At the beginning of 2018, the Dutch social scientist Cas Mudde made a prediction: 2017, he observed, had been the year when academics competed to explain the seemingly unstoppable populist wave that had resulted in (to name just two examples) the Brexit referendum in June 2016 and Donald Trump’s election that November. In the coming year, the message was going to be: “Democracy is dying, but you can save it… if you buy my book.”

Now, in 2019, we can safely say that his prediction has proved right. Apocalyptic talk of authoritarianism abounds, and a veritable democracy-defense industry has emerged. Dozens of books—The People vs. Democracy; Can It Happen Here?; Fascism: A Warning; How Fascism Works—fill the publishers’ catalogs, detailing democracy’s sadly dimming prospects. Their authors mostly make predictions on the basis of historical analogies—it’s the 1930s all over again—or extrapolate from recent authoritarian takeovers in countries like Russia or Turkey. But how plausible are such reference points? Of course, ideally everyone should want to learn from the past, but easily prepackaged “lessons from history” or forced analogies with countries that have never had a proper liberal democracy pose the danger that we will fail to grasp precisely what is peculiar about our age.

Of all the books that this new democracy-defense industry has produced, Steven Levitsky and Daniel Ziblatt’s How Democracies Die makes the most coherent case, by way of comparison, for why Trump’s presidency may well endanger one of the world’s oldest republics. As scholars who have worked primarily on Latin America and Europe, Levitsky and Ziblatt demonstrate how a global perspective should shake many people out of the complacency created by their cherished beliefs about American exceptionalism. Like all students of comparative politics, they are mindful that 1990s Venezuela, post-1945 West Germany, and interwar Belgium—all of which make an appearance in their book—differ profoundly. And yet they think one lesson can be generalized: that democracies depend not just on institutions like courts committed to protecting the rule of law; they also require informal norms that all political players need to observe to keep the democratic game going. Like many liberals, they think that serious norm violators such as Trump should be kept out of the game entirely, and so they call for reinforcing the power of elite gatekeepers as a powerful line of defense.

In his similarly titled How Democracy Ends, David Runciman, the most original theorist of democracy writing in the United Kingdom today, provides a convincing alternative to the products that come off the assembly line of the democracy-defense industry. The Cambridge professor is deeply wary of historical analogies. He worries that, by becoming fixated on fascism and other instances of democratic self-destruction, we will miss today’s real challenges—the catastrophe of climate change, above all, but also how social-media networks are undermining democracies in subtle yet potentially fatal ways. Democracy, Runciman says, is about keeping the future open and enabling people to change their minds after encountering different views and new information; the Internet giants, by contrast, profit from always giving us more of the same. Combine the power of algorithms with a state committed to all-out surveillance of its citizens and you get contemporary China, an authoritarian model that Runciman regards as a serious rival to democracies today.

Levitsky and Ziblatt’s book starts with what by now should be an uncontroversial observation: Democracies do not necessarily go out with a bang (as in the case of a military coup); they can also end with a whimper. True, coups have not disappeared altogether—think of Egypt and Thailand in recent years—but our situation is clearly different from the Cold War era, when coups d’état accounted for nearly three out of every four democratic breakdowns. Instead, today we face what the democracy-defense industry variously calls democratic “erosion,” “decay,” or “backsliding.”

The choice of these metaphors to describe democracy is not neutral; the terms suggest quasi-natural processes, when in fact democracy’s demise is the result of many conscious decisions. Levitsky and Ziblatt also use these metaphors, but they point the finger at particular people; for them, democracy’s breakdown is about elites abandoning the norms needed to hold politics together. They provide a checklist of what these norms are and when their trespass should set off alarm bells: Do politicians reject the rules of the democratic game—for instance, by questioning the legitimacy of their opponents’ winning an election? Do they deny the legitimacy of their rivals altogether? Do they tolerate or encourage violence in politics? And do they threaten to curtail the liberties of their political adversaries and possibly also of the media?

No prizes for guessing who checks off all four of the items on Levitsky and Ziblatt’s list. Trump made it clear enough that he would not be ready to accept the legitimacy of a Hillary Clinton victory when he promised voters, “I will tell you at the time.” As president, he has consistently demonized opponents, encouraged brutality against demonstrators at his rallies, and attempted to restrict the rights to political participation—most obviously through phony claims of electoral fraud in order to legitimate voter suppression. Levitsky and Ziblatt rightly insist that Trump is not a lone demagogue who came out of nowhere; instead, they argue, he was produced by a culture of “extreme polarization” that created an environment in which the normative “guardrails” of democracy were first loosened and now might get broken off altogether. In particular, political actors no longer accept the legitimacy of the other contenders for power, and they cease to exercise “forbearance”—a willingness to not always push institutional prerogatives to the limit.

