Monday, May 15, 2023

Our Continuing Constitutional Crisis


This post was prepared for a roundtable on Constitutional Crises, convened as part of LevinsonFest 2022—a year-long series gathering scholars from diverse disciplines and viewpoints to reflect on Sandy Levinson’s influential work in constitutional law.
In 2009 Sandy Levinson and I wrote an article in the University of Pennsylvania Law Review, which offered a typology of constitutional crises. We believed that it was important to distinguish constitutional crisis from emergency. In the history of constitutional republics, emergencies happen all the time, but constitutional crisis refers to a very specific kind of event.
Constitutions have many purposes, but one of them is to make politics possible. Constitutions channel conflict and struggles for power into legal and constitutional forms. They cause people to struggle for power within the boundaries of the politics made possible by the constitutional system, instead of going outside those boundaries.
According to this account, a constitutional crisis occurs when constitutions fail—or people reasonably believe that they are about to fail—at this central task.
In 2009, we suggested that there were three ways this might happen.
In Type One constitutional crises, government officials declare that they will no longer abide by the constitutional rules of the system; they go outside of the Constitution in order to achieve their goals or stay in power. There is a related situation in which a court issues a direct order to a government official, and the official simply refuses to obey it without attempting to appeal the decision, or in the case of a direct order by the U.S. Supreme Court, simply defies the order.
By contrast, when the President takes an action contrary to what others think is required by the Constitution and then fights out the issues in the courts, the term “constitutional crisis” is inappropriate. That is not a constitutional crisis. That's just litigation. Indeed, in a well-functioning constitution we would want political actors to go to the courts to resolve disputes about the meaning of the Constitution.
If Type One crises involve political actors dispensing with constitutional fidelity, Type Two crises involve what we might think of as an excess of fidelity, or a misplaced concern with fidelity. These are situations in which everyone is playing by the rules of the Constitution—at least as they understand them—and as a result they are unable to deal with some problem or emergency that causes catastrophe. A special case of this kind of crisis is one in which the Constitution appears to be silent on who has power to act in a particular situation. Hence the government is paralyzed and disaster occurs.
Generally speaking, and with caveats I'll mention later, Type Two crises don't last very long. They don't last very long because somebody comes up with a theory that allows them to act and deal with the problem. So Type Two crises either resolve into a practical solution, into an ordinary political dispute, or turn into one of the other kinds of constitutional crisis.
Sandy and I emphasize that constitutional crisis is not the same thing as constitutional disagreement. People disagree all the time about what the Constitution means and how to apply it. These fights are waged in politics and in the court system. As long as that happens, there is no constitutional crisis.
But sometimes the disagreements become so heated that the system breaks down. People take to the streets, the military stages a coup, political actors attempt to seize power or try to secede, and so on. In these cases, we have more than simple disagreement about the interpretation of the Constitution. We have disagreement that breaks into widespread violence, coups, revolutions or attempts at secession. These are Type Three crises. This situation is a constitutional crisis because the Constitution has clearly failed in its central purpose of channeling conflicts over power into the forms and practices of law.
Not all violence is political violence, and not all political violence constitutes a constitutional crisis. A riot in response to an unjust jury verdict is not a constitutional crisis, even if the rioting lasts many days. But if the violence spreads so that there is a genuine danger of a coup or a revolution, this presents a constitutional crisis.
Writing from the perspective of 2009, Sandy and I pointed out that there were very few examples of genuine constitutional crises in American history, although there had been many political crises and many emergencies, as well as many acts of political violence.  People have often referred to political crises as constitutional crises, but usually this was not the case. The Constitution was still working, and people were struggling with each other within the terms of the Constitution. Type One and Type Three crises were uncommon, and Type Two was exceedingly rare. From the perspective of 2009, at least, the United States had been relatively fortunate.
That's not to say that there weren't any constitutional crises in the country’s history. The most obvious example of a constitutional crisis is the Civil War, which was a crisis of all three types. It was a Type One crisis because members of the Confederacy refused to obey the Constitution. It was a Type Three crisis because a dispute about the Constitution turned into a civil war.
It was also, for a brief period of time, a Type Two crisis as well, because then-president James Buchanan took the position that although secession was illegal, there wasn't anything he could do to stop it. This was an example of misplaced fidelity to the Constitution that leads to disaster. But as Sandy and I pointed out, Type Two crises don't last long. When Abraham Lincoln became president, he immediately took a contrary interpretation from his predecessor Buchanan; he believed that the Constitution allowed him to take steps to resist Southern secession, leading to the Civil War.
