The recent presidential election campaign reminded us of an often-neglected fact: libertarianism has both a populist and an elitist strand. The populist strand was represented (perhaps with a vengeance) by those libertarians who supported populist Donald Trump. A representative of the elitist wing would be H.L. Mencken, the early 20th-century journalist and critic, who denounced democracy as “the theory that the common people know what they want, and deserve to get it good and hard.” No doubt, Mencken would see Trump as an exemplum.

Jason Brennan’s Against Democracy comes from the same perspective as Mencken. It is a well-argued and challenging book.

Ruled by ignoramuses / The author, a libertarian philosopher and professor in the McDonough School of Business at Georgetown University, starts by analyzing the irrationality of voters, in the tradition of Geoffrey Brennan, Loren Lomasky, Ilya Somin, and Bryan Caplan. But Brennan’s critique goes further.

It is quite obvious that some people know more than others. And some know very little; they ignore simple, relevant facts and do not understand the general theories (in economics, political science, philosophy, etc.) necessary to interpret those facts. Anybody who reads this article must agree that we—my readers and I—know things that others do not know, and we also admit that others know things that we don’t. We have to start from there. Against Democracy helps.

Brennan cites much evidence of ignorance, cognitive biases, and groupthink among the general electorate. The typical American does not know which party controls Congress, has no idea of the broad distribution of the federal budget (for example, people typically think that foreign aid makes up 28% of the federal budget, while it is closer to 1%), doesn’t know that crime has decreased since the 1990s, and knows virtually nothing about the political parties’ platforms. Voters tend to cast their ballots for the better-looking candidates. More than a quarter of Americans don’t know which foreign state the colonists fought during the Revolutionary War. These are just a few examples.

Brennan divides the electorate into three categories. Most voters are “hobbits,” who know nothing about political issues and social science. The rest—including “many, perhaps most libertarians”—are mostly “hooligans,” people who have strong, irrational, and biased views. Finally, “vulcans” are the small number of voters who know the facts and theories and analyze policies rationally. Only vulcans “can explain contrary points of view in a way that people holding those views would find satisfactory”—what Caplan has called “the ideological Turing test.”

The mistakes that democratic electorates make are systematic; they are not random and don’t cancel each other. For example, “the U.S. public as a whole makes systematic mistakes about economics, including most of the mistakes [Adam] Smith warned us not to make back in 1776.”

The nature of democratic voting does not encourage voters to learn. This is because the individual voter will have an actual influence on the result only if there would otherwise have been a tie in the vote. In any large group of electors, the probability of a tie is infinitesimal. Depending on the assumptions made, this probability, and thus the probability of a single voter changing the result of the vote, is at most equal to the probability of winning the lottery jackpot. Have you ever met anyone who can say, “Without my vote, the other presidential candidate would have been elected”?

Thus, most voters spend little time gathering and analyzing information about the issues or candidates. Since the individual voter knows that his own vote does not count, it is rational for him to remain ignorant and to think irrationally. He can indulge in his biases or follow his political tribe. For the same reason, as evidence shows, the voter who has an opinion will not vote his interest but his opinion. For example, he tends to be nationalist because he mistakenly believes that this will further the common good.

The consequence is that an ignorant electorate elects ignoramuses to coercively rule over us. Even the voters who vote for what they think is the “common good” will often harm those they want to help.

Despite its deification, democracy is not the right to govern oneself, but the power to rule others. And it is against justice to be ruled by an incompetent government.

Here comes epistocracy / To solve this problem, Against Democracy argues for “epistocracy,” that is, “the rule of the knowledgeable” or, more exactly, the distribution of political power, including the right to vote, in their favor. (In ancient Greek, episteme means knowledge.) Just as physicians must demonstrate their knowledge and competence before they may practice medicine, citizens who want to vote should have to prove a minimum level of knowledge or analytical competence.

A “license to vote”—or to cast multiple votes—would be granted to citizens who successfully pass an exam showing they are sufficiently knowledgeable—for example, the civics test portion of the U.S. citizenship exam. Another means to this end would be to create an “epistocratic council” that would have the power to veto any democratic decision. Whatever the means, the principle is that there is no right to vote, except for the informed voters.

Brennan emphasizes that his choice of epistocracy over democracy is instrumental: better political decisions would be made this way. Individuals would get what they’d really want if they understood the facts and consequences of different policies.

