How safe is liberal democracy? The elections and popular referendums of the past year, especially in the West, raise many questions — but much of the discussion has focused on the popular appeal of democracy.

Political scientists Yascha Mounk and Roberto Foa made a big splash when they argued that younger generations are falling out of love with democratic institutions. Erik Voeten, a professor at Georgetown University, quickly responded that no, public opinion polling data actually shows little evidence that attitudes have changed.

Millennials and other generations may have different views on democracy, but these differences tended to be modest and largely confined to the United States. Voeten conceded, however, that there were plenty of other things to worry about.

Who’s really undermining democracy?

In our recent study of “dysfunctional democracy” for the journal Government & Opposition, we focus on the role of politicians and democratic institutions. It turns out that there are many ways in which the day-to-day practices of relatively anonymous politicians can cripple the functioning of liberal democracy, no matter how many institutional safeguards are in place.

This is, of course, hardly a new insight. Jean-Jacques Rousseau made much the same observation more than 200 years ago in his pamphlet “Considerations on the Government of Poland.” A century later, Woodrow Wilson used a similar argument to frame his study of American politics and critique of the federal system, “Congressional Government.

Lessons learned sometimes require repetition to stick, however. We think this is one of those moments where the insights of the past can actually make a big difference for our understanding of democracy in the present.

Here’s an illustration. Foa and Mounk explain how the assault on democratic norms in Hungary and Poland help make their case for the loss of faith in democracy. Hungarian Prime Minister Viktor Orban, for instance, quickly managed to consolidate his power in 2010 by replacing the constitution. Since then, there has been a steady illiberal slide in the country while Orban is showing increasingly autocratic instincts.

After eliminating various constitutional checks and balances, Orban’s government has gradually ground down the independence of the judiciary and worked to silence the press, mainly through self-censorship. Most recently, his government has passed legislation that threatens to shut down the Central European University in Budapest, in a further effort to stifle academic freedom.

In Poland, the nationalist-populist Law and Justice party (PiS) of former prime minister Jarosław Kaczynski took office in the autumn of 2015 and attacked the independence of its high court and the public media, almost following the same “script” that Orban’s Fidesz party deployed in Hungary. The PiS government’s assault on Poland’s democratic institutions has provoked an ongoing constitutional crisis in the country.

Yes, democratic norms are under pressure in both countries. But two things are worth noting: 1) this is not the first time that either leader has come to power; and 2) both countries are embedded in a wider web of European Union institutions, having been full E.U. members since 2004, as well as Western norms, having been NATO members since the mid-1990s.

Politicians undermine democracy

These two factors help us paint a rather different picture — and a more troubling one. As R. Daniel Kelemen argues, a frequent cause of democratic backsliding is the complacency of national politicians who need to build support from the ground up but aren’t too picky about how that is accomplished. A classic U.S. example would be the way Democrats at the national level ignored and even insulated Huey Long while he was governor of Louisiana from 1928 to 1932 (even if much of that insulation disappeared once Long left Louisiana for the Senate).

This phenomenon unfolds at the multinational level as well. So it’s not that European institutions lack mechanisms to reinforce democracy in Hungary and Poland — the problem is that European politicians would rather get Hungarian and Polish support for other projects. At a minimum, they seek to avoid facing Hungarian and Polish opposition. As long as the assault on democratic norms is not too dramatic, it is easy to avoid an overt conflict.

Kelemen’s argument helps explain the differences between what’s happening in Hungary and Poland. Orban’s Fidesz in Hungary is part of the mainstream center-right European People’s Party, which dominates in the European Parliament. Orban’s government has also moved slowly and systematically, pushing right up against the soft boundaries of democratic norms in its reform of key institutions.

This means European criticism of Hungary has been relatively muted. By contrast, Kaczynski’s PiS party in Poland has moved brusquely to challenge state institutions. For example, while Orban was able to entrench his party’s hold on power through legal constitutional amendments, PiS is blatantly violating the Polish constitution and crushing the high court, which is in charge of defending it. Given that PiS does not belong to an important European political group, European criticism of Poland has been much more confrontational and persistent.

There are parallels in the United States, too

Kelemen’s argument helps connect what is happening in Europe to what’s happening elsewhere in the world. Political elites can “do the right thing” from a democratic perspective — or they can stand back and watch as democratic norms suffer from negligence. This is as true in the United States as it is in Europe.

Consider the protection of civil liberties. Desmond King shows how the success and failure of the U.S. civil rights movement correlates directly with the intensity and consistency of U.S. federal intervention. When U.S. federal courts and the executive branch actively apply civil rights laws, minorities have benefited from strengthened protections. But where federal action has been withheld, those protections have rolled back as a consequence. This sounds obvious — and nothing magical — but this doesn’t make the consequences of the rollback any less tragic.

Our overall findings suggest that, if liberal democracy is indeed failing at the moment, we need to look at the combination of political inertia and institutional constraints. To understand the phenomenon, we should, therefore, focus less on public opinion and more on elite behavior.

This means we should ask which politicians are complicit in undermining democratic norms and what their interests are in doing so. We should inquire whether a country’s government is protecting the underprivileged and, if not, who really stands to benefit.

But we should also examine whether all citizens are brought equally into the system. The question is not just whether democracy is for everyone or only for the powerful, it is whether our political elites are using democracy in an inclusive manner — or whether politicians are trying to exclude voices they think are inappropriate or inconvenient.

Most importantly, democracy is not just about popular attitudes or controversial leaders. Instead, it is about everyone who has the opportunity and the incentive to influence democratic performance.

This concept of “democratic dysfunction” is more about governing elites and their use or abuse of existing institutions than it is about great leaders or apathetic masses. The great beauty of democracy is that everyone and anyone can aspire to this “elite” status.

But here’s the flip side of this argument: If democracy is what we make of it, we are the only ones to blame if we make a mess of it.

Erik Jones is professor of European studies and international political economy at Johns Hopkins University’s School of Advanced International Studies.

Matthias Matthijs is assistant professor of international political economy at Johns Hopkins University’s School of Advanced International Studies.