Fishkin and National Plebiscites

A few reactions to Democracy When the People Are Thinking (Author: James Fishkin)

In my new book, one of the seven detailed proposals for fixing polarization is called “National Plebiscites.” Congress would mandate bi-monthly, national events, each focused on one public affairs issue. (E.g. immigration, foreign policy, journalism, and so on.)

This particular proposal ventures into territory established by the iconic James Fishkin, whose “Deliberative Polling” for the last 25 years has aimed at rescuing democracy for the people. If you’re unfamiliar, key elements of the method include:

  • Polling a representative sample of citizens on a target issue, and inviting a sizable number of them to come in person to explore and discuss the issue.
  • Sending them briefing materials in advance.
  • Live, facilitated meetings: with experts, in small groups, in plenary meetings.
  • Explaining and observing rules of interaction.
  • Arranging publicity for the event, e.g. broadcast television.
  • At the end, measuring the changes in opinion which—in theory at least—is what the public would think if it were more engaged and informed.

I greatly enjoyed and admire Fishkin’s 2018 book, When the People Are Thinking: Revitalizing Our Politics Through Public Deliberation. (It may possibly have been his swan song, given that he is now 76.) It evinces a deep knowledge of democratic theory, and an impressive record of experimentation around the globe.

My main aims here are (1) to compare and contrast Fishkin’s approach with the one I proposed in my book, and (2) to articulate some doubts I have about Fishkin’s approach.


Deliberative Polling vs. National Plebiscites

Elements that the two approaches share include:

  • Recruiting (and compensating) a demographically-balanced “mini-public” of a few hundred participants to get their opinions on an issue
  • Providing them with briefing materials before polling
    • Notably, mine would be in the form of accessible, professionally-produced multimedia, and would require passing of an (easy) test to continue participating.
  • Secret-ballot voting
  • Results of the event are vigorously publicized

A key philosophical difference between the two is that Fishkin’s Deliberative Polling is aimed—as directly as possible—at affecting policy or exercising some control over government decision-making; in other words, that there should be substantial, formal consequences. (Fishkin would be the first to admit that, in practice, such a strong influence has been achieved only infrequently.)

My National Plebiscites, in contrast, do not have such an aspiration. Rather, mine are primarily aimed at educating the public, by creating a regular spectacle that audiences are as eager to see and discuss as they might a good reality-TV show.

Specific elements of my National Plebiscites that are different are:

  • While participants are preparing and voting, they do not interact with each other at all. This is primarily to make it scalable and to keep costs low enough for bi-monthly events.
  • Participants must also select and rank reasons for their votes. Similar kinds of information were collected at Fishkin’s events by various means, but in National Plebiscites it is a core and standardized part.
  • Massive social media marketing campaigns follow each Plebiscite, promoting both the results and the multimedia preparatory materials. Participants are supported in (voluntarily) putting their own reasoning online.

Stimulating widespread public discussion about the theory and the actual facts for each issue is the real purpose of the whole program. It attracts popular interest when participants put their identities on the line.

Choices and Reasons
 Conceptual mockup of sample plebiscite voting form, from Chapter 4.4 of my book.
After passing the test and indicating their choice,
compensated participants force-rank their selected reasons.


My Doubts about Fishkin’s Approach

Although every event orchestrated by Fishkin created value, and although I would support huge amounts of taxpayer funding for more of them, I nevertheless do have some doubts—some theoretical, some practical.

Non-Expert Judgment

What reasons do we have to trust that the actions recommended by the demographically-balanced mini public would actually be good for the country or the state? Obviously, the answer always is “it depends”: it depends on the type of the issue, it depends on the present national context, it depends on whether the issue is technical or is just about discovering “informed” preferences.

But in general I am skeptical. I am reminded too much of Walter Lippmann’s critique of a democracy that assumes all citizens are capable legislators. Government administration is highly evolved, complicated, and technical, and often requires an understanding of economics, law, history, finance, or political science. (To take just one nonrepresentative example: my wife works in banking regulation, a domain that has taken me a long time to understand indirectly; the thought of regular citizens weighing in on any of the issues there after a couple days of study is ludicrous.)

I was particularly surprised about deliberative events for constitutional amendments and for proposals for amending the California ballot process. Really? Wouldn’t knowledgeable constitutional experts and policy wonks and legal scholars be better designers of government processes?

I was disappointed at one point when Fishkin gave a simplistic view of experts:

Responsible experts and technocrats cannot make contested value decisions on hard choices without considering the values at stake. If they do not somehow take account of the people’s values, and value-laden priorities, they will need to substitute their own to make decisions.

Uh, not a false dilemma? Can’t experts equally well analyze and factor in all of the competing values and develop models that make the tradeoffs and effects clearer? Is it such a great mystery what values are in conflict, and are all “values” equally relevant to policy formulation and legislation? Without mini-public deliberations, do experts have no options but to substitute in their own personal ideological biases?

Obviously, the public’s desires and judgments matter. At the extreme, governments that have not satisfied the public’s wishes have been overthrown in chaos. But in the more stable and normal ranges, with modern, gigantic, mixed-economy, liberal democracies, how helpful are amateur mini-publics?

