Debating Democracy is part of a book series that features a debate between two prominent authors on a key issue. The book consists of an initial, lengthy essay from each author articulating their view and then concludes with a shorter, rebuttal essay from each author in which they address some of the points of their interlocutor.
In this case prominent political theorists Jason Brennan and Hélène Landemore debate questions relating to knowledge and democracy, a subfield that is now commonly referred to as epistemic democracy. Put simply, the book is very engaging and both are forceful proponents of their perspectives. I have regularly referenced both of them in several books.
In terms of the debate, Brennan argues that ordinary citizens in a democracy like the United States are relatively ignorant and incompetent, whereas Landemore argues the opposite. Each of them have expressed their perspectives in more detail in other, longer works, such as Brennan’s Against Democracy and Landemore’s Democratic Reason.
To begin, Brennan runs through the standard litany of complaints against democracy. Namely, voters are deeply ignorant of policy specifics and they are not instrumentally rational in pursuing what they want politically. Brennan does not think that people are inherently irrational but rather because many people are politically ignorant, this leads to behavior that he sees as irrational.
Some thoughts: As I’ve argued before, the first point just isn’t that interesting. Why are Americans ignorant on many specifics? Does this matter? These questions are basically unanswered and we are left wondering, as Page and Shapiro do in The Rational Public, why should we care? For instance, being able to name the nine Supreme Court justices sounds like a trivia question. What do questions like this tell us? Do they indicate or correlate with anything important? As Landemore later says, these questions regarding the “inputs” into the political system don’t necessarily tell us much about how real world democracies operate.
The rationality part is harder to justify for Brennan. He provides a few selective samples but there is probably more information against his perspective. People do indeed turn up and vote for candidates, parties, and causes they believe in and care about. Abramowitz provides good evidence on this point in The Great Alignment. Brennan also frequently cites polls of Americans to indicate that voters are ignorant but we must distinguish between these two different groups, voters versus the broader American public. Voters in general are more educated, wealthier, and more informed than non-voters. And again, Brennan’s presentation of evidence is selective. There is much evidence that voters are instrumentally rational in their political behavior—one would not have to look too hard to write an updated version of Page and Shapiro's book on the rational political behavior of the public.
For instance, consider that voters will defy partisan perspectives to vote for specific issues that they believe in, such as when voters in heavily Republican states pass referenda raising the minimum wage or when voters in heavily Democratic states pass referenda limiting affirmative action. In each case voters defy elites from their own parties to pick out and pass specific policies that they believe in. This is also why Democrats are currently trying to use referenda to legalize abortion in red states. Because even in states where Republican elected officials are popular, severe abortion restrictions usually are not.
Brennan points to evidence from Caplan and many others that more educated voters have different opinions from less educated voters on many issues. I think this sounds like a good and obvious case for universal suffrage, so that all perspectives are represented. Landemore later makes a similar point. After all, I and most non-economists disagree with many of the predominant opinions of professional economists. We precisely don’t want only their opinions represented politically. We think they are wrong.
This takes us to the following point. All of Brennan’s proposed reforms here, in Against Democracy, and elsewhere, are designed to increase the political power of the most informed people (and hence, most educated, credentialed, highest income/wealth, status, and generally higher SES). But the empirical research, and a good amount of theoretical research as well, suggests that these people already have far more power than anyone else. The ultra rich, and secondarily the professional classes, basically run the political system and dominate the upper reaches of the economy. Why, precisely, must they be more empowered? Isn’t it possible, in fact, that their disproportionate power is part of the problem? Contra Brennan, I argue in Does Democracy Have a Future? that we already have a de facto version of Brennan’s desired polity. And it ain’t pretty. Landemore also makes a roughly similar point.
But what about the claim that voters care more about identity than policy? If voters act largely on group identity because of tribalistic motivations, why does this entail we should make America less democratic? Why would this entail, as Brennan suggests, that some people should not be able to vote? Katherine Cramer examines the role of identity in Wisconsin politics without coming to any elitist conclusions in The Politics of Resentment. Even Achen and Bartels’ book Democracy for Realists, which stresses the role of identity and offers an elitist perspective, does not suggest we should disenfranchise anyone just because the role of identity is larger than policy preference with regard to political behavior.
