Epistocracy: a political theorist’s case for letting only the informed vote
A political theorist’s provocative idea for how to fix democracy.
In 2016, Georgetown University political philosopher Jason Brennan published a controversial book, Against Democracy. He argued that democracy is overrated — that it isn’t necessarily more just than other forms of government, and that it doesn’t empower citizens or create more equitable outcomes.
According to Brennan, we’d be better off if we replaced democracy with a form of government known as “epistocracy.” Epistocracy is a system in which the votes of people who can prove their political knowledge count more than the votes of people who can’t. In other words, it’s a system that privileges the most politically informed citizens.
Brennan’s proposal sounds like a grandiose troll, but it’s not. His book is a serious critique of the moral and structural foundations of democracy.
The book got a bit of attention from the usual suspects when it first came out, but it didn’t go much further than that, and I confess I completely missed it at the time. I have strong objections to Brennan’s proposal, but his argument is interesting enough to justify a discussion. So I reached out to him a few weeks ago and asked him to make the case for “epistocracy.”
A lightly edited transcript of our conversation follows.
Why is an “epistocracy” preferable to a democracy?
We know that an unfortunate side effect of democracy is that it incentivizes citizens to be ignorant, irrational, tribalistic, and to not use their votes in very serious ways. So this is an attempt to correct for that pathology while keeping what’s good about a democratic system.
We have to ask ourselves what we think government is actually for. Some people think it has the value a painting has, which is to say that it’s symbolic. In that view, you might think, “We should have democracy because it’s a way of civilizing and expressing the idea that all of us have equal value.”
There’s another way of looking at government, which is that it’s a tool, like a hammer, and the purpose of politics is to generate just and good outcomes, to generate efficiency and stability, and to avoid mistreating people. So if you think government is for that purpose, and I do, then you have to wonder if we should pick the form of government that best delivers the goods, whatever that might be.
There’s a lot to unspool there, and I’m with you so far — we should care most about outcomes. But first, let’s clarify a key point on which your argument rests. You seem to believe that voting is a form of power. When citizens vote, they’re exercising power over others, and if they wield that power arbitrarily or incompetently, they’ve negated their right to vote. Is that a fair characterization?
Yes, I call this the “competence principle.” The idea is that anyone or any deliberative body that exercises power over anyone else has an obligation to use that power in good faith, and has the obligation to use that power competently. If they’re not going to use it in good faith, and they’re not going to use it competently, that’s a claim against them having any kind of authority or any kind of legitimacy.
I’ll circle back to the competence principle in a second. Tell me why we should expect the citizens you exclude from the democratic process to submit to rule by epistocrats. Do you not expect resistance to such a proposal?
Two points here. First, there’s this idea that only democracy can be seen as legitimate, and that if you have a nondemocratic system, people will rebel against it. I’m not that worried about that. When you read studies about conformity, or studies about deference to authority, or about how people in nondemocratic countries perceive their governments, you find that people tend to think whatever system they have is legitimate.
Russia has a very corrupt system, and yet people there have surprisingly positive views of their government. People in China tend to view their government as legitimate, even when they’re being surveyed outside of China and they’re not going to get caught or hurt if they say something negative about their government.
So, if anything, I think what makes people think their government is legitimate and authoritative is simply that they’re used to it.
That said, not every form of epistocracy involves excluding people. You could, for example, have a system where you only get to vote if you pass a test, and that’s probably the worst way to do it. But there are other ways to do epistocracy that don’t involve this kind of exclusion.
Are there any examples of epistocracies, today or previously, that you can point to as successful models?
Technically speaking, no. There was a time in British history (from around 1600 until 1950) where people who had college degrees would get an extra vote, but this was a stupid idea. It pains me to say this as an educator, but it turns out that university education has very little impact on how much people know. In general, college-educated people know more than non-college-educated people, but it’s not the college that’s making the difference.
People love to celebrate ancient Athens as a wonderful example of direct democracy, but they’re really talking about a form of epistocracy. That’s because only a very small number of people were actually voting, and they were the most educated members of society — the people who had the most political knowledge and the time to spend working on politics.
When you look at Singapore, I would call that more of a technocracy than an epistocracy, but you do have a system that is being run by elites for the common good.
Is there any fair way to determine who is and isn’t competent? Whoever defines the criteria has an immense amount of power in society, and the potential for abuse seems almost unavoidable. Although I know you’re against voting tests, I’m thinking here of racist literacy tests and poll taxes used in the Jim Crow South to keep black people from voting. Do you not worry about this kind of abuse?
Yeah, I do. Every kind of political system is abused, and we should guard against that. Here’s what I propose we do: Everyone can vote, even children. No one gets excluded. But when you vote, you do three things.
First, you tell us what you want. You cast your vote for a politician, or for a party, or you take a position on a referendum, whatever it might be. Second, you tell us who you are. We get your demographic information, which is anonymously coded, because that stuff affects how you vote and what you support.
And the third thing you do is take a quiz of very basic political knowledge. When we have those three bits of information, we can then statistically estimate what the public would have wanted if it was fully informed.
Under this system, it’s not really the case that you have more power than I do. We can’t really point to any individual and say you were excluded, or your vote counted for more. The idea is to gauge what the public would actually want if it had all the information it needed.
Okay, I’ve got a few issues with that, but let’s stick to the original question, which is who determines the criteria? Who decides what goes on that test?
