Response to Critics of When the People Speak: The Deliberative Deficit and What To Do About It

These essays do a marvelous job of raising fundamental issues about deliberative democracy in general and Deliberative Polling (DP) in particular. They create a real deliberation about my proposals and I have learned from each of them.

Sanders: Beyond Polling Alone

Lynne Sanders wrote a famous article "Against Deliberation" some years ago. In that article she identified the most serious deficiencies of citizen deliberation, mostly drawing on the jury literature.¹ That article was an important challenge to our work on deliberative democracy and we have worked hard to see if we could answer it. In particular she focused on the various distortions from inequality and domination by the more privileged. I was gratified to see her conclude that the results for the DP have been "quite reassuring."

In her contribution here, Sanders raises a number of new and interesting issues. She foregrounds a role that is often left in the background, that of the moderator: "What is the analogue in our politics for the role of the moderator? Is it the head of the jury, or the judge who instructs the jury? Is it the Chair of a committee in a legislature? What springs most immediately to mind are examples from schools, not politics ..." She concludes, "We do not have an American democratic theory of group discussion moderator." While I believe she is right that we do not have such a theory, we do have a useful analogue in the focus group moderator. I was amazed when Bruce Ackerman and I did research for Deliberation Day to learn the sheer scale of the focus group industry. Billions are spent every year on group discussion. Of course the aim is different than that of the DP. The aim is to uncover the most effective basis for advertising and persuasion. But the method of leading impartial discussions in which all participate in a safe public space without betraying any hint of one's own views requires a set of skills that can be easily developed and that are already widely shared. We have, in fact, found that the best DP moderators are focus group leaders, who undergo a slight retraining for the special tasks in the DP.

The book charts a journey from the founders' ideal of deliberation applied to representatives to the modern persuasion industry's practice of molding public opinion. It is a journey "from Madison to Madison Avenue." If the repertoire of skills which group discussion moderators cultivate for Madison Avenue could be returned to the service of deliberation, that would be the completion of a virtuous circle.

Sanders notes that I embrace the position that random sampling is a democracy enhancing device (in that it embodies political equality in the sense of an equal chance to have one's views counted). But she also queries the fact that it seems to serve in my current analysis as a "democracy limiting device." She is correct but this is not a contradiction. It is enhancing for at least one key democratic value and limiting for at least another. It is enhancing for political equality and limiting for mass participation. Precisely because only those drawn into the sample can participate, it counts views equally by giving everyone, in theory, an equal chance of being selected and of having one's views counted. Random samples are immune to the notorious difficulties with SLOPs (self selected listener opinion polls in the terminology of Norman Bradburn). Groups cannot mobilize to capture random samples by voting over and over and creating an impression of public opinion about candidates or policies. But precisely because groups cannot mobilize to capture the membership in a scientific sample, the participation of those not invited is barred. So the equal counting requires a limit on the value of mass participation.

The democratic theory portion of the book focuses on the tradeoffs between competing democratic values. If one takes the three values that are internal to the design of democratic processes—political equality, participation and deliberation, I argue that there is normally a trilemma in attempts to realize all...