Levitsky and Ziblatt recognize that polarization in America is not symmetrical. Only Republicans have, in living memory, denied a hearing to a Supreme Court nominee with a view to capturing the Court, in defiance of long-standing norms. And only Republicans and conservatives have made it their business to deepen cultural and even racial divisions through what the authors call a “conservative entertainment complex.” Yet Levitsky and Ziblatt do not just blame right-wing elites; they also blame the people. Give citizens a chance to participate in primaries and let them speak their minds on social media, they warn, and democracy might go to hell. In the absence of gatekeepers—party leaders and professional journalists who uphold democracy’s norms—the great unwashed could destroy the very machinery that enables self-rule.

This unashamedly elitist account sits uneasily with the book’s underlying narrative about the republic’s decline and possible fall. Levitsky and Ziblatt argue that there was a long stretch in 20th-century US history when norms were broadly observed—until the passage of civil rights made the South go Republican. Of course, racist attitudes were not created ex nihilo by elites, but the strategy to consistently strengthen them was a choice that predates Trump by a considerable period of time—and it was made by elites, including icons of responsible centrism like the Republican patrician George H.W. Bush.

Contrary to the clichéd talk of the United States as one of the world’s oldest and most enduring democracies, Levitsky and Ziblatt argue that we should think of the South in the 20th century as something almost akin to what the independent watchdog organization Freedom House calls a “country in transition” to democracy—and that therefore has all the attendant problems of transitioning. Hence, while Levitsky and Ziblatt insist, on the one hand, that America’s democratic guardrails have been dangerously weakened, they implicitly argue, on the other, that the earlier norm compliance depended on the country not being properly democratic to begin with. The emphasis on racial inequality here, however, also casts doubt on the usefulness of comparing the United States with countries that do not have a legacy of slavery and Jim Crow. Its omnipresence today, too, also makes our future look much bleaker. As Levitsky and Ziblatt note, “few societies in history have managed to be both multiracial and truly democratic.” Their concern is vindicated in that, quite apart from Trump, Republicans appear to be committed to occupying institutions like the Supreme Court and, if necessary, bending or breaking norms, all in order to defend an old order against an ever-stronger Democratic majority.

In the end, Levitsky and Ziblatt have no real answer to this challenge. The only thing they are absolutely sure about is that norm-breaking should not be countered in kind. From a strategic perspective, this is far from obvious—sometimes only tit-for-tat can encourage a return to compliance. Nancy Pelosi’s recent hard-nosed response to Trump’s antics might also point in that direction.

It is a mistake to think that democracy only works if there is niceness or consensus about observing norms—as if it were okay to run an apartheid state, so long as polite Southern gentlemen do it. What really matters is not how to prevent conflicts; rather, it’s how to understand them. Though the commentariat constantly laments that the US is so “divided,” it is important to remember that democracies exist precisely so that we can deal with our divisions. The question is whether a political battle can be waged without denying the legitimacy of one’s opponents, and it is here that a president who maligns his critics as “un-American” has done so much damage.

There is also more room for “the people” than Levitsky and Ziblatt may allow. The midterms have made it plausible to believe that the answer to democracy’s problems is not to reempower the gatekeepers—after all, the gatekeepers on the right caved to Trump and keep on enabling him—but rather to get more people to participate. Last November’s election and the impact of the Democratic freshmen in Congress also suggest that we don’t all have to be actors in Trump’s brand of reality-TV-style political theater.

In How Democracy Ends, David Runciman suggests as much: “If Trump is the answer, we are no longer asking the right question,” he declares. Runciman is politely dismissive of much of what the democracy-defense industry is producing today. He sees no real danger of fascism returning; after all, “our societies are too different—too affluent, too elderly, too networked—and our collective historical knowledge of what went wrong then is too entrenched.” That might sound complacent, but Runciman is right that nowhere in the West do states propagate the same kind of systematic cult of violence and racism that was at the core of fascism in the 20th century. We are also not witnessing a comprehensive mobilization, let alone militarization, of societies (though Brazil’s Jair Bolsonaro, who also checks all the items on Levitsky and Ziblatt’s list, will likely become a test case for this proposition).

Yet Runciman also cannot entirely resist the historical-analogies game. For him, the relevant comparison is with the Gilded Age at the end of the 19th century, which saw rapid technological change, rising inequality, and plenty of populism. But he only makes the comparison to highlight how different our own times are. During that period, and the Progressive era that followed, there was still what he calls “slack” in the political system, and so a generation of social democrats on both sides of the Atlantic were able to extend the franchise, enlarge the tax base, and, above all, deepen and expand the public’s sense of trust in the state. Today, that slack simply isn’t there, and as a result, citizens see political institutions as unresponsive.