That's where our analysis stood, more or less, when we wrote the piece in 2009, at the end of the George W. Bush Presidency. We have had occasion to rethink our views in the last decade and a half. Here I want to mention three points.
First, Type Two crises may be more frequent than we had originally suggested. In fact, there is a chance that we may be in the midst of a Type Two crisis very soon. Currently Republicans in the House of Representatives are engaged in hostage taking over the debt ceiling: They argue that they won't raise the debt ceiling unless the President agrees to draconian cuts that, among other things, would unravel his signature domestic policy achievements. So far President Biden has refused to budge, demanding a clean debt ceiling increase, although at the same time, the two sides are working on a compromise on spending.
Now suppose that negotiations break down, and the President of the United States decides that he has no way to do an end run around Congress. The results would be catastrophic—a global financial meltdown. This would be a Type Two crisis.
Like other Type Two crises, this one is likely to be temporary. If matters came to a head, the President has tools at his disposal to resolve the situation. I wrote about these alternatives at length in the summer of 2011, during the debt ceiling crisis in President Obama's first term. I won’t repeat my arguments here but you can look at my posts from that period if you are interested in the details.
More broadly, I think that Sandy’s work on constitutional design sheds an interesting perspective on the idea of a Type Two crisis. Sandy's view is that poorly designed features of Constitution itself may lead us into untenable situations from which it is very hard to extricate ourselves. In this sense, although we described Type Two crises as the rarest of the three types, they may always be lurking in the background because of problems of constitutional design. Perhaps Sandy might say that the U.S. Constitution is so badly designed that we are experiencing a Type Two crisis in slow motion.
Second, Trump's presidency led me to suggest a minor amendment to our original model. In our initial formulation we said that Type One crises required a political actor—here the President—to publicly state that he would not abide by the Constitution. In hindsight, this was too stringent a requirement. It is enough if the President, while outwardly professing fidelity to the Constitution, secretly plots to violate or subvert it.
Moreover, if the President’s public defenses of his actions are sufficiently implausible, we are entitled to interpret the situation differently, and conclude that he is, in fact, trying to subvert the Constitution. Put another way, the fact that the President is gaslighting the public or acting in bad faith doesn't mean that we don't have a constitutional crisis. Quite the contrary, in a genuine constitutional crisis—such as an attempted coup—we can expect that at least some officials will be acting in bad faith and will attempt to gaslight the public.
The events of January 6th clarify our argument. January 6th was a Type Three constitutional crisis. Trump’s acolytes, goaded by the President, marched on the Capitol and tried to prevent the peaceful transfer of power by disrupting Congress's constitutional duty to count the electoral votes in the 2020 election.
But January 6th is also a Type One crisis, despite Trump's repeated claims that the election was rigged and that he was just trying to prevent the election from being stolen. The January 6th House Committee showed what people had long suspected: that Trump understood that he lost the election, but he nevertheless plotted and schemed to find a way to stay in power.
Trump made specious legal and constitutional arguments to justify his position. But he did far more than attempt to litigate his claims in the courts, where he repeatedly lost.
Trump tried to bully state officials into changing votes on his behalf. He tried to bully the Vice-President into manipulating the electoral count his way. He plotted to arrange for alternate sets of phony presidential electors. And he called for a Stop the Steal rally on January 6th and whipped up the crowd to attack Congress with the goal of throwing the process of counting the electoral votes into confusion, so that he might be able to stay in power.
Trump publicly maintained that he was abiding by the Constitution and that his opponents were rigging the elections and violating the Constitution, but the evidence is clear that he was simply doing and saying whatever it took to stay in power. This is a Type One crisis.
One might also argue, along the lines of Sandy's work on constitutional design, that the last days of the Trump presidency were also a Type Two crisis. That is because the Trump presidency has shown that there is no effective way to remove or sanction a president who plots to subvert an election and prevent the peaceful transfer of power. In a polarized politics, the impeachment provisions of the Constitution are basically a dead letter. The Republican Party has rallied around the former President, and it has now become an article of faith in large parts of the Republican Party that the 2020 election was stolen, even though there is no evidence to justify this.
This brings me to my final point about our 2009 article. It is the relationship between constitutional crisis and what I have called constitutional rot.
Constitutional rot is the process by which a constitutional republic becomes increasingly less democratic and less republican over time. In previous work I've pointed to four recurrent features, which I call the Four Horsemen of Constitutional Rot: They are (1) increasing wealth inequality; (2) increasing political polarization; (3) increasing social distrust, including distrust among government officials of different parties and distrust by the public of politicians and other elite institutions more generally; and (4) policy disasters, by which I mean disasters in government decisionmaking that show that the government is not responsive to the public and is not looking out for the interests of ordinary people.
