Every year, I teach a course titled “Knowledge, Ignorance, and Democracy.” And every year, my students are struck by the same troubling fact: most people are shockingly ignorant about politics. We discuss an ocean of data showing that many voters don’t know even the most basic political facts, such as who their elected officials are, what their opponents believe, and which important laws or policies were passed in recent years. When confronted with this evidence, my students generally accept it. (So, too, do many political scientists, philosophers, and economists.)

Further, my students agree that a healthy democracy requires an informed public. Without knowledgeable citizens, elections become little more than guesswork, and misinformation spreads unchecked. They recognize that core democratic principles like accountability and representation lose their meaning if citizens are uninformed about the very system they’re meant to influence.

In other words, my students accept that many voters aren’t knowledgeable enough to vote responsibly. And once they see the problem, they agree that something must be done to fix it.

Yet, when the conversation shifts to potential solutions, something strange happens. If widespread political ignorance threatens democracy, one logical response is to ensure that more knowledgeable citizens have greater influence over political decisions. This is the basic idea behind epistocracy—a system in which political power is distributed in proportion to knowledge or competence. However, despite their deep concern about voter ignorance, my students are overwhelmingly unsympathetic to epistocracy. They reject the idea that we should privilege the voices of the more knowledgeable, not just on moral or democratic grounds, but for a more basic reason: they insist that we cannot reliably determine who counts as knowledgeable.

This has long puzzled me. On the one hand, my students are deeply troubled by the evidence of widespread political ignorance, which presupposes that we can meaningfully measure such ignorance. On the other hand, they are quick to dismiss epistocratic proposals—such as limiting voting rights to the more informed—on the basis that we cannot accurately identify who is knowledgeable. This contrast is striking. If we can determine who is politically ignorant, why can’t we determine who is politically knowledgeable? Aren’t these just two sides of the same coin?

Is Ignorance Is Easier to Spot Than Knowledge?

One explanation for this asymmetry is that we have clear benchmarks for detecting ignorance, whereas establishing the threshold for political competence is a deeply contested issue. For example, few would dispute that a voter who cannot name their elected representatives, misidentifies key policies, or holds wildly inaccurate beliefs about the positions of major political parties is politically uninformed. These errors are straightforward and easily measured through factual tests. But defining what it means to be sufficiently knowledgeable is far more controversial. Is it enough to be aware of major political issues and debates, or must voters also grasp the mechanics of policy implementation? Does competence require familiarity with opposing viewpoints, or just a coherent understanding of one’s own? Unlike ignorance, which can often be identified through clear failures, the threshold for adequate knowledge depends on contested judgments about what citizens should understand to participate responsibly in democracy.

An analogy may be helpful here. Consider that many people are skeptical about whether we can identify the “right” answers in politics. Yet, these same individuals often accept that some political decisions are clearly disastrous. For example, few would dispute that policies leading to famine, economic collapse, or genocide are catastrophic failures. These are outcomes that any reasonable person would agree are important to avoid. Thus, even if we cannot determine the single best course of action, we can still recognize decisions that are demonstrably worse than others. Similarly, while it’s difficult or controversial to determine what constitutes sufficient political knowledge, it’s easier to identify clear cases of political ignorance. It’s much less contentious to say that a voter who believes blatant falsehoods (e.g., that one political party supports policies it explicitly opposes) is politically uninformed than it is to specify exactly what level of knowledge qualifies someone as adequately informed.

This reply is reasonable as far as it goes, but it doesn’t go very far. Even if there is no consensus on what constitutes adequate political knowledge, that does not mean we cannot establish a reasonable threshold for minimal competence. While there may be debate about what it takes to be well-informed, we can likely agree that certain forms of ignorance are so extreme that they undermine responsible democratic participation. (After all, this is precisely why my students worry about voter ignorance in the first place.) A voter who cannot identify key policy issues in an election isn’t just missing deeper political insights—they are fundamentally ignorant. If we have reliable ways to measure such ignorance, why not use them to set a baseline for political competence?

Consider the idea of a voter qualification exam. While many oppose such exams because we lack clear benchmarks for political competence, this doesn’t mean we cannot establish a minimal threshold to exclude only the most politically uninformed. As Jason Brennan suggests, a voter competence exam need not be overly demanding. It could be designed to exclude only the least informed 10% of the population—those who demonstrably lack even a rudimentary grasp of political reality. By setting the threshold low, such an exam would preserve broad democratic participation while preventing the most uninformed individuals from exercising political influence. We could use our ability to measure ignorance as a tool for ensuring a minimally competent electorate, rather than letting the difficulty of defining sufficient knowledge prevent us from addressing clear cases of political incompetence.

At this point, one might object that there is no clear consensus on what counts as minimally sufficient political knowledge. But if that’s the case, why is there overwhelming agreement that political ignorance is a serious problem? The widespread concern over uninformed voters suggests that we already accept at least some objective standards for political competence—otherwise, diagnosing ignorance with such confidence would be incoherent. If we can readily recognize that many voters lack even the most basic understanding of politics, then it’s difficult to see why we cannot apply that same recognition to establish a minimal threshold for participation.

Moreover, we already exclude certain groups from voting based on a presumed lack of competence. Babies, young children, and in some cases, individuals with severe cognitive impairments are not granted voting rights because they are widely assumed to lack the necessary understanding of political decision-making. This exclusion is rarely controversial, and yet it reflects a competence standard: we recognize that some individuals are so uninformed that their participation would not contribute meaningfully to democratic decision-making.

