Condemnations of our Constitution as undemocratic or evil are almost as old as the Republic. Abolitionist William Lloyd Garrison condemned the deal as a “pact with the devil”; President Woodrow Wilson slighted it as a barrier to modern administration; and historian Charles Beard painted it as a naked power grab by landed elites. The chorus of protest has not abated. Frequent targets of criticism in recent decades include the anti-majoritarian Senate wherein California and Wyoming have equal influence on national policy; an electoral college that has twice in the last five elections selected a “winner” who lost the popular vote; and the influence of the federal courts, where judges serve for life, and seem to have free-wheeling discretion to interpret vague constitutional text in light of their own policy preferences.
Today, a new global era of democratic backsliding has brought a more nefarious concern to the fore. This is the way in which our founding document may facilitate action by a president or party intended to undermine the integrity of democratic elections. It is not obvious from the text of the Constitution, for example, that a president could not pardon himself, perhaps even prospectively. Nor would anything prevent a president from pardoning his subordinates who engaged in willful criminal action at his behest, including violence or subterfuge aimed at disabling political opponents. According to the current status quo, as put forward in a memo from the Office of Legal Counsel, the President cannot be indicted for crimes while in office. So the only remedy for an anti-democratic presidential agenda is impeachment. But whatever one thinks of the Trump impeachment trial of 2020, it exposed arguments that presidents have unfettered power to resist any oversight. Most notorious—and rightly vilified—was Alan Dershowitz’s assertion in the course of the impeachment trial that “if a president does something which he believes will help him get elected in the public interest, that cannot be the kind of quid pro quo that results in impeachment.” Come back Richard Nixon—all is forgiven!
All this bodes ill for our democracy. Further facilitating its decay is the theory of the “unitary executive.” This elusive and somewhat multifarious term has in the last two decades migrated from being an obscure academic idea of relatively modest scope to a reading of the Constitution preferred by the President himself, and perhaps many of his judicial appointees. Although the theory has different forms in the scholarly and the public debate, at its core is the Constitution’s assertion that executive power “shall be vested in a President.” In some scholarly treatments, this simply means that the president must be able to fire subordinates. More ambitiously, it can be taken to mean that the President must retain ultimate power over any executive action. Advocates of this theory have long drawn on Justice Scalia’s dissent in Morrison v. Olson, a 1988 challenge to the independent counsel statute. Today, Morrison can be wielded to the sinister conclusion that the president has unfettered control of all federal investigations and prosecutions—including those into his own malfeasance. Were this theory to be adopted, the President could openly and notoriously mandate criminal indictments of his political opponents, while immunizing his supporters from prosecution for interfering with elections.
If this possibility seems a touch dystopic, ask yourself what forces in the United States stand against it. Normally, we think of political parties as restraining forces, playing a central role in extending political time-horizons past the life of current officeholders. But in the United States, as in many countries, partisans seem increasingly willing to tolerate undemocratic action so long as it helps their co-partisans. Another constraint might be the judiciary, which would presumably refuse to uphold spurious prosecutions. But the Supreme Court has crafted rules that make it very difficult to challenge prosecutions as politically biased. In any event, investigations and prosecutions are themselves a punishment of sorts, and can be used to harass the opposition. The increasingly partisan character of judicial appointments also means that judges may turn a blind eye, even to cases of naked political interference. Indeed, the recent closely divided opinion on the addition of a citizenship question to the census is a master class in turning a blind eye to self-evident partisan impropriety.
In response to various nightmare scenarios, we might propose reforms at the federal level. But here we quickly run into the Constitution itself. The notoriously difficult amendment rule in Article V effectively insulates it from reform. The amendment rule is so onerous that we likely will never clarify the pardon power’s scope, let alone establish the constitutionalized accountability machinery that many other countries have. The Constitution also seems to doom us, for the foreseeable future, to a system in which Supreme Court justices will only be appointed when the Senate and Presidency are of the same party, and where lower court judges will be drawn from ever-younger cohorts so they can serve for many decades.
There is, however, one constitutional feature which gives some comfort: federalism. The presence of state governments with vast reserves of legal and practical power remains a robust check in certain parts of the country, even though the balance between states and the center has shrunk over time in favor of the latter. The American federal scheme provides states with resources for resistance, and forces the possibility of negotiation between different governments over core policies.
In the 19th century, resistance was embodied in the idea of nullification—the doctrine celebrated by John Calhoun and others that states had the right to nullify even valid federal laws—and ultimately secession. In our own era, less extreme measures are available. Consider marijuana policy, in which a decision by the Attorney General to classify the drug at the highest level of restriction has been effectively nullified by the 33 states that have legalized some form of use. (An amusing by-product we observed recently at O’Hare Airport is the installation of large, public bins, where one can deposit one’s excess cannabis before entering federal airspace.) Although the ensuing legal limbo remains a nuisance for those in the industry, there is no chance that the federal government will effectively reverse the policy of the states—it simply lacks the coercive resources to do so.
