The Big Kozinski By Emily Bazelon
Righting the Ship of Democracy By Bruce Ackerman and James Fishkin
Smooth Sailing By Richard Posner
The Wrong Tack By Arthur Lupia
Righting the Ship of Democracy
Presenting Deliberation Day: A radical proposal to help voters make better decisions.
In our soon-to-be-released book, we offer a new way of thinking about democratic reform, proposing a new national holidayDeliberation Day. It would replace Presidents' Day, which does no service to the memories of Washington and Lincoln, and would be held two weeks before major national elections.
Registered voters would be called together in neighborhood meeting places, in small groups of 15 and larger groups of 500, to discuss the central issues raised by the campaign. Each deliberator would be paid $150 for the day's work of citizenship. To allow the business of the world to carry on and as many as possible to participate, the holiday would be a two-day affair.
If Deliberation Day succeeded, everything else would change: the candidates, the media, the activists, the interest groups, the spin doctors, the advertisers, the pollsters, the fundraisers, the lobbyists, and the political parties. All would have no choice but to adapt to a more attentive and informed public. When the election arrived, the people would speak with a better chance of knowing what they wanted and which candidates were more likely to pursue the popular mandate.
WHY CAN'T PEOPLE SIMPLY ORGANIZE THEMSELVES without the assistance of a new civic holiday and its associated social engineering? After all, we don't live in a civic vacuum. Sustained conversations do take place in countless settings, from the breakfast table to the coffee break at the office to the meeting at the neighborhood church or union hall. And their intensity and frequency do increase during election campaigns. But the social context that motivates public deliberation is usually lacking, and the resulting levels of public information are disappointing.
If six decades of modern public opinion research have established anything, it is that the general public's political ignorance is appalling by any standard. As one influential researcher concludes, "the political ignorance of the American voter is one of the best-documented features of contemporary politics." And another: "The verdict is stunningly, depressingly clear: most people know very little about politics, and the distribution behind that statement has changed very little if at all over the survey era."
George Bishop and his colleagues at the University of Cincinnati dramatized this point in their study of attitudes toward the "Public Affairs Act of 1975." Asked for their opinion of the act, large percentages of the public either supported or opposed it, even though no such act was ever passed. In 1995, The Washington Post celebrated the "twentieth unanniversary" of the nonexistent act by asking respondents about its "repeal." Half the respondents were told that President Clinton wanted to repeal the act; the other half were informed that the "Republican Congress" favored its repeal. The respondents apparently used these cues to guide their answers, without recognizing the fictional character of the entire endeavor.
Even opinions about actual issues are often highly unstable impressions based on sound bites and headlines. Some researchers argue that citizens can function effectively without the kinds of specific knowledge called for by most survey questions. What is really important, they claim, is for voters to place candidates or political parties in the broader framework of the basic liberal-conservative spectrum. Yet Robert Luskin, a political scientist at the University of Texas, has shown that the American public does a terrible job at this task as well. Once corrections for guessing and nonresponses are introduced, surveys show that the American public does slightly worse in classifying parties or policies as "liberal" or "conservative" than it would do if it proceeded by flipping a coin.
None of this is really controversial. Indeed, the past generation of political economists has gone to great lengths to explain why voter ignorance is only to be expected. Acquiring and analyzing information is a time-consuming business. Time spent on public affairs competes with time acquiring information on more personal matters, like the price and quality of cars or houses. In these areas, each of us suffers a direct cost for ignorant decisionsI may buy a lemon unless I am careful to analyze my options ahead of time. In contrast, nobody pays a price for voting ignorantly, since the outcome of a major election never hinges on a single ballot. As a consequence, it may well be "rational" for individual voters to remain ignorant about public matters. They can then reserve all their time for analyzing information on cars, houses, and other matters of personal consumption, where the sanction for ignorant decisions is felt directly.
This point doesn't depend on whether voters are public-spirited citizens. Even if they are deeply concerned about the nation's future, their individual votes still don't make a difference, and so there isn't an instrumental reason to make their choice a well-informed one.
THE PROBLEM THIS RAISES IS OBVIOUS: Why should politicians consider the interests of all citizens if most voters are uninformed and selfish?
Since the days of Madison, constitutional thought has struggled with this problem, and there is no reason to think it will ever be solved definitively. Nevertheless, the political world is constantly changing, and these changes alter the terms in which the problem is expressed and the institutional modes through which it may be ameliorated, if not resolved.
Madison famously focused on the capacities of political elites to filter out the most irrational and self-interested aspects of public opinion. One of the aims of The Federalist Papers was to defend a constitutional framework that subtly rewarded elites for filtering, rather than mirroring, these selfish impulses, thereby encouraging leaders to steer the republic in more enlightened directions. Our first task is to consider how the introduction of Deliberation Day might contribute to this filtering process.
Our argument proceeds by appealing to the "law of anticipated reaction": Politicians try to look ahead and adapt their behavior to new risks as they emerge on the horizon. Suppose that a groundswell of popular support finally pushes our proposal through Congress in the form of the Deliberation Day Act of 2012. Put yourself in the place of the man or woman who wins the presidential election that year. Other things being equal, how will the introduction of Deliberation Day modify your political calculus as president?