If we accept that infants and young children lack the competence to vote responsibly, then we have already conceded that some threshold for political knowledge exists. The question, then, is not whether we can draw such a line, but rather where that line should be drawn. If the principle behind age-based voting restrictions is sound—that a certain level of competence is required for meaningful participation—why should we reject the possibility of applying a similar standard to exclude only the most politically ignorant adults?

Is Political Ignorance More Visible in the Aggregate?

What else might explain this puzzling asymmetry? Why do so many people readily accept that political ignorance is real and measurable, yet resist the idea that we can identify those who are minimally politically knowledgeable?

Perhaps people don’t actually believe we can accurately assess who is politically ignorant on an individual level, but they do think we can make broad judgments about the overall ignorance level in the electorate. When surveys reveal that large portions of the population fail basic political knowledge tests, this is taken as evidence of widespread ignorance—but not necessarily as a means of identifying which specific individuals are uninformed. In other words, the claim is not, “We can tell exactly who is too ignorant to vote,” but rather, “There is a lot of ignorance in the electorate.”

This distinction may help explain the asymmetry. A person might readily accept that voter ignorance is a problem in the abstract while hesitating to endorse any mechanism that requires assessing individuals. However, this response assumes a sharper divide between identifying general ignorance and identifying individual ignorance than actually exists. The very surveys that reveal widespread political ignorance rely on individual-level assessments—respondents answer specific political knowledge questions, and their responses are aggregated to determine the overall level of ignorance in the electorate. If these instruments are reliable enough to diagnose ignorance at a societal level, why should they suddenly become unreliable when applied to individuals in a non-anonymous way?

If anything, the only difference is how we choose to use this information. When ignorance is measured collectively, it’s treated as a general concern about democratic health. But when these same measures apply to individuals, the results carry weightier implications, especially if used to determine voting eligibility. The resistance to identifying who possesses adequate knowledge may therefore stem less from the reliability of these assessments and more from their ethical and political consequences. I’ll consider this idea next.

When Ignorance Matters, We’re More Skeptical

In philosophy, an increasingly popular idea is that the ethical stakes of a belief can influence how much evidence you need to justifiably hold that belief. This view, known as moral encroachment, challenges the traditional assumption that knowledge and justification are “purely epistemic” matters, that is, independent of moral considerations. Instead, it holds that when a belief has significant ethical or practical consequences (e.g., when it affects people’s rights, dignity, or social standing), the standards for holding that belief justifiably become higher.

Applied to the case of political ignorance, this suggests that the bar for confidently identifying ignorance depends on what is at stake. When the question is merely an abstract empirical inquiry—Are citizens adequately politically informed?—the standard for judging them ignorant is relatively low. Public opinion surveys and knowledge tests routinely reveal widespread political ignorance, and such findings are widely accepted without too much controversy. However, when the question shifts from mere description to prescription—Should politically ignorant individuals be disenfranchised?—the standards for making that judgment are suddenly much higher.

This heightened standard is not arbitrary; it reflects the moral risks involved in implementing a voter competence threshold. If we misjudge ignorance at an individual level, we risk unfairly stripping people of their political rights, further entrenching inequalities, and disproportionately excluding marginalized groups. Moreover, a voter competence test would inevitably involve ranking individuals in ways that can be seen as degrading or offensive—explicitly labeling some as less qualified to participate in democratic decision-making than others. Given these stakes, moral encroachment suggests that we should demand much stronger justification before concluding that someone is too ignorant to participate in democracy. In other words, the more consequential the belief, the greater the epistemic burden for holding it.

If this hypothesis is correct, it helps explain why people readily accept evidence of widespread political ignorance but resist the idea that we can design tests to determine who is politically knowledgeable. This is the puzzle we started with. The issue isn’t that ignorance is easier to identify than knowledge; rather, the consequences of error are far more severe when deciding who is knowledgeable enough to vote responsibly. Declaring the public generally ignorant has relatively diffuse consequences—it may fuel debates about civic education or institutional reforms, but it does not directly strip anyone of their rights. By contrast, wrongly labeling specific individuals or groups as politically ignorant and excluding them from voting carries much greater risks. It threatens to violate individual rights, deepen systemic inequalities, and erode the democratic principle that all citizens should have an equal say in governance. When the moral stakes are low, we are comfortable diagnosing ignorance on the basis of social science surveys evidence, but when the consequences involve restricting rights, the standards for justified belief become far more demanding.

A final explanation goes beyond the ethical stakes of assessing political ignorance and focuses on the risk of manipulation. When measures of political knowledge serve only an informational purpose, there is little incentive to distort them. But once they determine who can vote, they become a battleground for partisan influence. Those in power would have strong incentives to shape these assessments to favor their own supporters, turning them from neutral tests of civic competence into tools for ideological gatekeeping. This politicization would undermine their reliability, making the tests less about measuring knowledge and more about enforcing political conformity. In other words, the issue isn’t just that higher stakes demand stricter epistemic standards; it’s that tying knowledge assessments to political power invites efforts to control and distort them.

In the end, my students may be right to accept the evidence on political ignorance while resisting the idea that we can reliably identify the politically knowledgeable. Their skepticism is not necessarily inconsistent—it may stem from a deeper moral and political intuition about the risks of making such judgments. While surveys provide compelling evidence that voters, on average, lack basic political knowledge, applying these findings to individual assessments with real political consequences is far more fraught. The historical abuses of voter restrictions, the potential for competence tests to be exploited for partisan gain, and the risk of perpetuating social inequalities all render their skepticism not only intelligible but even well justified.