Keynote Address:
Dialogue: A Solution to the Public Policy Disconnect?
May 16, 1998
9:00 - 10:00 AM

Introduction: Bruce Jennings, MA
Keynote Speaker: Daniel Yankelovich, President, Public Agenda Foundation, New York, New York, Author, Coming to Public Judgment

Thank you and good morning. I have been asked to talk about dialogue as a possible solution to the troubling public policy disconnect in this country. I was invited to focus on three questions: (1) the need for public dialogue; (2) its potential in assisting sound policy-making on genetic issues; and (3) the role of institutions such as the media, universities, political leaders and opinion surveys in policy development.

These questions are very much to the point, so I will use them as the framework for my comments. In the first part of my talk, I will address the question of the need for more public dialogue. Then I will talk about the special requirements for dialogue on genetic issues, and I will conclude with a discussion of the changes our institutions will have to make to accommodate a dialogic model of public engagement.

My daughter, who always goes to the heart of a matter quickly, put the point to me when she learned I was going to speak at this conference. She said, "Please don't let the geneticists do to humans what they have done to the tomato." That comment sums up the mistrust, the feeling on the part of the public that government and science, left to their own devices, are going to do to humans what they did to the tomato: shine up the appearance, and ruin the flavor.

Most public policy issues, I think it is fair to say, are suffering from lack of resolution because of failure of genuine public engagement. The result has been a serious and increasing disconnect between public policy and the public. Evidence of the mounting failure of standard methods of public engagement is piling up. Standard methods, such as public education, advocacy, and PR, are not working very well, either in practice or in theory. They are not working as well as they once did for a variety of reasons.

Experts and élites live in one world, the public in another. These worlds have different concerns, agendas, vocabularies, and subcultures, and are barely connecting with each other. Consider education as an example. Our studies show that parents are concerned, sometimes to the point of desperation, with disruptive kids who are preventing their own kids from getting the kind of education they want for them, the kind of education that will enable them to compete in the world that is emerging. But most experts and professionals in the public education field are concerned with other issues, like preserving the public school system and being attentive to the self-esteem needs of the students. These are legitimate concerns, but they are different ones. So you have these two groups—the public and the experts—passing each other by, each preoccupied with its own concerns, and standard methods of communication aren't helping them to connect.

In the focus groups done for this Genome Policy Project, there was a fascinating example of this sort of disconnect. Many of the questions raised by genetic issues cause people a great deal of anxiety, and in their anxiety they are looking for help and counseling in order to reach their own decisions. But the focus groups with physicians showed a very different set of concerns. The doctors had no interest in helping people reach their own decisions. They wanted to ladle out advice: "I lay out the options and I tell them which one is the best one for them." On the surface this may seem like a simple stylistic difference, but it goes much deeper: on such fundamental matters people want to reach their own decisions rather than being told what to do in a very directive way. Consider another example: in the debate over the Clinton health care plan several years ago, the Clinton Administration's concern was with extending coverage to everyone, while the public's concern was with mounting costs. The two were never able to get together, with results that we all know. In the realm of foreign policy, the University of Maryland Survey Research Group recently completed a fascinating study. It shows that Congress thinks the public is much more isolationist than it in fact is. It seems to serve the ideological commitments of some lawmakers to make this allegation. They have a different agenda than the public and many gestures, like refusing to pay UN dues, are made in the name of a public resistance that does not actually exist. So everywhere you look, you see the concerns and agendas of élites moving in different directions than the concerns and agendas of the public.

This divide between public and elite subcultures is reinforced by a number of factors. Television is an all-important medium for involving the public, but it has its own subculture which features an addiction to drama and conflict. TV’s approach to issues is to pick representatives from the extremes and let them duke it out. With an issue like abortion, for example, the pro-life and pro-choice extremes together represent about 10% of the public. But since the extremes are the only ones represented on TV, the moderate 90% of the public gets very confused because they gain a false impression. TV’s message on abortion is that the country is sharply divided between the two ideological positions (pro-choice and pro-life) when in fact it is nothing of the sort. In recent years, advocacy groups have been entering local election campaigns whether the local candidate wants them or not, and whether they help the candidate or not. This is all very confusing to the public, a public that is turned off by such total unresponsiveness.

