1
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2022
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Brennan, Jason, 1979- author. | Landemore, Hé lène, 1976- author.
Title: Debating democracy : do we need more or less? / Jason Brennan and Hé lène Landemore
Description: New York : Oxford University Press, 2022. |
Series: Debating ethics | Includes bibliographical references and index.
Identifiers: LCCN 2021023444 (print) | LCCN 2021023445 (ebook) | ISBN 9780197540817 (hardback) | ISBN 9780197540824 (paperback) | ISBN 9780197540848 (epub)
Subjects: LCSH: Democracy—Philosophy. | Democracy—Moral and ethical aspects.
Classification: LCC JC423 .B7835 2021 (print) | LCC JC423 (ebook) | DDC 321.8—dc23
LC record available at https://lccn.loc.gov/2021023444
LC ebook record available at https://lccn.loc.gov/2021023445
DOI: 10.1093/oso/9780197540817.001.0001
1 3 5 7 9 8 6 4 2
Paperback printed by Marquis, Canada
Hardback printed by Bridgeport National Bindery, Inc., United States of America
8. If Democracy Is Such a Smart Regime, Why Are Democracies Doing So Poorly at the Moment and How Can We Fix Them?
PART 3 RESPONSES
9. Brennan: Response to Landemore
10. Landemore: Response to Brennan
Introduction: How to Fix
What Ails Democracy?
WHEN WE WROTE THE FIRST draft of these sentences, pro- democracy protesters in Hong Kong had rallied for six straight months. They demanded safeguards for Hong Kong’s political autonomy and universal suffrage for the election of the devolved legislature and chief executive. When we wrote the second draft of this paragraph, it appeared their pro- democratic movement had been crushed.
Democracy is in retreat in a number of countries around the world. In the past decade, Russia, Turkey, and Venezuela transitioned from flawed democracies into authoritarian regimes. Even established democracies experience turmoil. Brexit, the election of Trump, the success of the Five Star Movement in Italy, the prolonged protests against Macron in France, and the rise of extreme parties show that even stable Western democracies are suffering. Is democracy itself, as a normative ideal, in crisis? What other regime could possibly look superior? Regardless, democracy remains the gold standard form of government. Most people regard democracy as the best and most legitimate form of government. Even dictators
claim to speak for and derive their authority from the people; they go to great pains to imitate democracy including through sham elections.
We agree that the best places to live today are generally well-functioning democracies, rather than sham or partial democracies, one- party states, or something else. We also agree that real-life democracies suffer from significant flaws. We’ll debate what we should do about it. Can we, as Landemore argues, fix the problems of democracy with more democracy? Or, as Brennan argues, should we explore alternatives to democracy?
WHAT DO WE MEAN BY “DEMOCRACY”?
We regard a system as democratic to the extent that all members of society share equal fundamental political power. That is an intentionally broad definition. Some regard this definition as too broad—we will consider objections based on this worry later.
This definition includes direct democracies where citizens make the law and indirect or representative democracies where democratic representatives make the law. It includes systems in which majority rule is constrained and unconstrained. We’ll start from the common conception of democracy as characterized by universal suffrage, periodic elections, and the peaceful alternation of parties in power, even as we will consider alternative understandings of this regime form (including regimes where political offices are staffed by lottery rather than elections).
American high school students taking Advanced Placement US civics are trained to say, “America is a
republic, not a democracy!” This objection insists on using these words in an obsolete way yet it is onto something about the real nature of the American regime. In the late 1700s, political theorists used the word “republic” to refer to representative governments with constitutional limits on their power; they used “democracy” as a pejorative label for mob rule. Today, political scientists and philosophers use the word “democracy” as a broader category to refer to both limited and unlimited representative government with some degree of popular sovereignty. Whether those are real democracies, however, can still be questioned.1
Indeed, in practice and by law, no system is perfectly democratic so described. No actual democracy in fact distributes power equally among all members of society. For one thing, not every member of society qualifies as part of the voting public or as a legal citizen. Nearly every democratic government restricts foreign residents from voting. Every country restricts children from voting, usually on the grounds that they do not yet have competent or autonomous political judgment. Many disenfranchise felons or citizens living abroad.