The author of Against Democracy criticizes the philosophers who believe that democracy is necessary to symbolize and implement the equal dignity of individuals. He writes:

Democracy does not empower individuals. It disempowers individuals and instead empowers the majority of the moment. In a democracy, individual citizens are nearly powerless. (emphasis in original)

He believes that individuals, including minority individuals, would be better treated in an epistocratic regime. For example, “U.S. voters tend to be ignorant of the effects of the drug war on minorities … and how being ‘tough on crime’ tends to cause disproportionate harm to minorities.” Brennan also argues that democratic politics appeals to people’s baser instincts, “makes us worse,” and “tends to make us hate each other.”

His proposed regime is not as radical as it appears at first glance. It can be interpreted as democracy corrected by a knowledge requirement. The current democratic regime already includes epistocratic elements such as the courts. “Most epistocrats also want the rule of the many,” he writes, but of “the many-but-not-quite-everybody.”

He wants prudent, experimental, and gradual moves toward (more) epistocracy. He explains that if voters are incompetent to choose policies, they may be competent to establish general standards of political competence. “Whether we should prefer epistocracy to democracy is in part an empirical question, which I am not fully able to answer,” he writes in the last chapter of Against Democracy.

Epistocratic problems / There are many problems with epistocracy. I see three main ones.

The first is hubris. Even if we grant the easy point that “We the People,” taken in a collective sense, don’t know what we’re talking about, it does not follow that some elites know what is good for the rest of us. It may be true regarding some genuine public goods, but it is certainly not true with regard to most government interventions. Regarding personal tastes (is dark chocolate better than white chocolate?) and most personal goals, the Latin dictum espoused by economists applies: De gustibus non est disputandum (“There is no disputing about tastes”). This objection becomes moot if epistocracy produces a minimum state, but this is what remains to be proven.

Brennan claims that “the case for epistocracy does not rest on the authority tenet; it is based on something closer to an antiauthority tenet” (emphasis in original). He also argues that “as people … become more informed, they favor overall less government intervention and control of the economy.” But we are not sure that this is how an epistocratic system would actually develop.

The second problem is that epistocracy would further promote a society where the individual is trained to request permits and licenses. “The unequal distribution of power on the basis of competence seems elitist,” Brennan writes, “but it is not inherently more elitist than the unequal distribution of plumbing or hairdressing licenses.” There probably are better reasons to regulate the right to vote than the right to, say, repair a leaky toilet, but any license analogy plays into the hands of the regulatory state. Proponents of parenting licenses invoke the example of drivers’ licenses; indeed, raising children certainly has more social consequences than driving a clunker. We have to escape the licensing logic, not give it another lease on life.

It is one thing to make the easy argument that some people know things that others don’t. It is another to believe that, in this Brave New World, everybody has to be trained before he gains the right to do something. From this point of view, John Stuart Mill’s idea of giving additional votes to citizens with certain academic degrees is a more attractive idea than setting a formal system of state exams for political competence. But even this form of epistocracy is risky.

A third problem is that we have tried epistocracy and found it does not work as in Brennan’s model. From the Progressive era to our own days, academics, high-level bureaucrats, and pundits—in brief, the intelligentsia—have constituted an informal epistocracy, in cahoots with the state. Imagine what would have happened to the Second Amendment if the reigning intelligentsia had had still more power. Imagine what would happen to the First Amendment if the high priests of college political correctness were to form the government’s epistocratic council. Imagine the proponents of the parenting license among the top epistocrats of the future. If a mild level of epistocracy has not worked, it is not clear that more of the same is the solution.

Classical liberal solution / An alternative solution, which is the classical liberal solution, is to limit the scope of state action. If the state did little, the damage done by political ignorance would be limited, especially in a representative democracy, and more to the point in a republic based on the division of power. Brennan does not disagree with this, but he believes that more epistocracy would lead us to less government, which is a contentious claim.

Power to the knowers? No. But power to the ignorant neither. Some mild epistocratic elements could be added to the current political system. State propaganda should not actively encourage the ignorant to vote. (An ignorant nonvoter is more respectable than an ignorant voter.) If politically palatable, a poll tax would reduce the electoral participation of those who vote only to follow the crowd and entertain themselves. (Brennan hints at a similar measure.) The voting age could be raised.

Libertarian economist and political philosopher James Buchanan’s way of conceiving the foundations of classical liberalism may be useful here: “Each man counts for one, and that is that,” he wrote in The Limits of Liberty (University of Chicago Press, 1975). Is Brennan too quick to discard the symbolic value of democracy? What is the best form of government (if we need any) to protect individual liberty? On these deep issues, political philosophers disagree after 2,500 years of analysis and debate. This observation does not invalidate the necessary pursuit of truth, but it does cast further doubts on epistocracy.

Perhaps libertarians should be content to dwell somewhere between wild populism and undisguised epistocracy. This may be the true lesson of Against Democracy.