More pertinent here, the collective power of state-level law enforcement has emerged as a key feature of accountability in the Trump era. As conventionally understood, the federal pardon power does not extend to state crimes, and the Double Jeopardy Clause does not preclude a president from being investigated or prosecuted after being pardoned for a federal offense. Although the supremacy clause might prevent a criminal prosecution of the president from proceeding while he is in office, that proposition has not been tested. Further, the ongoing lawsuit about whether the President can be forced to turn over his tax returns to New York state authorities is relevant toward ensuring accountability. State Attorney Generals, then, could be important checks on executive lawlessness.
More broadly, federalism enables policy variance. Decentralization promises greater alignment between preferences and policies, and allows for local experiments—captured in Brandeis’ famous phrase, “laboratories of democracy.” Many innovations of American democracy originated in states and spread among them, especially during the progressive era. Some of these are well known: nonpartisan electoral commissions; direct or primary elections; special councils or commissions for judicial appointments; referendums and initiatives. Direct election of senators and female suffrage started at the state level. Only later was it adopted by the Federal government. Another one, not copied at the federal level, is the plural executive. Many U.S. states have rejected the unitary executive, providing separate elections for Governor, Lieutenant Governor, Attorney General, and Secretary of State. And today, good government reforms continue: In 2020, for example, Maine will be the first U.S. state to use ranked-choice voting in a presidential election.
To be sure, federalism is no panacea. For a century it facilitated what scholars call “subnational authoritarianism” in the Jim Crow South. Even as Brandeis was celebrating the states as laboratories of democracy, they were experimenting with new forms of repression, and this is a concern today as well: Many states are exploring ways of suppressing the vote and restricting political speech. Call them the “laboratories of anti-democracy.”
One important innovation was the decision by North Carolina Republicans, after losing the state governorship to Roy Cooper in 2016, to try to strip the office of many powers before he could take office. (To be fair, this move had been tried before outside the United States: Hugo Chavez had pulled the same move over the mayoralty of Caracas in 2008.) This innovation was partly successful, though other parts were struck down by courts. The technique was then borrowed in Wisconsin and Michigan in 2018. It is one of the tragedies of our highly partisan era that outgoing Michigan Governor Rick Snyder, who vetoed the bill proposed by his co-partisans, was not celebrated as a hero of democracy.
The problem is especially acute given political sorting. At this writing, 36 state governments are in the hands of a single party—more than has been the case in many decades. Single-party government provides a greater incentive to lock in power and also fewer constraints on doing so.
What can we do to encourage states to adopt pro-democratic reforms rather than anti-democratic reforms? One useful wrinkle is that state constitutions are slightly different creatures than their federal counterparts. Their subject matter is limited, precisely because the federal government takes care of many things that a constitution would normally include. They tend to be much more detailed, easier to amend, and so much less enduring: The average state has had several different constitutions in its lifetime. They have many more rights, and much more variety in institutions.
For the foreseeable future, state constitutions may thus offer a promising arena for pro-democratic reforms. We focus on one feature in particular: the management of elections and the role of secretaries of state.
There is a great deal that can be done at the state level to keep democracy competitive in ways that have spillover effects beyond a state’s border. An obvious thing is non-partisan or bipartisan redistricting process, now adopted in some form in 19 states. These reforms are often adopted through initiative and referendum, themselves governance innovations. These institutions have received a good deal of attention: Their example may well prove contagious—seeding a beneficial form of policy contagion.
Less well considered is the management of elections. The office of Secretary of State is in many U.S. states a partisan, elected office. But most of its responsibilities are hardly partisan. In most states these include managing driver’s licenses, chartering businesses, and overseeing elections. So why in 36 states, is the Secretary elected by the public in a partisan election? And why in four more is she appointed by a partisan legislature? Treating this office as inherently partisan is anachronistic, something like the elected coroner running on a party ticket.
It is also hazardous to democracy. In the 2018 election cycle, two secretaries of state ran for offices in elections they were responsible for running. This is compatible with no definition of democracy with which we are familiar. To the contrary, it violates the common law idea of “natural justice”, namely that no person should be judge of her own case. In the end, Kansas Secretary of State Chris Kobach lost his bid for the Governorship, while Georgia’s Brian Kemp won his. Neither recused themselves from overseeing the counting of ballots, which the most basic rules of political morality would seem to require. And both had long records of vote suppression.
Democracy needs neutral referees, and it is a mistake to consider that it can operate without them. In many countries, the Speaker of the parliament resigns his or her party affiliation upon election; the Senate parliamentarian in the United States is the keeper of norms and traditions, and routinely makes rulings that are followed by the parties in politics. The Secretary of State, like these other offices, ought to be a nonpartisan one, or have its powers over elections shifted elsewhere.
One might object that these reforms are hardly likely to be incentive-compatible in a landscape of increasingly one-party states, and a polarized national environment. And yet we have been here before. The Progressives in the 19th century arose against a corrupt gilded age background, in which parties were seen as hotbeds of corruption. If we did it then, we can do it now. No doubt, reform efforts will be bitterly opposed, and sometimes rolled back. But this is no reason to give up on the grand experiment of American democracy. Indeed, state-level reform may be the best and perhaps only response available to the skepticism of the Garrisons, the Wilsons, and the Beards.