By contrast, for Levitsky and Ziblatt there is still plenty of slack, and democracy remains to be fully realized in the United States. But for Runciman, democracy is tired and worn out. This exhaustion is not a terminal condition, he notes, but rather the misery of a midlife crisis. People are no longer energized by the prospect of change in which all might gain; if citizens act at all, they do so only to get something back that they feel they’ve lost—which means a great deal of energy goes into blaming those who were allegedly responsible for the losses.

So what is the takeaway for Runciman? He argues that liberal democracies certainly can make the most of their middle age, but that they cannot expect to get their youthful enthusiasm back. If they try, they might end up with a Trump, who appears as the political equivalent of a motorbike for middle-aged men.

Yet Runciman’s analogy to middle age does inescapably imply that death is on its way, and he sketches three main scenarios for democracy’s demise. One mortal threat, he argues, is climate change, which he says lacks an immediacy to inspire the necessary political action for combatting it. It is indeed telling that the Yellow Vest protesters in France complain that Emmanuel Macron, in order to justify raising green taxes on gas, can only talk about the end of the world—whereas they worry about the end of the month. This does not mean that democracy will perish anytime soon; it is just, as Runciman laconically puts it, that climate change is too much for democracy to cope with, but also not enough to kill it off.

Another challenge, Runciman argues, is Facebook and the rise of social-media networks. He really does believe that the Internet changed everything, and that social networks might become a serious rival to the state. This seems implausible at first sight: Facebook’s power is connective rather than coercive, and unlike the Leviathan, sword in hand on the cover of Hobbes’s foundational book, Mark Zuckerberg merely wields the smartphone. The former promises security; the latter only “likes.” But in a manner comparable with Hobbes’s sovereign, Facebook hoards its authority: Its horizontal networks also come with a steep organizational hierarchy, and there is no accountability to democratic institutions.

Above all, Runciman argues, Facebook’s challenge to democracy is epistemological. The Internet giants want to get to know us so that they can always give us more of what we want. By contrast, democracy is a form of institutionalized uncertainty: You cannot know what will happen in elections and political developments more broadly. This openness is a source of strength; it means that democracies can learn and adapt. The Internet, despite all its promises about gathering the world’s knowledge at our fingertips, actually closes our minds by tailoring that knowledge to reinforce our political biases.

Facebook exercises a soft, subtle despotism, and in one respect, the third major threat that Runciman identifies is comparable with it. China is increasingly presenting itself as a force for global harmony; Runciman thinks that this market-friendly dictatorship is a genuine rival to democracy. The latter used to promise a combination of dignity for the individual—one person, one vote—plus collective benefits in the form of things like the United States’ Social Security or Britain’s National Health Service. China, by contrast, makes no promises to the individual other than the chance to enrich oneself. True, Chinese citizens also want recognition, but they get it in the form of nationalism and great-power status. This collective dignity is perfectly compatible with individual political exclusion, and Runciman wonders: If democratic states cannot deliver collective goods, will individual respect for democracy be enough to help it survive?

Runciman doesn’t provide a direct answer to this question, but he does pose another that he believes might serve as a possible response: Where would you rather be when something goes wrong? Yet the idea that worst-case scenarios bring out the best in democracies seems hard to square with Runciman’s fretting about democracy’s incapacity to deal with climate change. Those who think that an intellectual Cold War with China is now on will find precious little ammunition here.

But maybe that isn’t the point. Runciman, in his urbane, self-deprecating way, admits at the end of his book that he has no solutions. How Democracy Ends could not be more different from the sometimes rather authoritarian-sounding political self-help manuals of the democracy-defense industry, which proclaim: “This is what you must do to save democracy, as dictated by history.” Runciman instead only suggests what we should think about, not what we should think, and his book is an impressive exercise in political imagination, even if it sometimes comes out as pure speculation.

After all, how does Runciman—or anyone else, for that matter—really know whether his assertion that “democracy is over the hill” is actually correct? One might argue that Trump is not the political equivalent of a midlife-crisis motorbike; instead, he’s the product of a youthful lapse in judgment that, once its costs become painfully clear, at last concentrates the mind and makes democracy more mature—not in the sense of, say, finally appreciating the great norms and institutions that our parents left us (that would be the mainstream liberal intuition written up so well by Levitsky and Ziblatt), but in understanding how, in some ways, we have never been serious about some of our democratic ideals to begin with, and that we need to be so now.

Runciman’s volume will be of interest long after Trump has left the presidency, and long after the democracy-defense industry’s products have been consigned to the scrap heap. Its only real weakness is a perhaps all too British penchant for being contrarian for contrarianism’s sake. Plenty of paradoxes and arresting phrases (things like “Al Gore did not invent the internet—Gandhi did”) are never really explained and might better have not left the senior common room. But that’s a small price to pay for a brilliant book.