clock menu more-arrow no yes mobile

Filed under:

How to lose a constitutional democracy

Turkey's President Recep Tayyip Erdogan reviews an honor guard before attending the oath-taking ceremony of the newly elected members of parliament in Ankara, June 2015.
Turkey's President Recep Tayyip Erdogan has curtailed the media and rooted out opponents in the civil service — a recent example of the worldwide trend of “de-democratization.”
Kayhan Ozer / Anadolu Agency / Getty

As the Trump administration finds its feet, a fear of autocracy is in the air. Some spy the beginning of a sustained assault on our democratic order — envisioning a world in which “keeping the president happy” becomes a widespread corporate goal, as the president metes out warnings on Twitter to companies that threaten his business interests. They sense disdain for constitutional limits on presidential power in Trump’s attack on a “so-called judge” who dared to issue a stay against his hastily issued executive order banning refugees. And they wonder if he will comply if the courts eventually rule against him.

Others worry about the powers the bellicose Trump might assert — and be granted by a supine Republican Congress — in the event of a terrorist attack. On the other hand, plenty of conservatives accuse liberals of crying wolf, confident that no such democratic crisis is imminent.

Who’s right? One reason for the uncertainty is that Americans don’t really know what backsliding from democracy looks like, at least not firsthand. The United States has the world’s oldest democratic Constitution still in force. Despite the Civil War, two world wars, and countless emergencies, national elections have never been postponed. Britain, by contrast, canceled elections during World War II.

It is true that Lincoln suspended the writ of habeas corpus while waging war against the South, that antiwar activism was effectively criminalized in World War I, and that human and civil rights have been violated during other crises. But we lack a history of the systematic corrosion of the three main pillars of our democratic institutions: elections, the rule of law, and freedom of speech. As a result, we lack the historical experience needed to evaluate the current risk to key national institutions.

The rest of the world, however, hasn’t been so lucky — and it is there that we can turn for hints about the dangers of the current situation. In the past decade, an increasing number of seemingly stable, reasonably wealthy democracies have retreated from previously robust democratic regimes toward autocracy. These states are literally all over the map: They range from Eastern Europe (Hungary and Poland) to the Mediterranean (Turkey) to Latin America (Bolivia and Venezuela). Once-anticipated democratic gains in Russia and China haven’t materialized. Meanwhile, a hoped-for “fourth wave” of democracy in the Arab Spring’s wake has dissipated into bitter civil war or authoritarianism.

It was once thought that when a reasonably wealthy country achieved democracy, it would almost certainly maintain it. No more.

Democratic backsliding is far less rare than political scientists used to believe. In a recent academic paper, we identified 37 instances in 25 different countries in the postwar period in which democratic quality declined significantly (though a fully authoritarian regime didn’t emerge). That is, roughly one out of eight countries experienced measurable decay in the quality of their democratic institutions.

Scholars used to argue that democracy, once attained in a fairly wealthy state, would become a permanent fixture. As the late Juan Linz put it, democracy would become “the only game in town.” That belief turned out to be merely hopeful, not a reality.

As a result, the global trend for democracies — the other categories being partial or complete autocracies — does not look positive, as the following chart shows. While we are not yet to the point where democracies are rare, as in the 1970s, it is quite possible that the “third wave” of democratization has peaked. And the recent de-democratization trend stands out:

A graph showing that the number of democracies in the world increased from the 70s to the mid aughts, but have declined since.
Regime types in the third wave.
Adapted from Freedom House data, 2016.

Does the experience of the rest of the world matter for the United States? That might seem like an odd question. But since at least the time of Alexis de Tocqueville, commentators have argued that our country has a distinctly strong democratic tradition and temperament. Indeed, the phrase “American exceptionalism” emerged in US communist circles in the 1920s in the course of efforts to explain the apparent immunity of the United States to proletarian revolution. American exceptionalism has since become something of national credo. It’s all but obligatory, at least in political circles, to say that the founders created a marvelous system of checks and balances that would defeat any attempt at a power grab.

But having carefully studied the experiences of other countries as well as our own Constitution, we think complacency is unwise. The United States is not exceptional. It is instead vulnerable to the most prevalent form of democratic backsliding — a slow descent toward partial autocracy.

Coups are rare (and growing rarer). Democratic “backsliding” is common.

