition.” All the proposals were also combined into a single initiative, rather than separated into individual initiatives that might have facilitated a more coherent public debate.
This example highlights the importance of the broader political context. No matter how good a deliberative polling process may be, its ultimate success depends on outside factors as well. It’s a powerful tool to help democracy work better, but it can’t do everything by itself. It needs broader democratic support. In that spirit, it can benefit from the two ideas mentioned above: Journalists adopting a "citizen's agenda" approach to campaign coverage, and pollsters adopting a public-interest polling approach.
The citizen's agenda
“When I first encountered the concept, a deliberative opinion poll was about improving the quality of public opinion,” Rosen told Salon. “The citizen's agenda model for election coverage is different. It might employ opinion polling and public deliberation as elements in a broader approach.”
More specifically, he said, the "animating unknown" of the citizen's agenda approach is not, "'Who's likely to win and why?’ which is an analyst's question, but ‘What do you want the candidates to be talking about as they compete for votes?’ That is something you can ask of voters, the people you are trying to inform." Rosen continued:
There is no one way of getting a reliable answer. You need multiple methods. Polling can be one. Others would be one-on-one interviews, "call this number and leave a message," "fill out this form," focus groups, tweet your reply, talk to us on Facebook, come to our listening event, or we'll come to your listening event.
That is where I can see the deliberative opinion poll making a difference. At a public event with voters, it would be instructive for journalists to hear the answers to "What do you want the candidates to be talking about as they compete for votes?" It would be even more instructive to know the answers after people deliberate with each other on that question.
We can also take it one step further: Present voters with results from a deliberative poll like “America in One Room,” as a starting point, along with the question of what they want candidates to talk about. In other words, use the initial intensive deliberative process as a foundation to build further conversation.
Public interest polling
This concept, as developed by Alan Kay, can also build on and synergize with deliberative polling. I’ll just highlight excerpts of just two examples I described in 2015, drawing from Kay's work with Americans Talk Issues (ATI) and reported in his book, “Locating Consensus for Democracy - A Ten-Year U.S. Experiment,” as well some online material.
The first concerns energy policy that's good for the economy and the environment — reflecting public opinion in 1991. Out of 18 proposals tested, using different wording in two different surveys, Kay found five “triple winners” (getting  the energy we need, while helping  the economy and  the environment) getting majority support (except for one, which received “only” 49% in one survey). Two examples will suffice:
- Install new renewable-energy electric generating systems based on wind, solar and hydro or water power. (76 percent, 71 percent)
- Adopt, on a national scale, some new fuel — like hydrogen or alcohol fuels — to begin to replace gasoline. (59 percent, 60 percent)
As I wrote then, “That was ATI's poll result in 1991! Imagine how much farther along we'd be today in fighting global warming, if the public voice, reflected in ATI's results, had been heeded by politicians at the time.”
At the very least, the results of the environmental deliberations from America in One Room could be used as the basis for public interest polling, to see to what extent public opinion already reflects those deliberative results, and thus to further promote their discussion — both by journalists taking a "citizen's agenda" approach, and by politicians running for office, from president on down.
The second example concerns government reform — a topic integral to Ross Perot’s independent candidacy in 1992. Kay examined support for 38 possible policy responses. The first group included relatively well-known, punitive measures, such as reducing the pay and staff of Congress, introducing term limits and passing a balanced budget amendment. The second more proactive group of reform proposals included five with 70 percent support or more, including the “motor voter” plan and these two:
- In the same way we've developed and use the Gross National Product to measure the growth of the economy, develop and use a scorecard of new indicators for holding politicians responsible for progress toward other national goals, like improving education, extending health care, preserving the environment and making the military meet today's needs.
- Require Congress to conduct scientific, nonpartisan, large-sample surveys of public opinion on all important national issues and to promptly release the results to the media so Congress and the public will know what most Americans want for legislation.
As I wrote in 2015, “Unfortunately, only the motor-voter plan was adopted by Congress. All the other measures that could have been adopted to produce a more vibrant, responsive democracy and build public trust were ignored.” What’s changed since then is that we now have a much more aroused progressive community that is strongly invested in citizen empowerment — on everything from the climate crisis to gun safety to redistricting reform. The time is ripe for public interest polling on a wide range of proactive government reforms, up to and including ideas for making deliberative polling a normal part of our political process.
As I said in the beginning, American democracy is badly broken, but there’s no shortage of possible ways to improve it. America in One Room will provide a much-needed window on our possible future. But it’s only one facet of what we need to do in order to move forward.
“I think democracy is in trouble because people see it leading to deadlock, division, polarization and name-calling, and it’s not solving the public's problems,” Fishkin told me. “They see the political class is out of touch and maybe not to be trusted, yet the policymakers and political elites see the specter of an angry populism. But those who are really angry don't necessarily speak for everybody else —they're the ones who speak the loudest. So you need a mechanism that represents everybody," and one that creates "conditions where they can think. That's what we’re trying to accomplish.”
Some Americans just yearn to return to "normal" after the intense disruption of the last few years. In some respects, that’s just what we should do. But in other ways, what’s normal in our system has always been problematic, lacking both universal inclusion and vibrant participation. We have it in our power to change that.