When I wrote last week about the troubling eclipse of civic education, the solution I had in mind was a revival of education’s civic mission, powered by a vastly-increased investment in the resources needed to equip people with the knowledge and skills to become effective democratic citizens. But of course, there are cheaper options: you could just use civic tests to block people from voting. That is the suggestion of a puzzling op-ed in Friday’s Washington Post.

I say “puzzling” not because blocking people from voting, whether for their supposed ignorance or for a host of other reasons, is an alien idea. After all, it’s got a long historical lineage and is popular at the moment within one of our two major parties. Rather, what’s puzzling is the lack of a connection between this quite radical suggestion and any particular problem it will supposedly solve.

It ought to go without saying that stripping people of their voting rights is an extreme policy idea. A good argument for doing it is hard to imagine, but this piece doesn’t provide much of an argument at all. It cites a few disturbing facts about the number of voters who can’t name the Vice President, or who don’t know when the Declaration was signed, concludes that this must be behind our political doldrums, and suggests “weeding [them] out.” This is the closest it gets to identifying a particular concern:

A person need only survey the inanity of the ongoing presidential race to comprehend that the most pressing problem facing the nation isn’t Big Business, Big Labor, Big Media or even Big Money in politics. It’s you, the American voter. And by weeding out millions of irresponsible voters who can’t be bothered to learn the rudimentary workings of the Constitution, or their preferred candidate’s proposals or even their history, we may be able to mitigate the recklessness of the electorate.

The claims in this passage resist engagement. Why is the presidential race inane? We are never told. Who has made it that way? Allegedly, it’s “the American voter”—although it can’t be all American voters, since only primary voters in certain states have voted so far. And many of those people voted for losers, so it stands to reason that at least some of them can’t be blamed for the current state of the campaign.

In claiming that the “race” is inane, the piece absurdly assigns equal responsibility to the GOP’s choice and one or both of the Democratic candidates. Get ready for plenty of this over the next five months: language that goes out of its way to avoid grappling with the fact that one party’s nomination, but not the other’s, has been seized by a racist buffoon. The wide scope of the language implies that voters would have chosen neither this race-baiting demagogue, nor a longtime fixture of the Democratic mainstream, nor a democratic socialist if they had had better civic knowledge, or a better knowledge of their candidate’s positions or history. This is a pretty wide array of options that would have been sorted out by better voter knowledge, and it’s hard to square with the fact that the Sanders-Clinton showdown is not inane—in fact, it’s driven in no small part by arguments over the candidates’ respective histories.

If the piece is frustratingly vague in these respects, it’s downright misleading in others. It gives the impression that voters are a horde of illiterate rubes who can be directly blamed for—well, something. And yes, surveys suggest that many of them wouldn’t pass the citizenship test. But it would have been helpful to point out that on the whole, voters are a relatively educated subset of the American population. Research examining the presidential electorate in every race since 1972 has shown that “people with higher levels of education and income are far more likely to cast ballots.” The turnout rate for the highly educated routinely comes in above 75%, and turnout falls along with education levels. Whether they can answer civics-test questions or not (some of which verge on trivia), the poorly educated (and the poorly informed) have legitimate interests too, and it’s not at all clear that democracy is served by excluding them from voting. Tellingly, people who study political participation tend to worry about a different problem: “The well-educated and affluent have long enjoyed powerful megaphones to amplify their political voices. Large class-based gaps have been a feature of U.S. politics at least since the early 1950s, when national surveys started.” The Post’s piece dismisses educational gaps, arguing that “we have unlimited access to information […] it literally takes seconds to learn about the fundamentals of our republic and the positions of candidates.” That ignores the estimated 60 million American adults who don’t use the Internet, as well as the fact that home Internet use—and service reliability—falls along with income.

Still, even if access to information were equal, the existence of uninformed voters would not, in itself, represent a strong argument for stripping their voting rights. Democratic elections are chiefly a guarantor of political legitimacy, not a way of making the best expert decisions—and a democratic state is legitimate because it gives a say to all of its people, not just the informed among them.

Because the piece is so vague about the connections between its claims, it’s easy to overlook these objections while nodding along with the one part of it that is indisputable: many voters do, in fact, demonstrate a startling lack of knowledge about the workings of American government and the basics of American history. This is a problem. But what’s never shown, much less argued for, is that this lack of knowledge has generated some widespread inanity—when really, the madness infecting electoral politics is the result of a combination of long-term trends located within one party. Yet it is general voter ignorance that sounds the alarm at the piece’s outset: “Never have so many people with so little knowledge made so many consequential decisions for the rest of us.” This makes for a confusing moment a few paragraphs later, when the argument abruptly shifts course: “To be fair, the contemporary electorate is probably no less ignorant today than it was 50 or 100 years ago.” Well, which is it? If today’s electorate is no worse than that of 50 or 100 years ago, then why weren’t the candidates just as inane 50 or 100 years ago? Answering these questions might have required a little more detail. It almost certainly would have generated a piece with more to say about our current predicament.