The divide between élites and the public produces a number of secondary effects. The lack of institutional responsiveness leads to mistrust, especially toward the government. Over the last twenty-five years, there has been a dramatic reversal in public trust of government. Twenty-five years ago, more than three-quarters of the public said they trusted the government to do the right thing most of the time. Since then, there has been a gradual but steady erosion of trust, until today when the numbers are totally reversed. Now, more than three-quarters of the public do not trust the government. That is a massive reversal, and as a result of it, there is often an initial paranoid public response to anything the government wants to do. In the dialogues you conducted with the public, when you were dealing with public policy issues, the conflict for people was exacerbated whenever it looked like government might be a solution. This increasing mistrust in government results in huge public resistance to government-based solutions.

The public thinks that our country's problems are serious enough to warrant non-partisan approaches to them. Voters want the Republicans and Democrats to work together to deal with them. But because of their own subculture and their own agendas, the two parties take the partisan route, which frustrates the public.

Almost everywhere you look, you see that communication with the public is not working. The situation is even worse when examined from a theoretical perspective. For example, if you identify the premises of the public education model they go something like this: the objective of public education is a better informed public, a public that has more facts at its disposal. The method of choice is "top-down" communications where experts share a fraction of their knowledge and information with the public. The ideal is that of a value-free presentation, so that experts present their information to the public free from the bias of their own personal values and opinions.

Communication is assumed to occur in real time and in a single step—you give the information and the person on the other side receives it in the same real time frame. I always think of the example of two notes that two different women might leave for their husbands. One reads, "I am leaving your dinner in the fridge," and the other says, "I am leaving you." The information in the first note is transmitted in real time: it takes 10 seconds to read it and it takes 10 seconds to absorb it. In the case of the second message, it may take the husband 10 seconds to read it, but it may take him months and years to absorb and deal with it.

The public is also assumed to be a blank slate on which a message will be received and understood as the source of the message intends. You have, for example, a managed care doctor saying, "I do not think you need that test." The person hearing the message reacts, "I probably do need that test, but the doctor is putting the profits of the HMO ahead of the care I need." The public is never a tabula rasa; the information given is always altered by the recipient. If you go down this list of assumptions (and I want to contrast them with the assumptions made in a genuine dialogue), you find that every single one of them is problematic, and in most cases they are just plain wrong. We have a system of public education based on theoretical presuppositions that are mainly false.

My observations about how the public arrives at judgments have been reinforced by another form of experience. In recent years I have served on a number of corporate and non-profit boards. In attending board meetings, I began to notice that for the most important decisions, the facts were scanty and the analysis even more so. Something else was going on. Both with the public and in these boards, what was going on was, in effect, a dialogue, a way of people encountering each other and looking at a problem from a diversity of points of view. Through dialogue as a method of reaching judgment, people can affect each other's judgments and responses as a method of reaching judgment in a way that just does not correspond to the official epistemology, to the unwritten set of presuppositions we have about the way people actually arrive at judgments.

If you think about it, the dialogic process is very different from the public education process. Let me contrast the premises of a dialogic model with those of the public education model. First: in a dialogic model, the objective is not being better informed. The objective is either enhancing mutual understanding or seeking common ground. Second: dialogue is not a top-down method; people are encountering each other much more on a basis of equality. Third: in dialogue, the fact / value distinction, which is so important to social science, along with the notion of value-free social science, is totally abandoned in favor of a process where facts and values are dealt with simultaneously, which is the way the issues actually occur in the world and is the way that people like to deal with them. Fourth: the communication process is not a real-time, single-step process, but occurs over a period of time and goes through a number of different stages and phases.

In my book Coming to Public Judgment, I outline seven stages that the public passes through from their initial raw opinion to the final stage of thoughtful, considered judgment. The stages are complicated, and the process can vary from hours to years to decades. Dr. Fleck mentioned several times in his introductions to your dialogues that he thinks this process will take 20 years or so before people come to a considered judgment on genome issues. That is probably a correct appraisal. As I stated earlier, the public is not a blank screen on which messages are written. The public is an active participant. Judgment is shaped by the interaction, and is not received passively as a simple message from the outside. We are not talking about a slightly different model; we are talking about a radically different one.