Consider a central aspect of modern politics: government by public-opinion poll. No sitting politician would think of taking an important step without hiring a pollster or two to test the waters, and modern White Houses are famous for their elaborate polling operations. Deliberation Day will require presidents to rethink their relation to this steady stream of polling data, and in ways that promise a more reflective relation to the public good.
Our new holiday need not attract a huge turnout to the nation's schools and community centers. The key question is not how many Americans use the holiday as a convenient excuse to catch up on their sleep, but whether the folks who do show up can swing the election. Call this a critical mass.
If a critical mass does show up, modern polling will suddenly seem old-fashioned, and present-day techniques will dramatically depreciate in political value. Today, what polls of citizens generally measure are raw, poorly formed preferences based on very weak information. But a politician has no reason to believe that her constituents' preferences will be much less raw on the distant day they go to the polls; the campaign mobilization may make some difference, but not a huge one. As a consequence, the politician can extrapolate existing polling data to the futurethat is, Election Daywith a certain degree of confidence.
If Deliberation Day is established and a critical mass shows up to participate in the holiday, sophisticated politicians will no longer be so interested in monitoring the existing patterns of raw preferences. They will want to know about their constituents' refined preferences: what the voters will think after they have engaged in the discussion and reflection precipitated by Deliberation Day.
DELIBERATION DAY IS A NEW IDEA, but it builds on a host of smaller experiments involving ordinary citizens deliberating on public issues. In many different forums, in different cities and countries around the world, citizens have gathered together for experiments in serious and balanced public discussion. Many of these experiments have proved remarkably successful, but we will focus on one particular method of citizen consultation, the deliberative poll. Because the deliberative poll, or DP, is designed as a social science experiment, it provides the best evidence for the viability of our proposal. Since one of us, Jim Fishkin, has spent the past decade of his professional life designing and observing deliberative polls on a wide variety of issues, we can use these experiments with a solid understanding of their strengths and limitations.
A deliberative poll is a survey of a random sample of citizens before and after the group has had a chance to deliberate seriously on an issue. The process begins by selecting a representative sample from the population and asking each person a set of questions on the issue raised at the deliberative poll. This initial survey is the standard sort conducted by social scientists doing public-opinion research. The respondents are then invited to a specified place for a weekend of discussion. A small honorarium and travel expenses are paid to recruit a representative sample.
In preparation for the event, the participants receive briefing materials to lay the groundwork for the discussion. These materials are typically supervised for balance and accuracy by an advisory board of relevant experts and stakeholders. On arrival, the participants are randomly assigned to small groups with trained moderators. When they meet, they not only discuss the general issue but try to identify key questions that merit further exploration. They then bring these questions to balanced panels of competing experts or policymakers in larger plenary sessions. The small groups and plenary sessions alternate throughout the weekend. At the end of the process, the respondents take the same questionnaire they were given on first contact.
What have we learned from our study of DPs that might be relevant to the viability of Deliberation Day?
First, deliberation makes a difference. When one compares the attitudes and opinions that respondents had at the end of the process with those they had on first contact, there are many large and statistically significant differences. As in the first deliberative poll on crime (broadcast on Britain's Channel Four in 1994), it is not unusual for deliberation to significantly change the balance of opinion on two-thirds of the policy questions. And more than half of the respondents typically change their positions on particular policy items after sustained conversation.
Carmel Meredith, a British participant in the DP on crime, put the point in human terms. Speaking on the TV broadcast, she said that "the questionnaire that I filled in four weeks ago, I might as well rip up now and put it in the bin. It was an absolute waste of time because I didn't know enough about it." After discussions, she had come to seriously re-evaluate the top-of-the-head views that she had given to the interviewer in the initial survey. In describing her experience, she was speaking for many other DP participants.
What is going on to produce this kind of change? Some of it occurs in anticipation of the event. During the British DP on crime, the spouse of one of the respondents came up to thank Jim Fishkin. In 30 years of marriage, her husband had never read a newspaper, but the DP invitation had prompted him to read "every newspaper every day" (and, as a result, he was going to be "much more interesting to live with in retirement"). This encapsulated our aspiration to create incentives to overcome rational ignorance and to motivate people to behave a bit more like ideal citizens.
The research that has come out of deliberative polls suggests not only that participants change their political attitudes but that these changes are driven by better information. It suggests not only that these changed attitudes generate different voting intentions but that these preferences become more public-spirited and collectively consistent. These changes occur throughout the population and aren't limited to the more educated. Finally, deliberation is intrinsically satisfying once people are given a serious chance to engage with one another in an appropriate setting.
THESE RESULTS CONTRAST SHARPLY WITH A GROWING LITERATURE in political science. On this view, citizens who became well informed wouldn't vote very differently than they do today. Advocates of this position suggest that poorly informed voters already use various simplifying devices to approximate the same conclusions that they would reach if they were well informed. If citizens know, for example, who is for a proposal or who is against it, they can use this cue to guide their voting behavior without taking the trouble to study the issue themselves.