People will try to manipulate that test for their own benefit. Republicans might want to make the test exclude certain groups; the Democrats will want to make the test exclude certain groups, or weigh certain issues.
So here’s my paradoxical-sounding idea: Let democracy decide what goes on the test. Randomly select, say, 500 citizens. Pay them a bunch of money and pass a law that says they can take time off from work without any kind of detriment to their career. Let them deliberate with one another, let them work together. They get to decide what’s going to go on the test. And then we use that test to weigh votes.
Why should we expect them to know the answers to this imagined test?
This sounds weird, but it’s really not. If you survey people and ask them what it takes to be an informed voter, they say the same kinds of things I would say, but you quickly find out that many of them don’t know the answers.
If I ask my 10-year-old son what he should look for in a spouse, he’d be surprisingly good at giving you a sensible answer about what makes for a good spouse. But no one thinks he’s competent right now to actually pick a spouse, or get married.
Voters know in the abstract what they ought to know; they just don’t actually know the things they think they should.
Let’s return to the “competence principle.” Why does the right to competent government trump other fundamental rights, like the right to participate in the democratic process?
I think the real question is why should we assume there’s a right to participate in democratic process? It’s actually quite weird and different from a lot of other rights we seem to have.
We have the right to choose our partner, to choose our religion, to choose what we’re going to eat, where we live, what job we’ll do, etc. While some of these things do impose costs on others, they’re primarily about carving out a sphere of autonomy for the individual, and about preventing other people from having control over you.
A right to participate in politics seems fundamentally different because it involves imposing your will upon other people. So I’m not sure that any of us should have that kind of right, at least not without any responsibilities.
But voting is not merely about imposing our will upon another. A lot of democratic theory holds that participation in the political process empowers the individual. Now, you claim this is wrong because the individual’s vote is meaningless and therefore voting is really about group empowerment. But isn’t it true that the individual is empowered when a group that shares their interests gains more political power? Isn’t the individual empowered through the group?
I would say that if a group that shares your interests takes power, then you will be empowered in the sense that your interests will be promoted. I’m not sure you’re empowered in the sense that you’re getting your way.
This is a weird metaphor, but imagine I couldn’t move my arms because they were tied. But then you were to give me coffee whenever I wanted it. That’s kind of what’s going on in a democracy: I’m still getting the coffee, and that’s awesome, but I’m not actually responsible for the coffee getting into my mouth — you’re the one that’s doing it.
No matter how you look at it, it’s really the group that has the power, not the individual, even if we’re in the group that has the power.
That’s a peculiar understanding of self-empowerment, but I don’t want to go down that theoretical rabbit hole. Part of your argument in the book is this idea that democratic politics undercuts social cooperation because it fuels identity-based conflicts. But I’d argue that social divides are byproducts of real and unavoidable differences in values and power. Could you just as easily argue that democracy provides a constructive means to channel these fundamental differences?
My worry is that people too often vote for basically arbitrary or historical reasons that have little to do with interests or ideologies. Certain identity groups get attached to certain parties and that’s just the way it is.
So I’m Boston Irish — that’s my identity. Because I’m Boston Irish, that predicts my loyalty to the New England Patriots and the Boston Red Sox. And it’s true — I’m a fan of both teams. It also predicts that I’m going to vote Democrat, even if I don’t know anything about the Democrats, or have no political beliefs.
And you find that, overwhelmingly, people with that identity will be assigned to the Democratic Party, even if they have no idea what the Democratic Party supports, and even if, when you ask them their opinion, it turns out their opinion more closely matches Republicans, or Libertarians, or Socialists, than it does Democrats. But they vote that way anyway.
So I tend to think that for the majority of people, your political affiliation is kind of like your sports team affiliation. And in the US, at least, sports team affiliations are not really that antagonistic. When I see a Yankees fan, I think, “Fuck the Yankees,” but I’m not looking to fight anyone. And I’ve been in Yankee Stadium wearing Red Sox gear and no one tries to beat me up. It’s just kind of a fun way to channel this divisiveness.
But in the political space, especially in the age of social media, we’re all engaged in constant grandstanding and the nastiness and division is ratcheted up all the time. Political division has gotten so dysfunctional and so ugly that it’s crippling to democracy.
Look, I’m sympathetic to much of what you’re saying, but let’s step back and try to get a little perspective. Democracy has always been a mess, and yet the democratic world has, over time, gotten more wealthy, more stable, and more tolerant. So democracy is self-evidently not a disaster. Why should we expect an epistocracy to produce a better outcome?
That’s a very good question. I like to say I’m a fan of democracy, and I’m also a fan of Iron Maiden, but I think Iron Maiden has quite a few albums that are terrible — and I think democracy is kind of like this. It’s great, it’s the best system we have so far, but we shouldn’t accept that it can’t be improved.
We might recognize that it’s better than anything else we’ve tried, and yet we can also see that there are all these persistent pathologies that exist, and so we should be asking, “How can we fix them?” We should be constantly experimenting and discovering what works and what doesn’t.
So epistocracy is just an idea, an attempt to do even better than we’re currently doing. There’s a lot at stake. We’ve eliminated a lot of problems. We have equal rights for LGBTQ people now; we treat African Americans better than we used to, though still much worse than we should. Women have more rights. We’ve reduced poverty. These are all good things.
But we’ve also bombed lots of countries and committed atrocities and engaged in all sorts of injustices at home and abroad. We can always do better.