Brennan’s discussion of the Iceland model that Landemore focuses on is not very compelling. To only simplify slightly, he says that the Iceland example of direct democracy is not a good example because Icelandic voters did not pick the freedom of religion principle that he would want. He also assumes, without argument, that there are correct answers to all important political questions. But as I ask in my work, are there? Many questions are political, and precisely of concern for the entire public, because they don’t have independent, correct answers. This overlaps with ideas on democracy found in Wolin, Pateman, Barber, Young, Mouffe, Rancière, and elsewhere.
Taking a different but related tack, Estlund also challenges Brennan’s assumption. Maybe there are correct political answers but who can identify them? Who gets to decide? As he says, who made you boss? Why should I accept that Brennan, or other intellectual elites, or economic or political elites, have the “correct” answers to questions that concern fundamental issues of identity, values, purpose, and meaning? Sure, it takes years of study to become an expert on East Asian politics but how does this give one insight into the “correct” answer of whether we should go to war with North Korea?
Both authors discuss ways to answer these questions. One is to divide possible perspectives into two camps: epistemic democrats versus proceduralist democrats: This is too neat of a distinction, first off. But in simple form it means the following: Epistemic democrats defend democracy because majority rule tends to produce the correct, or best outcomes. Proceduralist democrats defend democracy as a decision-making procedure that, when properly enacted, by definition produces the best results. They appeal to no outside, independent standard of right or wrong.
Now, pure proceduralism can’t be correct. If a democratic assembly decides to commit genocide, or invade another country, this is wrong, regardless of whether they adhered to proper democratic procedure. (I suppose they could respond by saying that proper democratic procedures would never yield such a terrible outcome, but this seems a cheat. It defines away the problem by saying real democracy will never do anything bad and it also violates its proceduralist commitments by appealing to an outside standard of good or bad).
But this does not mean we should become pure epistemic democrats. Elite thinkers like Brennan, Caplan, Achen, and Bartels assume that most, or perhaps all, political questions have independent, objective, correct answers (or at the bare minimum, better or worse answers. But at many points they appeal to explicitly independent standards to judge if political decisions were correct, right, false, and so on). But they simply assume the existence of such standards. Why? What is the metaphysical status of such political standards? And how can we identify them? Who is capable of identifying them? As Estlund says, who made you boss? For instance, why should I accept Brennan’s account of “correct” political decisions? Or the broader professional classes’ definition of “correct” political decisions? Or the wealthiest one percent? There is no agreed upon answer for most political questions which is why we must turn to everyone to produce a collective answer, one that doesn’t exist before the demos in effect creates it. This is what it means to be a political question!
Rather than identifying with either the epistemic democrat or the procedural democrat, I would defend something in between the two. I see Estlund as doing something like this. Maybe 80% proceduralist, 20% epistemic. What would this look like? On some moral issues there are answers we can clearly identify (clear to identify because there is almost universal agreement). Genocide, invading another country, persecution of political dissidents, institutionalized racism, sexism, and other forms of bigotry. These are wrong. In many respects one could look at the various international agreements that defined the post-World War 2 era for a sense of which things are universally prohibited. So these are independent standards. A government, no matter how democratic or otherwise admirable, can never rightly do these things. As for everything else…
As Benjamin Barber argues, the realm of politics emerges precisely in those areas where we don’t have clear answers, where math, science, logic, philosophy, common sense, and so on do not yield widely shared answers. Politics emerges precisely when we have to come together and make decisions on issues for which we do not all agree. And on almost all these issues it is up to whatever that particular group of democratic citizens wants. These are questions of group identity, values, future direction, purpose, weighing of competing interests and perspectives, and so on, which don’t have identifiable objective standards or right answers, save for the small number of wrongs that are barred. Other than that, groups of people can do as they wish.