In at least one regard, however, those who worry about an overreaction to Trump are correct. The sudden and dramatic end of democracy in the US, as in a military coup, is highly unlikely, even if this has often been the engine fueling dystopian fiction and film.

Coups, of course, do happen. In May 2014, for example, the Thai military suspended that country’s constitution and ended democratic rule. A year earlier, the Egyptian military ousted then-President Mohamed Morsi in favor of Gen. Abdel Fattah el-Sisi. By contrast, an attempted coup against Turkish President Recep Tayyip Erdoğan in 2016 failed.

But despite these high-profile examples, coups are in fact increasingly rare. A 2011 study of democratic backsliding identified 53 historical cases of democratic decline. Out of those, only five involved coups or other sudden collapses into authoritarianism.

What’s more, since the 1950s, coups have become increasingly infrequent. And they usually take place in a context quite different from the American situation. Full-on democratic collapse tends to occur in recently established, relatively impoverished democracies, in which civilian control of the military is tenuous. None of those conditions apply in the US (despite economic problems such as rising inequality).

What about the quick strangulation of democracy using emergency powers? The use of such powers is not uncommon. From 1985 to 2004, 137 countries invoked state-of-emergency procedures at least once. Commentators who worry about Trump’s behavior after a terrorist attack have something like this in mind.

It is certainly true that the Constitution lacks the careful restrictions on emergency powers that other countries’ constitutions employ. They typically place constraints on the length and scope of extraconstitutional behavior, and they name the constitutional actors who must sign off on the emergency measures. Not so the US Constitution.

Rather, American presidents and judges have inferred vague emergency powers into many of the Constitution’s key clauses and phrases — such as “Commander in Chief,” which is one basis for the president’s power to respond to sudden attacks, or the president’s power to “take Care” the laws are enforced, which has been used to justify expansive executive power under both Democrats and Republicans.

But this drafting failure in the Constitution may paradoxically work to democracy’s advantage. The very fact that government has a great deal of legal discretion in responding to perceived crisis — often to the detriment of important liberty and dignity interests — means there is far less plausible justification for calling off the regular processes of elections to deal with a crisis.

The Constitution, in other words, often fails to protect individuals when an emergency occurs, as it failed to protect Japanese Americans from internment, and failed to protect some foreign nationals from torture after 9/11, but in so doing it may be saving democracy writ large.

The most important reason the sudden collapse of democracy is rare — and a key reason it is unlikely in the US — is that a sudden derogation of democracy simply isn’t necessary. Would-be autocrats have a cheaper option to hand, one that is far less likely to catalyze opposition and resistance: the slow, insidious curtailment of democratic institutions and traditions.

Autocrats often target the three pillars of democratic society using tools permissible under law

To understand democratic backsliding, it’s important to understand the essential components of a democracy. First, there must be elections, which must be both free and fair. Elections by themselves are not enough: Both Russia and China, after all, have elections that formally reflect the choice of the people, but allow only limited choices.

Second, democracy needs liberal rights of speech and association so those with alternative views can challenge government on its policies, hold it accountable, and propose alternatives. Finally, democracy can’t work if the ruling party has the courts and bureaucracy firmly in its pocket. The rule of law — not just the rule of the powerful and influential — is essential.

Take away one of these attributes, and democracy might wobble. Sap all three, and the meaningful possibility of democratic competition recedes from view.

Comparative experience shows that would-be autocrats find it critical first to control the public narrative, often by directly attacking or intimidating the press. Libel suits — Putin notably recriminalized libel, after it had been decriminalized in 2011 under Dmitry Medvedev — drummed-up prosecutions, and vise-like media regulation accomplish the same ends.

Conjuring or overemphasizing a national security threat creates a sense of crisis, allowing would-be autocrats to malign critics as weak-willed or unpatriotic. Other rhetorical moves are common: Leaders who wish to roll back democratic institutions tend to depict those institutions’ defenders as representatives of a tired, insulated elite.

An independent judiciary and checks such as legislative oversight of administrative activity can prove significant barriers. Hence, we often see would-be autocrats trying to pack the courts or intimidate judges into getting with the program.