I have distinguished constitutional rot from constitutional crisis in two different ways. First, crisis is something that happens over a relatively short pace of time, whereas rot is cumulative over many years. Second, crisis applies to constitutions in general, whereas constitutional rot is a feature of democracies and republics.
But Sandy has pointed out to me, and I agree, that the two ideas are related. The more pronounced constitutional rot becomes, the easier it is for a constitutional crisis to occur. Prolonged constitutional rot eats away at the supports of representative government, a bit like slowly sawing a tree branch on which you are sitting. At some point the branch will give way and you will crash to the ground.
Increasing inequalities of wealth, increasing political polarization, and increasing levels of social distrust feed on each other. Distrust of political elites and institutions leads to a proliferation of conspiracy theories, making it easier for demagogues to spring up, and easier for authoritarians to spread disillusionment with democracy generally. Couple all this with a series of disastrous policies that suggest to the public that politicians don't care about them, and you create fertile ground for constitutional crisis. Put another way, a constitution is more likely to fail when its supports in civil society rot away.
Any theory of constitutional crisis must take into account both long-term questions of constitutional design, which has been Sandy's concern through his career, and questions of constitutional rot, which have been the focus of my work over the past decade or so. Both constitutional design and constitutional rot seem increasingly central to understanding why constitutional crises occur and how they develop.
One last point also relates to the interaction between rot and crisis—the question of how long constitutional crises last. In particular, I am interested in whether the constitutional crisis that began on January 6th is actually over.
In one sense, the constitutional crisis that occurred on January 6th is over. The insurrection was stopped, the Capitol was cleared, and the members of the House and Senate met together and finished the business of counting the electoral votes. Joe Biden was inaugurated as president on January 20th, 2021.
In another sense, however, the constitutional crisis that occurred on January 6th is still ongoing. Many Americans continue to believe that the electoral system is corrupt, and that the 2020 election was stolen. Former President Trump and his allies in Congress and in the conservative media have encouraged these beliefs. Conspiracy theories about the opposition lead people to see their political opponents as criminal, evil, satanic, and worse. MAGA followers have hounded election officials, causing them to resign. Trump's most unhinged supporters have called for violence as a way of solving the country's problems. They no longer believe that elections are fair or that democracy works unless their candidates win. To the extent this remains the case, or gets even worse, the supports for democratic constitutionalism remain weak and uncertain.
Adding to the sense of uncertainty is the fact that former President Trump, who conspired to overthrow the government after the last election, is currently the front runner for the Republican nomination, and he has not committed to respecting the results of the 2024 election if he loses. Quite the contrary, he has continued to insist that the 2020 election was stolen and stated that he will pardon people involved in the January 6th insurrection. He has given no indication that if he loses he will not try to overthrow the government a second time. After all, so far he has suffered no legal consequences for his last attempt.

And if Trump wins the 2024 election and gains power a second time, it is hardly clear that the constitutional system will be safe from crisis. Rather, constitutional rot will continue unabated.
All of this suggests that the constitutional system remains under significant threat. And when people reasonably believe that constitutions are about to fail at their central task of keeping struggles for power within the boundaries of the Constitution, that is a constitutional crisis.
To this extent I think it's fair to say that the crisis of January 6th is not over. The insurrection and Trump's plotting to take advantage of it are not isolated events; they may be the herald of something far more ominous. We do not yet know when the constitutional system will regain its footing and the danger will end.
It is not unheard of for a constitutional crisis to last for a long time. The Civil War was the most important constitutional crisis in American history. Hostilities lasted from 1861 to 1865, but as Gregory Downs has explained, there is a good argument that the crisis of the Civil War was not fully over until sometime in the 1870s, when Southern states were readmitted and federal troops begin to leave the South.
The crisis of January 6th was brought on by an advanced case of constitutional rot, which has not receded. If anything, the rot in our constitutional system appears to be even worse than it was five years ago. Recent disclosures suggest that the rot has spread to the Supreme Court, which is unable or unwilling to police its members’ conflicts of interests and corruption and which has lost the confidence of much of the public. Nor does the political system seem able to engage in the kinds of constitutional reforms that would help prevent constitutional failures in the future, in part because polarization and distrust have grown so strong. As long as constitutional rot remains severe, there is every reason to think that the constitutional crisis of January 6th is not over. Our democracy remains in grave danger, and that danger will continue for some time to come.
Jack M. Balkin is Knight Professor of Constitutional Law and the First Amendment at Yale Law School. You can reach him by e-mail at

Older Posts
Newer Posts