There is a similar analogy here to being an adult. There are some things I cannot do or be while continuing to be a decent person (i.e. I cannot be a serial killer or rapist). But there are an infinite number of lives I might choose to live while continuing to be an admirable person. That is precisely what it means to be an adult. I get to decide and in many respects the right answer for how to live my life is whatever life sounds best to me. As much as it may bother Brennan, this is what many democratic decisions are like.
So again, for Brennan et al, they see politics like Plato does, as a domain of knowledge that is not really distinct from chemistry, engineering, etc. It has correct, objective, knowable answers that can be ascertained through expertise. The problem, again, is that they don’t address or justify this assumption so much as assume it. But this framing does help to explain some of the differences in perspective and intuition between epistocrats (i.e. those like Brennan, who believe that the knowledgable should rule) and democrats (who believe that the people in their entirety should rule). Epistocrats see the political world as Plato does, democrats see it as Protagoras does (i.e. it is a domain of action and values, not specialized knowledge). As Landemore says, reflecting the democratic view, political leadership is about moral vision, not specialized knowledge.
And as Estlund argues, who is to identify the correct standards? Plato’s philosopher kings? Brennan? Political scientists? Harvard grads? Those who can pass a political knowledge quiz, as Brennan suggests? But this still leads to the follow-up question—why should we assume this subset of “enlightened” citizens will produce better results? And who will decide that their results are better? There are no universally shared standards to identify if the policies enacted by a participatory democracy, a representative democracy, a Brennan-style epistocracy, and a military dictatorship are better or worse, save for the limited set of abuses that are off-limits (and prohibited by various international agreements—the UN Charter, the Universal Declaration of Rights, Geneva Conventions, and so on spell out things such as torture, genocide, and aggressive war that are banned).
Some closing notes. Landemore operates with a good definition of politics: “the pursuit of the common good under conditions of uncertainty.” (page 143). With this in mind, let’s consider a few more points.
Landemore argues, and I agree, that Brennan’s empirical claims are less compelling when applied to real world democracies today. First, by Brennan’s own standards, democracies perform better than alternative regimes. Why is this? Given his belief in this claim, why is he skeptical of democracy? Brennan doesn’t offer a compelling answer to these questions, though he suggests that democracies of the world do better because they are not fully democratic and retain elitist elements. But this entails two follow-up questions: One, if this is so, shouldn’t more purely elitist regimes, such as Singapore (or perhaps China) perform even better? If not, why not? And second, if actual democracies such as the United States retain many elitist and oligarchic elements, isn’t it possible that their flaws and mistakes (whatever these may be) are also caused by the elites who dominate these societies, rather than the masses? But Brennan cannot dwell on this point for too long or it would undo the justification for his entire theory. If democracies are not run by the people then it does not really matter if the people are ignorant, for they do not have political power. Their ignorance would, in effect, have no efficacy. But to concede this would be to move more in the direction of Landemore and me—that problems in countries like the US largely result from elite dominance and might be resolved through more democracy.
Let me conclude by quoting Landemore at more length on the real world of politics, which is defined by uncertainty. “Uncertainty means that we cannot fully anticipate what the future will be like, whether in the short term or the long term, and whether, in particular, it will be sufficiently like the past (the way, say, business cycles may repeat familiar patterns) or different in radically new ways (as when a pandemic, a war, or a climate emergency suddenly complicate familiar political questions in unpredictable ways). As a result, no one person or group of persons can be trusted to have the relevant knowledge or skills that make them the superior decision-makers in every situation, and it becomes rational to follow instead a simple heuristic that distributes power over the entire group: give everyone the same participation rights in the decision-making process.” (162-163).
As she rightly argues, “by contrast, small oligarchies of “knowers” may be more prone to group think, especially if they don’t have to be accountable to a large, critical, and diverse society.” (175).
What this entails is a perspective on democracy, politics, and knowledge that is defined by humility and openness: “Political knowledge is a living thing and comes in various guises. We should be humble about who has it and about the possibility of defining or identifying it ex ante.” (p. 207).