Although there were surely other forces driving pro-Brexit sentiment, David Johnson’s recent Aeon essay brings the numbers to show that, by at least one count, the critics had a point: the EU simply isn’t built for democratic responsiveness. While the global average today is one representative per 146,000 people, the EU has one per 677,000. And by this measure, the U.S. fares even worse: the 435 members of the House of Representatives amount to one representative per 744,000 people. No wonder, writes Johnson, that “the alienation Americans feel from Washington is comparable to what many Europeans feel from Brussels.” To address this problem, Johnson thinks leaders on both sides of the Atlantic ought to consider “clos[ing] this distance by expanding their legislatures.”

Johnson’s argument boasts a prestigious ally: no less an authority on institutional design than James Madison proposed amending the Constitution so that the size of the House would grow along with the U.S. population. (Larry Sabato also advocated it in this journal.) But in the early twentieth century, Congress capped the body at its current number—with, Johnson suggests, a range of deleterious consequences. For one thing, such a relatively small number of representatives can’t possibly stand in for the varied interests and perspectives of a nation as large and diverse as the twenty-first century U.S. And even if it did more closely reflect the population it serves, its members would have too many constituents to sustain connections with them. Johnson frames this as a basic problem of democratic accountability: “Citizens want to know whom to complain to.”

I would put that last complaint a little differently. A high ratio of citizens to representatives doesn’t generate a knowledge problem, per se: people might still know very well whom they ought to complain to. And if they don’t, the problem might have less to do with numbers than with other features of the American system that diffuse responsibility, such as federalism, the separation of powers, or the current reality of a divided federal government. But high ratios do cut down on the time and resources that representatives have for each constituent, or the likelihood that they’ll have enough in common to understand each other. As Johnson writes:

For example, I live in Berkeley, California, a city of about 120,000 people. I feel that I can communicate and be heard by our mayor, although I suspect it reaches a limit of the sort of direct connection required between citizen and representative. My US Representative, by contrast, also serves nearby Oakland, Emeryville, Alameda, San Leandro, and Piedmont – which together total more than 700,000 constituents. Not only do I not feel a personal connection with her, but these communities, though similar to Berkeley in many respects, are also significantly different.

Responding to this problem by proposing an expansion of representative bodies places this essay in one of two camps in current thinking about the problems facing large modern democracies. For some writers, the deficiencies of contemporary representation should prompt an increase in opportunities for democratic participation—openings for ordinary citizens to exercise more direct control over the activities of government. But for those thinkers still committed to the criticisms of direct democracy that partly motivated the creation of representative institutions in the first place, the solution is not to shift powers away from our representatives, but rather to make them (or us) function better.

One example of this approach would be to increase the size of parliaments. Or, turning our focus to citizens, we might instead cultivate skills and habits suited to the simple fact that most people (to take an unsentimental view) are unlikely to have much more than a passive relationship to their government. On this view, favored by a number of self-identified realists, most of us will never participate directly in politics—but we could do a much better job of evaluating and choosing the people who do.

Of course, these approaches are not mutually exclusive; they might be distinguished chiefly by their points of emphasis. For that reason, reformers might find an all-of-the-above approach to be the most appealing. The contemporary crisis of democracy does not seem to recognize national boundaries, and it has revealed cracks and dysfunctions in the world’s most important political bodies. If we want to preserve them, it’s time to start thinking as seriously as their creators did about the questions of, and principles behind, institutional design.