The whole notion of public education as top-down information sharing where I, the expert, impart a tiny fraction of my profound knowledge and understanding and wisdom to the great unwashed masses just does not work. The dialogic model provides a different conception. Personally, I have come to the conclusion that dialogic communication has the potential to be the solution for some of the most serious problems our nation faces. But it is not an automatic or an easy fix. There are lots of pieces that are missing and must be put into place. The great advantage of the standard methods of public education is that we understand them. We have the skills and institutions to implement them. If we switch, even partly, to a dialogic model, we have to confront the reality that many of the skills and institutions we need are lacking, or actually work at cross purposes to what we do today.

Dialogue On Genetic Issues
With this context in mind, let me now turn to the second question: the potential for dialogue on policy associated with genetic issues. I think that as a society we are at the very earliest stages of a long road to fill the need for alternative forms of public engagement. Lots of experiments are going on, and lots is being learned about how to implement new forms of communication. The Clinton administration early in its tenure tried out Town Hall Meetings, and now is supporting dialogues on race and social security, for which they deserve much credit. Unfortunately, I don't think they understand the rules of the game, so that a lot of these experiments are not working. But at least they realize that a different format is needed.

At the University of Texas, political scientist Jim Fishkin has come up with a number of experiments on what he calls "deliberative polling," giving people a chance to wrestle with issues in a deliberative way. Your Genome Policy Project is also an innovation moving in the same direction. A number of the organizations with which I am affiliated, such as the Public Agenda, the Kettering Foundation and the National Issues Forum, represent other efforts in this direction. MIT has a Center for Dialogue under Bill Issacs and Peter Senge. The list goes on, and demonstrates that our society is trying to come up with innovative approaches in this direction. I guess I should add the book that I have just finished, entitled The Magic of Dialogue.

All of these approaches have common elements. First, they have a broader objective than simply imparting information; they focus on deliberation, finding common ground, resolving ambivalence and conflicts, and so forth. They avoid or minimize the top-down aspect; they are generally interactive and focus on give-and-take rather than one-way communication; and values are considered as well as facts. And they all assume that the public brings preexisting values, preconceptions and assumptions to the table.

But in addition to the common elements, each of these experiments differ, sometimes in far-reaching ways. This matters at several levels. First of all, everybody is using the word "dialogue" to refer to something different, so that when you think you are talking about dialogue, people may have in mind rather different processes. This semantic confusion may not matter much, but what is important is that at the level of substance these differences count heavily. And you may not realize which particular form of dialogue is appropriate for which purpose. In the race dialogues that the Clinton Administration launched, the chairman initially refused to include people who disagreed with the majority, so what you ended up with was a form of preaching to the converted, full of self-congratulation. It undercut one of the main purposes of dialogue, which is to find common ground among people who disagree with one another. The people who were excluded complained to the President, who invited them to the White House. They came out of the meeting with him feeling very good about the fact that he had listened and responded and understood their concerns. But he never followed-up or responded to them after the meeting, so they grew angry and have now formed their own organization, which repeats the same kind of miscarriage that was the case with the first effort.

The more experience I have with dialogue, the more I realize that some forms of it are good for some purposes and not so good for other purposes. I thought it might be useful for you to contrast the conception of dialogue you use in your Genome Policy Project with the conception of dialogue I develop in my book The Magic of Dialogue. They are different conceptions. This does not mean that you are right and I am wrong. And it does not mean the reverse. They are just different conceptions, and I think there is a way of resolving them and understanding which form is right for which purposes. In your Genome Policy Project, dialogue is used in a broad and general sense. It refers to a process that is interactive, that gives people propositions or questions to react to, and that devotes a certain amount of time to discussing each one. In my work, I reserve the term "dialogue" for a much more restrictive usage, which I would like to characterize briefly.

My approach mirrors that of a number of other theorists working in the dialogue field who make fairly sharp distinctions among the three D’s: debate, discussion, and dialogue. In the past there was little need to make these kinds of distinctions and to differentiate among various forms of conversation. You could use dialogue, debate and discussion interchangeably. On the Lehrer Report there is a segment called "Dialogue with David Gergen." The only difference I can make out between Gergen’s "dialogue" and an ordinary interview with an author is that Gergen has actually read the book. This is a compliment to the author, but somehow it underlines the fact that if there is any unusual aspect to a conversation it can get the label dialogue. However, for certain purposes, it is important to clarify how dialogue, debate and discussion differ from one another.

