Proposals for constitutional reform are afoot in ways that I have not seen in my lifetime.

All of this activity reflects a fear: the age of Trump, with its relentless populist assault, and its rule-by-falsity. The very, very real assault on our Capitol in January 2021 led the Chinese and the Russians to scoff at our democracy. If President Biden succeeds in breathing life into our self-government, he will be known to history as resuscitating a democracy that, during the Capitol siege, appeared on life-support.

The legal academy where I work has largely been unprepared for this moment. Liberal constitutionalists have been extolling rights and teaching about courts for decades. Progressives have been so enamored of “rights” and “courts” that they teach virtually nothing else about constitutions. There are almost no courses devoted to the institutions of Congress or the presidency in law schools, except perhaps in comparative law courses. The vast majority of students learn the following from constitutional law: Politics is bad, rights are good, and courts can fix everything. This is not a very good recipe for building a constitution or saving democracy.

More than a decade ago, I asked a roomful of academics at Harvard Law School what the most important rule in democracy was. No one answered. I said I’ll give you the number. I said Rule 22. Someone from the back of the room yelled “interpleader.” That answer pretty much says it all about legal education and democracy. “Interpleader” is the name of an arcane civil procedure rule about the parties in judicial cases. It has nothing to do with the basics of democracy, but it has everything to do with legal academics’ view that courts sit at the center of the constitutional universe. Rule 22 is the Senate rule that closes debates, which ends a filibuster. That rule requires a super-majority of 60 on legislation, all legislation—and has since the 1990s. As too many have now realized, the filibuster is a very deep problem for democracy, it mandates minority rule, but I doubt anyone teaches it in constitutional law even today.

Unlike most of the legal academics involved in this project, I’ve actually worked as a lawyer in the Senate and the White House. I have written and believe that the most basic thing to any constitution is “constituting” the public, borrowing a phrase from the political philosopher Philip Pettit. The Constitution’s institutional structure is at its most central an arrangement of voters. Among legal academics, I have taken the unpopular position that the structure of our separated powers Constitution can be defended as better than parliamentarianism on at least one dimension: It is better for voters. Matthew Stephenson and Jide Nzelibe, trained as both political scientists and lawyers, have shown that my ideas are not simply pipe dreams in a sophisticated piece in the Harvard Law Review. But the proposition is simple: If a dictator takes over the White House, you can thwart her in the House, or the Senate, or the states. Voters have more choice than in a parliamentary system in which parties unify institutions and players. Our system is wildly inefficient, yes. But efficiency has never been the virtue of representative government.

My recommendation to reformers: Look at the incentives that voting arrangements provide politicians. That is the heart of any constitution, its engine. The “constituting public”—the voting public—reconstitutes itself every election. If that electoral engine is broken or tainted, the institutional
superstructure will mirror those defects. To be sure, there must be brakes on the engine; no one denies that. The reason that all those “rule of law” movements in foreign countries a few decades ago did not actually yield democracy was that you have to have a people devoted to the democracy as much as a set of laws that restrain democracy. One cannot know the proper brakes to use unless one understands the engine.

I have serious doubts about many of the proposals one sees in the Democracy constitution. A basically unicameral legislature fractured by parties or local interests will not resist a demagogic President in practice even if in theory it has broad legal powers. But at the very least, this effort has shown how important and difficult structural decisions can be. Progressives have finally engaged on a subject that that they have generally ignored, at least progressives in law schools. Before now, they have largely ceded structural matters to the Federalist Society, who claim the separation of powers as a mantra. We know what that gave us: a textual analysis that allowed Donald Trump to say that he could do anything he
wanted because Justice Antonin Scalia once wrote in a dissent that the President had “all” executive power.

It should not be surprising to hear that the most important thing about this exercise was that liberal constitutionalists came to understand the difficulty and importance of structure. It is also not surprising that the most important set of reforms in this constitution, in my view, are those focused on the electoral system. The current Constitution gives Congress not nearly enough power over voting, and the Supreme Court is poised to read its powers narrowly. If we want to save democracy for the future, it is essential to work on the engine. The political science empirical literature is still out on some of these proposals. But we must begin to experiment with national standards on election safety and security, we must consider multi-member districts in the House, and ranked-choice voting and instant runoffs. And if we cannot, at least in my lifetime, end the gross malapportionment of the Senate, giving a few people in Wyoming more power than millions elsewhere, at the very least we can ditch the current version of the filibuster, which magnifies that most anti-democratic feature of our Constitution. In a perfect world, we would, as Ken Shepsle, Matt Stephenson, and Jon Gould have suggested in a new piece, turn filibuster reform into malapportionment reform by counting population in the Senate rather than the number of states voting. None of these ideas may sound visionary, but setting a federally mandated workweek of 40 hours in 1930 was not visionary either; it was an idea whose time had come.