When the state bureaucracy insists on rule-of-law norms, it too must be bullied into submission. Weakening civil service tenure protections is an underappreciated way for an executive to aggrandize power. When government workers hired on the basis of merit are replaced by partisans, this not only removes one potential source of opposition to the executive branch; it enables a would-be autocrat to direct formidable prosecutorial and investigative apparatuses against political foes. The recent fraud conviction of Putin opponent Alexei Navalny shows how such tools can be used against an opponent who threatens to amass power through electoral popularity.

Finally, political competition must be stanched, even if elections proceed in some form as a way of enabling leaders to claim a mantle of legitimacy. Modifying term limits is a common move, as is changing the rules of elections to permanently lock in temporary majorities.

Two case studies in de-democratization: Hungary and Poland

To see the full panoply of these measures being deployed against democracy, there are no better contemporary case studies than Hungary and Poland. Populist governments in both countries have straitjacketed independent courts, dismantled independent checks on political power, used regulation to muzzle the media or stack it with cronies, and conjured supposed security threats from immigrants and minorities as a justification for centralizing power and dismantling checks.

In Hungary, the Fidesz government used constitutional amendments to entrench its slim (53 percent) majority beyond easy electoral challenge by changing the composition and operation of a previously independent electoral commission. The result was that in 2014, it won two-thirds of the parliamentary seats with 45 percent of the vote.

In 2015, the Polish Law and Justice Party did not need a constitutional amendment in order to remake the judiciary in its image. It simply refused to seat judges that had been appointed to Poland’s highest court by the outgoing party. The Law and Justice Party declared those appointments unconstitutional, then named its own slate. The party also raised the voting threshold for the court to strike a law (to two-thirds), but this change was declared unconstitutional by the constitutional court itself. The government then refused to publish this and other rulings of the constitutional court, creating legal confusion and leading the outgoing chair of the court to say that Poland was “on the road to autocracy.”

Hungary and Poland are hardly unique. In Turkey, President Erdoǧan leveraged the 2016 coup attempt to deepen his massive purge of almost every state institution, leaving regime loyalists firmly in control. As of this writing, more than 135,000 soldiers, judges, police, university deans, and teachers have lost their jobs, in some cases without due process. His AK party has also suspended and manipulated media licenses, and arrested journalists on national security grounds.

In Venezuela, the Chávez regime has notoriously aggregated executive power, limited political opposition, attacked academia, and stifled independent media — a classic example of “de-democratization” under the color of law. Some moves have been especially creative. When a political opponent won at the municipal level, the Chávez regime responded by gutting the powers wielded by the new mayor.

Many of these examples of democratic backsliding proceeded through formally constitutional legislation or administrative processes. Alarm in response to each of them can thus be condemned as excessive or histrionic. But the cumulative effect of many small weakening steps is to dismantle the possibility of democratic competition, leaving only its facade. It is a death by a thousand cuts, rather than the clean slice of the coup maker.

This is what makes the slow road from democracy so alluring to seekers of power, and so dangerous for the rest of us. Because it can be masked with a veneer of legality, it can be cloaked with plausible deniability. It is always possible to justify each incremental step.

Paul Ryan talks with Donald Trump and Mike Pence, on stage at the GOP Congressional retreat, January 2017.
The founders assumed Congress would naturally check the executive branch. But that assumes Congress members would place institutional loyalty over party ambition — which is far from clear.
Bill Clark / Getty

Can we rely on the Constitution?

So could it happen here? Looking to these recent examples suggests that the US Constitution may be good at checking coups or the anti-democratic deployment of emergency powers, but it isn’t well-suited to stall the slow decay of democracy. Our 18th-century Constitution lacks provisions necessary to slow down a would-be autocrat bent on the slow dismantling of the republic.

To be sure, the cumbersome American process of constitutional amendment shuts off one avenue for a president who wishes to amass power — say, by ending term limits. But other much-cited checks and balances have been overrated.

James Madison thought the divergent “ambitions” of the legislative and executive branches would cause those institutions to balance one another. But he failed to anticipate the rise of parties, and how they would reshape incentives. Congress members today may have little reason to investigate or otherwise rein in an aggressive president of their own party, as we are now witnessing. That Republicans are not eager to investigate President Trump’s financial dealings, or his contacts with Russia, is entirely predictable, from an institutional standpoint.

Other constitutions give minority parties rights to demand information and make inquiries, but the US Constitution does not.