The difference between debate and the other two is clear. The purpose of debate is to win, not to achieve mutual understanding. There is no such thing as "losing" a dialogue. You can lose a debate, you cannot lose a dialogue. In a dialogue, either everybody involved loses or everybody involved wins.

The distinction between discussion and dialogue is trickier because these two forms of conversation often share the same purpose, that of reaching mutual understanding and agreement (which is the way the dictionary defines dialogue). However, they achieve this purpose in different ways. Dialogue is far more demanding than ordinary discussion, because it imposes three requirements that general discussion does not. First, it requires equality of standing in the context of the dialogue: the people sitting around the table may be very unequal in status before they come into the dialogue, but in the context of the dialogue they must participate as equals. Second, dialogue requires empathic listening, putting yourself in the shoes of the other person and demonstrating that you have done so. Third, it requires assumptions to be brought out into the open in an atmosphere of acceptance—people need to be able to bring out the assumptions that govern their points of view and not have everybody jump down their throat, but instead suspend judgment, entertain them, listen to them, respond to them.

Now, in my terminology, which admittedly is arbitrary, the Genome Policy Project would fall under the heading of a "guided discussion" rather than a "dialogue". Guided discussion can be very valuable. The technique used is one I have often used. I tend to think of it as a "stimulus-response" technique. You stimulate the group by provoking them in some way, usually with a stronger, more pungent, more extreme statement than they themselves might make. This evokes a response and gets a very good discussion going. But if you take the three requirements I have laid down for dialogue, you will see that the methods of the Genome Policy Project do not meet them. In the Genome Policy Project the moderator is definitely more equal than the others. He or she brings authority, structure, sets the agenda, formulates the issues and limits the discussion. At yesterday's mini-session there was lively participation, a great deal of openness and the glimmerings of empathy, but the structure did not permit this empathy to unfold because the group had to move on to the next question. So as soon as there was somebody who was beginning to understand the concerns of someone else, the discussion shifted ground. I understand that just from the simple, practical point of view, given the immense scope of what you are trying to accomplish, there is not time to have the process of empathic understanding develop and unfold. But this constraint undermines dialogue. As for the third requirement—bringing assumptions into the open and responding to them non-judgmentally—there is not enough time for that either. This is not an easy thing to do, and it also takes a certain amount of training and understanding.

The limitation of my concept of dialogue is that it is demanding; it is time consuming; and it requires special skills. So it ought not be applied when other simpler methods can be used. If you can accomplish the same result through top-down communication or through debate or through guided discussion, you should do so. But there are certain circumstances when you cannot accomplish your goals through these methods and you need dialogue in my sense of the term.

I would cite four conditions under which you need to resort to dialogue in my sense. One is where subculture differences divide the participants. This condition does not apply to the Genome Policy Project as it is now structured, because your consumers do not belong to different subcultures. If you included the physicians and other professionals, this condition would apply. The second condition, probably the most essential condition for dialogue, is when mistrust among participants exists. Issues like race, managed care, social security, labor-management relations, deep conflicts such as exist between the Israelis and the Palestinians, and so forth, these are the classic conditions where you need my kind of dialogue. Here again, this condition doesn't apply to the Genome Policy Project. There is some mistrust of government, but that is not mistrust among the participants themselves. So the first two conditions for dialogue do not apply here.

However, the third and fourth conditions are applicable. The third condition is when you are seeking resolution of value-laden issues where there is either ambivalence within the person or conflict among the participants. On the Genome issue there is profound ambivalence, and it is ambivalence of a very interesting sort, as revealed by the focus group research that you did.

From a practical point of view, genetic testing makes a lot of sense to people. However, it evokes some fundamental value conflicts. I have been trying to understand what the common denominators of these conflicts are. It seems to me that they have something to do with the values of hope and courage. People say things like, "we feel that things are going to work out for the best anyway" and "you have to cope with whatever comes along" and "maybe God wanted it to work out this way." There is a whole range of feelings that represent the way people deal with lives of adversity where they are confronted with difficulties at every stage, and yet keep up their hope and their courage. Genetic testing interferes with this process, creating ambivalence. This ambivalence is not a surface one; a brief discussion won't permit people to resolve it.