It points, incidentally, to uncomfortable, fundamental questions about the desirability of democracy. Perhaps rank-and-file citizens should just be satisfied with competent governmental decision making, and the regular opportunity to kick out incumbent figureheads. Or, should they, rather, have the right to run things themselves? (Whoops, another false dilemma.)

Certainly, the notion of “audience democracy” is unromantic. And, at minimum, we must admit that deliberative mini-publics could at least give citizens a stronger impression of self-governance.

Can This Catch On in the US?

I’m not sure how to articulate my second doubt without it sounding awkward. Here goes: This stuff is too boring. Too earnest. And it’s too laborious to explain to anyone why elections and polls and the Constitution and the Fourth Estate together are not “good enough.” There, I said it.

As a wonk who is really interested in this topic, I love it, and I can’t help but be impressed by all of Fishkin’s frameworks. Everything he says is accurate and sensible, and he brings pragmatism to his projects, never letting perfect be the enemy of the good. Yet, the arguments and language he brings to it (“deliberation,” “evidence-based”, “nomothetai”, “knowledge gain”, and so on) I think is something that tends to make even most journalists’ and politicians’ eyes glaze over.

So the difficulty of getting people interested in supporting this in a bigger way is a challenge. The proposal also prompts vague doubts about logistical matters, such as the costs. Bringing a mini-public together is not, I assume, terribly simple or cheap, and may be vulnerable to prejudices or funny business. And if you could, alternatively, direct three full-time policy analysts to accomplish something not the same but similar, how could anyone prove that Deliberative Polling is worth the hassle?

Maybe this challenge is partly due to the culture of the United States. Fishkin alludes to this possibility in a passage where he discussed the success of mini-public designs in authoritarian countries such as China:

The interest of such cases is precisely that, within a limited domain, our main criteria for the success of a given application of deliberative methods can be fully satisfied. Indeed, sometimes those criteria are satisfied better than in many cases in western party-competition-based democracies. In the latter, the sample recruitment may be more challenging and sometimes the results of the deliberation are treated as just a media story rather than a serious input to decision.

Perhaps it is related to the nature of competitive democracies and modern media systems:

One of the main challenges for competitive democracy is the obvious incentives it offers for political actors to attempt to manipulate. Our quest is for institutional designs that will incentivize deliberation, both by political actors and by the people themselves.

Will deliberative mini-public events cause any of the political actors’ manipulations to diminish? Will candidates who want to win elections have any incentive to change their behaviors, when the vast majority of their electorates are barely aware of such events? Will they reduce their reliance on partisan talking points? Will the media’s incentives change?

In summary, it seems like there are some fundamental kinds of headwinds. Personally, I have trouble picturing how support for these mini-public events can be ratcheted up.

Can Real Deliberation Happen at All Today?

My third doubt is not about Deliberative Polling per se, but rather is about the way the public discourse environment has been degrading in the past two decades. Including the way media audiences have fragmented into partisan camps, social media has lowered the general quality of discourse, and propagandists have been able to exploit all of it by flooding the zone with talking points. Public discourse is less and less about issues and more and more about culture war. Affective polarization is skyrocketing.

The results are that (1) mini-public participants are likely to be less open-minded, (2) the public is more likely to be suspicious of the events, and (3) debates about substantive issues are less likely to draw much interest.

The problem hit me hardest when reading a proposal by Fishkin for a national “Deliberation Day” just prior to US presidential elections, in which citizens are assigned to groups to debate public issues. A slew of objections flooded into my mind. Who are we kidding about this? No one will come to these with an open mind. Candidates do not have positions, just narratives. No one on the left will want to have to listen to MAGA Republicans, no conservatives will want to listen to the lefty sanctimony. No one will trust the pre-reads, no one will trust anything anyone has to say. Everyone will leave with a sour taste in their mouths, and few people will bother attending.

All right, maybe that is catastrophizing, but you get the point. Thanks to partisan media, social media, and disappearing newsrooms, the public discourse environment has changed in ways that make deliberation extremely difficult.


What We Really Need

My first doubt (about the citizens’ expertise) might only be a problem if Deliberative Polling were pushed too far, e.g. if their results were binding on the government. My second doubt isn’t a criticism of Deliberative Polling per se, but skepticism about its potential to get any traction in the US.

The third doubt, though, I think is a killer. Public discourse is not about issues anymore, and it’s not likely to be until we systematically work to “fix” the environment. (That is what the seven proposals in my book are about.) The temperature needs to be brought down.

So, I believe my National Plebiscites are better suited to these times. We need to find ways to make factual and rational points harder for the public to avoid. We need to cause the more egregious partisan talking points to lose their luster. And to thus shrink the spaces that partisan media can exploit.

A key to this is scalability, and that is why an important component of National Plebiscites is lavish spending on post-plebiscite promotion—especially on social media, where each event has the potential to reach wide swaths of the politically-aware public with factual language. And it has to have some entertainment value.

We all of course would like to see the voices of thinking citizens elevated. However, the more urgent need is to thwart propagandists, to teach the public about political arguments, and to reduce the widespread intellectual hubris that partisan media encourages. Learning facts that undercut your favored narrative—however unwelcome they may be—has a way of reducing that hubris.

There are many dire, festering issues that need legislative attention, as well as reforms to sensibly manage the public speech environment itself. The National Plebiscites program is an intermediate step that may enable us to start getting to some of them.

Leave a Reply

Your email address will not be published. Required fields are marked *