Where other nations have independent election officials, too many of our election rules depend on the good faith of the party in power. As the omnipresence of gerrymandering shows, good faith may not be enough. After the 2010 redistricting in Wisconsin, the GOP was able to win 60 of 99 seats in the state legislature, despite winning less than half of the statewide vote. (A case challenging Wisconsin’s gerrymandering will be heard by the Supreme Court.)

North Carolina Republicans tried a strategy that was straight out of the Chávez playbook when their party’s candidate lost the governor’s race: They cut the governor’s staff by 80 percent, eliminated his ability to name trustees of the state university, and required that cabinet appointees be approved by the legislature. They also restructured the elections board so that they would hold the chairmanship during all statewide elections. These moves remain tied up in court.

The courts are critical in upholding the rule of law. But there is a growing acceptance in American jurisprudence of “deference” to the political branches. That ideology, in combination with aggressively partisan appointments — Trump is in a position to fill 112 federal judicial vacancies, out of 870 seats — could erode public confidence in judges’ ability to stand up to government overreach, and thus lead to democratic retrogression.

The independence of even of the Supreme Court is dependent on “norms,” not constitutional rules — and norms can change. In a less polarized time, the US Senate would have held confirmation hearings for Merrick Garland, President Obama’s last Supreme Court nominee, yet by playing hardball, Republicans may end up reshaping how laws are interpreted for decades to come.

Similarly, in the United States, the civil service, which scholars understand as a bulwark against autocracy, is protected largely by tradition. That is why the Republican move to lay off federal workers and reduce the benefits of those who remain is so significant, as is a gratuitous revival of a rule that lets them punish individual bureaucrats by slashing their pay. US attorneys also serve “at the pleasure” of the president; it is largely self-restraint (not always exercised) that prevents presidents from punishing them or rewarding them for partisan legal attacks.

Yet other constitutions create independent ombudsmen offices to monitor corruption or human rights compliance. Not so ours.

While the First Amendment (currently) limits the misuse of libel law, it does not hedge the risk of partisan media regulation by the Federal Communications Commission or other agencies. Media companies seeking to keep regulators’ favor have now lots of reasons to trim the sails of their political coverage. And the First Amendment, for good or ill, arguably protects sources of outright propaganda — sites spreading lies about politicians, for example — which could in tandem with presidential attacks on the media as an “enemy of the American people” lead citizens to distrust all news sources.

There is, in short, nothing particularly exceptional about the American Constitution — at least in any positive sense. Because of its age, the Constitution doesn’t reflect the learning from recent generations of constitutional designers. If anything, it is more vulnerable to backsliding than the regimes that failed in Poland, Hungary, Venezuela, Turkey, and elsewhere.

What will prevent backsliding from occurring here?

Whether or not the United States moves away from its best democratic traditions doesn’t rest on the Constitution or on simple fidelity to constitutional institutions. Those won’t be enough. Nor will it be enough to belabor the technical legal merits or demerits of specific executive actions, or their opponents’ responses. To do so misses the forest for the trees.

Rather, the degree to which democratic norms and practices are lost in the United States over the next four years will depend on how both politicians and citizens react. The quality of our democracy will depend on what happens on the streets, what happens in legislative backrooms (especially on the Republican side), and, most importantly, what happens at the polls. But it won’t depend, in any simple way, on the Constitution. And at least in this regard, there is nothing exceptional about our current predicament.

Surveying democratic backsliding around the world, the clear lesson is: Not every wolf threatening democracy howls and bares its teeth. Many threats are stealthy. The founders certainly knew that. They did not devise autocrat-proof institutions, but they were keenly aware of politicians’ autocratic tendencies, and felt a great trepidation about whether American democracy should endure.

We would do well to reject feel-good talk about American exceptionalism and embrace some of the founders’ trepidation.

Aziz Huq is the Frank and Bernice J. Greenberg professor of law at the University of Chicago Law School. Tom Ginsburg is the Leo Spitz professor of law at the University of Chicago Law School and is on Twitter @tomginsburg. They are the co-editors of Assessing Constitutional Performance.

The Big Idea is Vox’s home for smart, often scholarly excursions into the most important issues and ideas in politics, science, and culture — typically written by outside contributors. If you have an idea for a piece, pitch us at

Sign up for the newsletter Sign up for Vox Recommends

Get curated picks of the best Vox journalism to read, watch, and listen to every week, from our editors.