The fourth condition is when people are confronting an unprecedented change that requires a lot of intensive working-through time. In this case, dialogue can accelerate the working-through process. Something that might otherwise take years could be accelerated enormously by this kind of dialogue encounter.

My general observation is that the Genome Policy Project has been a very valuable experiment and learning experience. As it evolves and you go on to the next stage you might, in replicating it, simplify it or elaborate on it in two ways: Some of the questions you deal with don't require as much discussion, as much structure, or as many meetings as you now give them. Other questions require much more. For example, in the questions about laws requiring people to take genetic testing, there is a shift before and after the discussion. About 77% are against it to start with and after discussion this number goes up to 88%. It does not take a lot of discussion to take already-existing concerns and reinforce them. Another question asked is whether society should pick up 80% of the cost for in-vitro testing to avoid genetic disorders. In the pre-discussion survey, 40% thought this was a good idea, while in the post-discussion survey this number was cut in half, to about 22%, because people realized how high the costs would be. These are important issues that come up quickly through discussion and you can deal with them and move on quickly.

On other issues, however, where public policy decisions are urgent or where there is ambivalence or conflict, or where practicality collides with deeply-held values, you might want to resort to the kind of dialogue that is closer to the model I am talking about. It is my guess that your kind of guided discussion, which can be quite expeditious and brief, will be suitable for a majority of the issues you are concerned with. But the more extensive kind of dialogue I am talking about is appropriate for some of the other issues. You surely need dialogue for the times when you bring experts and the public together in order to look at changes that people have difficulty confronting because they are so huge, and also at changes that stir up a lot of ambivalence and conflict.

Strategies
In my book, I identify 15 different strategies for carrying out my conception of dialogue, and I want to just give you a few quick examples of the sorts of things I suggest. One is the importance of keeping dialogue and decision making compartmentalized. If you try to mix them together you create an unholy mess because two radically different processes are involved. For decision-making you do not need equality of understanding and you don't need empathic listening; you need a different process. So you have to have a method of keeping them separate.

In order to bring assumptions into the open non-judgmentally, one strategy would be to bring forth your own assumptions before speculating on those of other people in the dialogue group—this avoids a little bit of hassle. Another strategy for dialogue is a technique that we have developed at the Public Agenda in conjunction with Kettering, which we call "choice work." The choice work concept is to package choices for people to consider in advance. Scenarios are created where real choices are presented, and their values and pros and cons made explicit. This saves an enormous amount of time. It is also useful in that values and facts are brought together at the same time.

In my book, I also identify some ten potholes that people fall into when carrying out this sort of dialogue. Here again I give you just a couple of quickie examples. The most dominant one is holding back. A typical experience would be where a day-and-a-half is set aside for a dialogue because the need for an enormous amount of time is recognized. And at 11 o'clock on the second day somebody finally begins to open up. You need to have some techniques and methods for getting people to let down their guards. People are not on their own territory, so they are guarded and careful and unwilling to open up unless other people open up first.

Another pothole is different starting points. In your discussions, some people who had dealt with genetic issues were much further along than people who were just beginning to think about them. Another kind of pothole is that there is always, in every dialogue group, at least one person who has a pet preoccupation. If you don't listen to it two or three times and let them get it out of their system, you can't go on. It is a practical matter—you have to let people sing their arias so that they can get their preoccupations out of their systems.

There is a lot going on in dialogue, but as a society we don't understand it very well; it is not our thing. We can become skilled at dialogue, but there is a lot that would need to be done first by way of institutionalizing it and understanding it.

Institutions
Let me turn very briefly to the role of institutions. I think the role of institutions is indispensable to deploying the dialogic model. Most institutions now support the standard methods of public engagement. We now need to make room for new methods, and we probably need new institutions. Right now I can only touch on this vast topic, on a few of the institutions in question.