Further, people have unequal ability to run for office. Bernard Manin argued long ago that elections treat citizens unequally as candidates, distributing power to those
1. In the same way, in the late 1700s, philosophers used the word “experiment” to refer not to what we would call experiments today (carefully controlled artificial interventions) but to what we would now simply call “observations.” Saying today that the US is a republic, not a democracy, is thus like saying that what physicists do are not really experiments, but something else. Words change meaning over time.
with extraordinary, socially salient qualities. In practice, elections tend to empower social and economic elites. 2 In the US, a Kennedy, Clinton, or Bush is far more likely to win office than an average Joe. In most democracies, the rich have more influence de facto than the poor. 3 Majority ethnic groups usually have more effective power and influence than minorities. Handsome candidates tend to beat ugly ones.4 In the UK, an Oxford grad is more likely to become prime minister than an East Anglia grad. In France, National School of Administration grads are more likely to become president than graduates of regional noname universities.
WHY NOT
DEMOCRACY?
We will debate how best to fix democracy, but we also agree that certain arguments for democracy are weak. Some people say what justifies democracy is fairness: it gives everyone an equal say.
There is an intuitive appeal to this argument. People have fought and died for democracy for this reason. However, it faces serious philosophical problems. In fact, in real-life democracies, persistent ethnic or ideological minorities often have no chance of getting their way. They will always be outvoted. In other words, equal say does not mean equality of influence.
2. Manin 1995.
3. Gilens 2012. See also Cagé 2020.
4. Lens and Chappell 2008.
If you genuinely and only wanted to equalize influence, you would not hold elections, which both introduce various kinds of class and racial biases and guarantee unpopular views have no shot. Instead, you would choose decisionmakers by lottery, as the ancient Greeks did. For instance, we could pick the president at random from all citizens between the ages 35 and 65. Many think this a bad way to pick a president, but it is perfectly fair: everyone, regardless of class, race, or sex, would have an equal (albeit infinitesimal) chance of being chosen. Or, we could let people vote for candidates, but then assign percentages to each candidate based on the percentage of votes she receives, and, finally, pick winners at random using these weighted percentages. Either system would be far fairer. Whether they would be more desirable is the harder question.
Relatedly, some people say that democracy is just because it treats individuals equally. Giving everyone an equal vote means they are all equals in the eyes of the law. However, giving everyone an equal vote is but one of many ways we could enshrine and publicly express the ideal of equality. Many liberals of various sorts think the more important way to enshrine equality is to ensure that every person has an extensive and equal set of rights protected by law.
Again, giving everyone an equal vote could undermine other forms of equality. For instance, imagine all white people are virulent racists. Suppose the electorate is 90% white and 10% black. Imagine there are no constitutional restrictions on what laws the society can pass, except that it must remain democratic and everyone must have equal voting rights. Here, the white majority might outvote the black minority and use the law to oppress them, even
though each individual white person has the same power as each individual black person.
A closely related argument says that democracy empowers us as individuals. But that seems false, at least in most modern election-based representative democracies. In most elections, how we vote matters a great deal, but how any one of us votes matters little. Note that the whole point of democracy is to empower groups, not individuals. If how you individually voted regularly made much of a difference, then something would be wrong. The probability that an individual vote will make a difference in a typical election is very low.
DEMOCRACY IS GOOD BECAUSE/ IF IT WORKS
Instead of those arguments, we think that democracy is good primarily because it works.
Consider, by analogy, a hammer. We value hammers not as ends in themselves, but because they do a job. We would not insist on using an inferior hammer over a better hammer. We would not insist on using a hammer when we need a wrench.
For the purposes of this book, this is how we’ll evaluate democracy. We put aside questions about whether democracy is an end in itself or valuable for what it expresses. Instead, we debate whether democracy is the best tool for the job.
Let’s introduce a technical term. “Instrumentalism” refers to a set of views about what justifies having one form of government over another. Instrumentalists hold:
1. There are procedure-independent right answers to at least some political questions. To illustrate, think of a criminal jury trying to assess whether someone is guilty. There is an independent fact that they are trying to discover. This is in contrast with, say, a game of dice, where the “right” answer is procedurally defined: whoever rolls the highest number wins.
2. What justifies the distribution of power, or a particular way of making political decisions, is (at least in part) that this distribution or method tends to get the right answer.