Samuel Popkin, author of The Reasoning Voter, presents the example of President Gerald Ford choking on a tamale while campaigning against Ronald Reagan in the 1976 Republican primary in Texas. The president didn't know that he was supposed to shuck the tamale before eating it. Popkin claims that Mexican-American voters could reasonably infer from Ford's ignorance about their food that he knew little about them or their culture. As a consequence, they had good reason to prefer Reagan, without needing to know much about the detailed policy positions of the rival candidates. Popkin's example serves as a caution about the potential abuses of informational shortcuts: The unshucked tamale did not, in fact, serve as a very good indicator of the comparative positions of Ford and Reagan.
Apart from such anecdotes, the thesis, as the political scientist Arthur Lupia puts it, that "shortcuts" can produce the same voting behavior as "encyclopedic" knowledge rests primarily on statistical models developed from survey data. These models compare the preferences of uninformed voters with those who are better informed but are otherwise similar. But it is possible to construct equally plausible models from survey data that lead to a very different conclusion: that the public's ignorance makes a real difference in how elections come out. If people were better informed, they would vote differently, and these differences would be large enough to change election results. The deliberative polls provide convincing evidence to support this view. As we have seen in the cases of Australia and Britain, relatively ignorant voters do change their voting preferences substantially once they deliberate on the basis of better information.
UP UNTIL NOW WE HAVE DELAYED DISCUSSION OF COST. Deliberation Day will, of course, be expensive. But there is more at stake than totaling up the dollars and cents. We must ask what the dollar signs mean. Is Deliberation Day like an enormously expensive Mercedes-Benz, which we are free to reject merely because other consumer goods better satisfy our desires? Or are the costs involved more akin to those spent on educating the young or defending the country?
We can't answer these questions with economic reasoning alone. To motivate the requisite political reflection, we focus on one of our "big-ticket" items: the proposal to pay each participating citizen a stipend of $150. Such payments are hardly unprecedented; similar (but smaller) stipends are paid every day to Americans serving as jurors in criminal and civil trials throughout the nation. But these $150 payments will add up to many billions each Deliberation Day, especially if our initiative is successful and tens of millions of Americans turn out to discuss the issues.
This big budgetary item will lead us to reflect on a curious asymmetry in modern public finance. National and local governments spend hundreds of billions of dollars a year on highways and healthcare and other goods and services. When the state is viewed as a machine for satisfying needs and wants, our budgetary imagination knows no bounds. But when it comes to citizenship development, we spend almost nothing. Is this disparity justified? Is the $150 stipend a mere luxury or an essential aspect of our initiative?
From the layman's point of view, it may also seem very costly to use countless school classrooms and governmental office buildings as sites for tens of millions of deliberators. But this may not be the case. It should be said here that we are not giving Americans a new day off for Deliberation Day; we are appropriating an already existing holiday, Presidents' Day, for a better civic purpose. Currently school buildings remain empty on Presidents' Day, when the government closes for business. Since the buildings are empty for a day anyway, there is no opportunity cost involved in using them for deliberative purposes, and we should not add billions of dollars to our cost estimates to reflect their "rental value." The same point applies to the economic production lost when half the working population takes at least one day off during our two-day holiday. Since many people currently take Presidents' Day off, the opportunity cost is greatly reduced along this dimension as well.
This hardly implies that Deliberation Day will be cheap. Lots of work will go into preparing for the holiday, organizing its operations, and cleaning up afterward. (And don't forget the free lunch served to tens of millions of deliberators in school cafeterias throughout the land!) While these real resource costs will range in the billions, the magnitudes will seem quite manageable once placed within a suitable political and economic framework. (For those who are interested, a more detailed cost analysis is located in our book.)
Americans may well reject our initiative, but they should not do so because of its economic costs. The question is whether ordinary people can convince themselves that popular sovereignty has a future in America and that it remains possible to take back control over politics by constructing a new place for face-to-face dialogue. If Americans retain their democratic faith, they will find the economic costs easy to accept; if they give it up, much more than Deliberation Day is at stake.
WE HOPE THE NEW NATIONAL HOLIDAY, like the deliberative polls, will lift participants out of the routines of everyday life. There will be an air of anticipation as tens of millions prepare for the day's discussions. Participants won't want to make fools of themselves before their neighbors. Millions will start paying more attention to the news, and their dinner table conversations will begin to focus on Deliberation Day as it approaches. There will also be lots of hubbub during and after the holiday: exit polls, countless conversations, media commentary. The overall level of popular engagement will far exceed that of any DP in the past or future. And Deliberation Day will retain the features that have made DPs such a powerful tool for reflective decision making: Each deliberator is one voice in 15 in a small group and one voice in a few hundred at the plenary sessions. We may be old-fashioned, but we continue to believe that substance counts for more than media hype. Deliberation Day will provide tens of millions with the opportunity to take the great phantom of "public opinion" away from pollsters and political advertisers and make it a creation of popular deliberation. We should never underestimate the power of simple ideas to inspire genuine commitment.