First, I would like to address universities. I think that this is a wonderful opportunity for the social science and humanities departments of our great research universities to reconnect to mainstream America. One of the most interesting and important areas that the universities could take on would be to examine the creation of new institutions to create public space in today's world for dialogue and other interactive forms of public engagement. Existing institutions do this extremely well for élites. There is something that is sometimes called the "Invisible University." We are now attending a class in the Invisible University. All over the country, in every conceivable method, way and form, élites are meeting with each other. They are meeting within homogeneous groups, across groups, listening to outsiders, talking among themselves; it is an extraordinary phenomenon. But this opportunity does not exist for the public and it doesn't exist between the public and experts. Our democracy needs public space in which to do this, and it seems to me that this is a wonderful challenge for our universities.

I think that the media have an opportunity to move from being part of the problem to being part of the solution. A giant contribution would consist of helping to find a solution to the problem of how to go to scale. Dialogue is a process that works best on a small, intimate scale. So how do you adapt a format that works ideally for a dozen to 20 people at a time to thousands, millions, tens of millions of people? Talk radio is a move in that direction. I've also encountered a couple of other examples. On the eve of the Gulf War there was a so-called debate in Congress which was unique in my experience. One person after another in the Senate and House got up and dropped their lawyer, adversarial role and agonized in public about how difficult the decision was for them. They aired their feelings and concerns about putting soldiers in harm's way. This vicarious experience helped listeners to work through their own conflicts and lack of resolution. Bill Moyers’ show on Genesis, and the dialogue there, to which millions of people responded, again was a vicarious experience that people could identify with.

The principle is not novel; it taps into the dynamics of identification, story telling, and drama, but to adapt it to this mode would take creativity and imagination and experimentation, and the will to do it. The media have the creativity and the imagination, but there aren't many signs that they have the will.

I am going to talk very briefly about political leadership by mentioning an experience I had recently in consulting with the government of Canada. I was asked to talk about new ways of public engagement. I started to talk about some of the ways in which these officials could engage with the public and overcome the communication gap, and it became clearer and clearer that they did not want to do it. In the worst possible way, they did not want do it. The amount of resistance to going out there and listening and reacting and talking to people as if they were equals, rather than telling them what to think and informing them, was enormous. It just did not make any sense to them. They could see what the public could get out of being there, but they could not see what they could get out of listening to the public. So I do not think that one should underestimate the kinds of resistance there are to dealing with these sort of issues.

Let me conclude with a few words from my own field of public opinion research. I think that surveys can make a unique contribution to this model on an issue that it usually does not focus on, which is the enormously wide time frame for reaching judgment and consensus. On any one issue, the likely time required for resolution and judgment is a very important datum that you can gain through research, but almost never do. I think that survey research can help by monitoring progress along a time continuum through trend-tracking studies. For example, at DYG we have been doing a national trend study of changing social values in the country for 30 years now. And on issues like the changing role of women and attitudes towards this change you see resolution and movement that is extraordinary. Every year, this is probably the most dynamic area of American life, and if you track it you see it in a way that you can't see it in any other form.

Before and after studies of small groups exposed to varying degrees of dialogic interaction can isolate how much time and intensity of dialogue it takes to move people along the seven stages of judgment. The intensity of dialogue and the time required will vary enormously from issue to issue. Survey research can identify factors that cause people to shift their position. Greater confidence in government, for example, would make a huge difference on many policy issues.

Through comparing the results of small group dialogues with the public at large, surveys can produce insights into the direction of public sentiment, the factors that might accelerate the voyage to public judgment, and information about where the public stands at any one moment. In other words, you can learn from the laboratory experiment of a dialogue among a small group what might help move the country as a whole along on this voyage. In short, survey research is an indispensable part of the dialogic model, not simply to record change, but more fundamentally to advance the theory underlying the dialogic model and to help us practice it on the large scale required for issues that affect the nation as a whole.

In conclusion, I would like to congratulate the Genome Policy Project on a terrific experiment and simply make once again the modest suggestion that as it proceeds to later stages it uses the two tools of discussion and dialogue in conjunction with one another. That you can discriminate it and then evolve, and it seems to me that is the way that really fruitful and creative research does evolve. Thank you very much.

If you have any questions regarding this document please call 734-936-1226 or email genpolicy@umich.edu.
Copyright © 2000 Genome Technology & Reproduction: Values & Public Policy Project