To continue with the jury example, we can ask whether other trial decisions are more reliable, overall. Which procedure is most likely to get us to the truth of the matter about whether the accused committed the crime? Perhaps a tribunal of judges or an AI bot would be superior.
We might also worry that some systems have too high a chance of selecting horrible answers. Aristotle argued that the main virtue of republican forms of government is not that they tend to pick the best policies, but that they rarely pick very disastrous policies. In contrast, he thought monarchies were more likely to pick excellent policies, but also more likely to pick disastrous policies. He thought republicanism was best because it minimized the risk of very bad outcomes, even at the expense of a reduced chance of exceptionally good outcomes.
We both agree that we should judge democracy primarily by how well it serves the ends of the polity, whether defined as justice, stability, efficiency, or some other value. Or, more abstractly, by how well it works.
We reject a strange moral relativism that certain democrats accept. In Plato’s Euthyphro, Socrates asks Euthyphro whether things are good because the gods say so or whether the gods say those things are good because they are good. There must be some underlying reason why some things are good and others evil; they are not good or evil simply because Zeus decided they are on a whim. Similarly, we agree that what makes many things good or bad in a democracy is that they are good or bad for independent reasons, not because a democracy judges them to be so. If 95% of people vote to do something evil, awful, or stupid, it remains evil, awful, and stupid.
We both think democracy should be judged primarily by its results. We reject the view that people can always just decide, by fiat, what counts as good results. But then, you should ask, what counts as good results? Here, we will try to rest our cases on the broad values and goals that people across the world tend to share, such as human rights protection, economic prosperity, peace, security, basic dignity, individual freedom, education, health, and general individual well- being.
How to weigh all these values when they conflict is a hard question. The good news is that we doubt we’ll need to do that to make our arguments go through. Instead, we’ll discuss certain features of democracy, aggregative voting, deliberation, voter psychology, political incentives, and so on, and explain why we have independent reasons to think these phenomena are likely to track the truth about the common good of a given polity and lead to good decisions, or not.
In the same way, a philosopher of science— or your middle school science teacher— can explain to you why using scientific methods are generally reliable and trustworthy ways to track the truth, without having to first tell you what all of the truths of science are. We can explain to you why wishful thinking, in contrast, is not a reliable guide to truth in mathematics without first having to discover all the truths of math. In this book, for the most part, we can debate whether certain political decision-making methods are reliable or not based upon outside evidence, but without having to pin our arguments on our particular political views. We will occasionally use some specific examples where we think democracy got the right or wrong answer, but our arguments will rarely hinge on these specific examples.
Of course, when we ask whether democracy is reliable, we have to ask, compared to what? Our answer: compared to the feasible alternatives; that is, compared to other forms of government that could be implemented given the constraints of human nature. In this book, we’re not interested in asking what the ideal form of government would be, if only everyone were a moral saint who always did the right thing. We want to know which political regime will best tend to make good decisions and produce good outcomes, given that people’s willingness and ability to comply with institutions are imperfect, given that people are sometimes incompetent and corrupt, and given that institutions are not guaranteed to work as intended.
BRENNAN: DEMOCRACY WITHOUT ROMANCE
In the opening chapter, Brennan will argue the reasons most people accept democracy are nearly irrelevant to realworld democracies. Their mental model of democracy does not match how democracies in fact perform or could be made to perform under realistic circumstances. They think citizens form their political affiliations on the basis of their background beliefs and values. When citizens vote, they support politicians who will advance their favored ideas. In the end, democracies deliver, if not the will of the people, at least a compromise position among their separate wills. In contrast, he’ll argue, the empirical work shows that most citizens lack any stable ideology or political beliefs, and their political affiliations are largely arbitrary. Their votes do not communicate their genuine support for different policies or values. Citizens are ignorant, misinformed, and tribalistic despite lacking firm beliefs. As a result, the more power we give them, the more we suffer the consequences. Whatever we say about democracy, we need to be realistic about how people behave.
In chapter 2, Brennan will note that most philosophers try to “solve” these problems by arguing we need more and better democracy. In particular, they think certain kinds of democratic systems could unleash the hidden “wisdom of the crowds.” However, he’ll argue, they either rely upon mistaken applications of certain mathematical theorems, or they end up retreating toward unrealistic ideals of how people ought to behave. In effect, they say that democracy would be wonderful if only people behaved the right way.
But, he’ll argue, there is no realistic mechanism to get them to behave the right way.
In chapter 3, Brennan will argue that part of the solution is to reduce the sphere of politics and also the sphere of political control. Certain issues, such as trade policy, immigration policy, central banking interest rates, who serves as district attorney or judge, and various kinds of regulation, should be kept out of citizens’ hands, for everybody’s own good.
In chapter 4, Brennan will argue that we should be open to experimenting with certain non- democratic forms of government. In particular, he’ll argue that Enlightened Preference Voting is likely to be superior to our current system. In Enlightened Preference Voting, all citizens may vote. When they vote, they (1) register their preferences, (2) indicate their demographic categories, and (3) take a short test of basic, easily verifiable political knowledge. Afterward, all three sets of “data” are anonymized and made public. The government—and any decent political scientist or newspaper— can then calculate what a demographically identical public would have supported if only everyone got a perfect score on the test. He’ll argue that while such a system would be subject to special interest rent- seeking, it is nevertheless likely to be superior to any realistic democracy as we find it.
LANDEMORE: LET’S TRY REAL DEMOCRACY
In her first chapter, chapter 5, Landemore will put forward a general theoretical case for the benefits of distributing
political decision- power in an inclusive and egalitarian manner and for locating the legitimacy of laws and policies in the deliberations of the people or their democratic representatives. The core idea is that many minds are better than few to deal with the uncertainty and complexity of the world and figuring out solutions that work for all in it. This argument builds on the formal properties of two key decision-making mechanisms of democracy, namely inclusive deliberation on equal grounds and majority rule with universal suffrage. Properly used in sequence and under the right conditions, these two mechanisms ensure that no information and viewpoint is ignored and maximize the cognitive diversity brought to bear on collective political problems and predictions. Building on existing formal results by Lu Hong and Scott Page, the chapter introduces the “Numbers Trump Ability” theorem, which formalizes the intuition that many minds are smarter than just a few. Under the right conditions systems governed by democratic decision- procedures can be expected to deliver greater epistemic performance than less inclusive and egalitarian systems.
In chapter 6, Landemore addresses theoretical and empirical objections to the epistemic argument for democracy presented in the previous chapter (the argument from collective wisdom). The objections this chapter addresses include those based on the average voter’s alleged incompetence and systematic biases, as well as those that challenge the relevance of deductive arguments for democracy. The main counterpoint is that focusing on individual input into the democratic process to infer the quality of democratic outcomes (the model “garbage in, garbage out” used by Brennan) is misguided. It fails to consider that collective
intelligence is an emergent property that crucially depends on group properties not captured by measures of individual input. Systematic biases would be, and often are, a problem for democracy but no more than for oligarchies of knowers. In a free and diverse public sphere the public and its democratic representatives have more opportunities to debias themselves, at least over time, than small groups of homogenously thinking elites.
In chapter 7, Landemore argues against both oligarchic and majoritarian rule by knowers, or epistocracies. Such regimes are necessarily blind to a number of interests and perspectives, rendering them epistemically inferior to fully inclusive democracies over the long term. The chapter first consider the classic defense of Chinese-style epistocracy by Daniel Bell and then turns to the more puzzling rule by the knowledgeable 95% defended by Jason Brennan. While Bell’s Chinese model is much more vulnerable to epistemic failure due to the blindspots it structurally builds in its decision- process, even Brennan’s majoritarian epistocracy takes the unjustifiable epistemic risk of silencing what could be the most relevant voices on crucial issues.
In chapter 8, Landemore returns to the ideal of people’s power and argues that democracies as we know them are dubiously democratic. Most ordinary citizens in existing representative democracies have little deliberative input into the laws and policies that rule their lives. The chapter traces the problem to fundamental design mistakes made in the eighteenth century when elections, an oligarchic selection mechanism, rather than the traditional lot of Classical Athens, were privileged as the method for choosing representatives. This chapter also makes the case for open democracy, a new paradigm of democracy inspired by
a number of real-life experiments in participatory democracy and in which the center of power is accessible to all citizens on an equal basis. Central to this new paradigm are new forms of democratic representation that are truly inclusive and egalitarian, such as open mini-publics connected to the larger population via crowdsourcing platforms and moments of mass voting.
