Book after book in recent years has alerted us—as if we couldn’t tell by reading the news and absorbing the panicked media—that democracy is in crisis. Did it start with Trump or with Brexit? In Europe or the U.S.? The diagnosis varies among authors of different backgrounds and political persuasions, as do their prescriptions on what to do now. Not all of the books even share the premise that the loss of democracy is such a bad thing—at least one recent work argues that the real crisis was a democratic surplus.What the books have in common, beyond their shared subject matter, is a common confusion over what democracy actually is. This confusion about the differences between democracy’s complementary functions like lawmaking and voting is, in its own way, rather illuminating, as the shared shortcoming in the books suggests a broader breakdown in democratic understanding that helped engender the very crisis they were written to address.Probably the best, and certainly the most talked about entry in the “crisis of democracy” catalogue, is Yascha Mounk’s The People vs. Democracy.Mounk was, at the time of the book’s release, a Harvard lecturer, and his well-timed tome has catapulted him into the world of celebrity academics. This clearly came as no surprise to his publishers at Harvard University Press, who put the book out with no fewer than four gushing blurbs from an exhaustive array of celebrity intellectuals ranging all the way from a Harvard AB (Dani Rodrik), a Harvard Ph.D. (Francis Fukuyama), a Harvard JD (Anne-Marie Slaughter), and all the way to a current Harvard professor (Michael Sandel).The book is organized into three parts. The first lays out the “crisis of liberal democracy,” the second endeavors to explain the origins of the crisis, and the third proposes a set of remedies that are remarkably moderate in scope relative to the severity of the crisis outlined in Part 1. Though Mounk denies it throughout, his book is clearly informed by a certain nostalgia for the postwar consensus of strong but limited liberal tolerance, a welfare state undergirded by a broad social solidarity, and a deference for cultural and political elites enforced by shared media that were occasionally publicly owned and nearly always publicly minded.It is the first part of the book that contains the most original and the most compelling argument, and it is this argument that has received the most attention. Mounk identifies two long-term trends in the practice of democracy that have the same theoretical point of departure, namely the decoupling of liberalism from democracy. First there is the much-ballyhooed rise of “illiberal democracy.” What’s nice about this term is that it is more or less understood by everyone, and is used by both political theorists and everyday commentators to mean the same thing. Even better, it is mostly free of judgment. People like Mounk who are terribly worried about illiberal democracy call it that, while Hungarian Prime Minister Viktor Orban uses the exact same term to champion his own political vision.The second trend that has Mounk’s attention is the mirror image of illiberal democracy. “Undemocratic liberalism,” in addition to being less catchy also lacks the obvious tie-in with the twin electoral traumas of 2016 in the U.S. and U.K. or with the fashionable concern with rising populism, whatever that term is supposed to mean now. The book’s title (The People vs. Democracy) and it’s even more dramatic self-helpy subtitle (Why Our Freedom Is in Danger and How to Save It) reference only one of the two trends. In the Trump-Corbyn era that’s clearly good marketing, but it’s also an injustice to a book and a scholar far more sophisticated than the cover lets on.When Mounk peers out at the state of advanced democracies today, he sees a rising tide of both illiberal democracy and undemocratic liberalism—and, perhaps most worryingly, how each trend feeds off and then fortifies the other. This is Mounk’s greatest insight and his fundamental contribution to the conversation on contemporary democracy.But the elegance of the argument gets the best of him, occasionally. Under illiberal democracy he includes phenomena such as populism, xenophobia, majoritarianism, and attacks on the press. His enumeration of the characteristics of undemocratic liberalism starts out strong, including things like judicial review and global civil society. But then it veers off into some sociological observations about political elites which, while cogent and mostly accurate, don’t seem to be particularly “liberal” problems. Finally, still under the rubric of undemocratic liberalism, Mounk gets to the perversions of money and corruption in politics. This, too, is undeniably important and clearly “undemocratic” about this, but I have no idea what precisely is “liberal.”There’s a beauty to an academic hypothesis built on two contrary macrotrends occurring simultaneously, but not everything fits gracefully into Mounk’s categories, and he appears, at times, reluctant to rework his organizational concepts to accommodate realities. The distortion of big money in politics is real, but it’s not really a case of illiberal democracy or of undemocratic liberalism. A more analytically robust presentation might lose some of the elegance, but that is a price worth paying.The same is true of the recurring two-by-two matrix which is supposed to serve as a handy visual aid but only ends up obfuscating. As an illustration of how liberalism and democracy are conceptually separate, a matrix with each concept on a separate axis could be helpful (though arguably unnecessary). Where things get tricky is that each box in the matrix has the name of a country (or even the European Union) in it.This all seems a bit shallow in comparison to the more careful argument that the text of the book is advancing. The problem with advanced democracy in the early 21st century is not that we are becoming too much like Poland or Canada or Switzerland (or some superficial stereotype of Poland or Canada or Switzerland). It is, rather, that certain putatively democratic forces are undermining the rule of law and at the same time certain putatively liberal forces are undermining popular sovereignty. This confluence of reinforcements manifests itself differently, to be sure, in Warsaw than in Brussels. But the subtlety of the argument, and the trap the status quo lays for democracy, is contradicted by a visual aid that ignores or even contradicts the thesis of the book.The rest of the book is dedicated to explaining the origins of the democratic crisis, which Mounk locates in the rise of social media, economic stagnation in general and social inequality in particular, and the growth of identity politics. All these trends are noteworthy (if not particularly earth shattering), but where we place them chronologically will determine how we separate cause and effect. The economic stagnation chapter, by far the book’s shortest, is a familiar lament about the decline of the social democratic order that, whether it went under the name New Deal or Postwar Settlement, underpinned mid-20th-century social solidarity. At times Mounk presents his views on these issues as controversial or daring, but in fact they are fairly conventional and likely shared by most of his readers.In Mounk’s analysis, the rupture in news-consumption habits that precipitated the democratic crisis begins with the rise in social media. This, however, may be the wrong framework. Clearly, social media and the internet have wrought profound changes in how we consume news and interact with current events, but they may still have been less harmful to democratic norms than the privatization of broadcast news and the rise of cable television that occurred decades earlier. Until the early 1990s, most television in the advanced democracies was publicly owned and publicly minded. Even in the U.S., the three main networks, while owned by private corporations, operated as civic-minded institutions, with news divisions often losing money (before remote controls this made sense as it drove viewers into more profitable programming after the news) and political content regulated by “equal time” provisions of the FCC.It seems like a small point, but it’s not at all. Surely all these changes had an impact and hunting for one inflection point—one year that supposedly “changed” our media world—is a kind of vanity. Except that if all that matters is Twitter and Facebook, then economic stagnation and new media are distinct problems to be understood separately and remedied separately. But if it was privatization and cable television that changed our relationship to the news and opinion formation, then maybe the media shift isn’t coincidental to the socio-economic changes of the last 40 years, because the inflection points on both trend lines are roughly the same. Maybe both outcomes are outgrowths of the same across-the-board undermining of the foundations of social solidarity. Privatizing media and eliminating gatekeepers went hand in hand with the anti-regulation market boosterism known on the right as libertarianism and on the left as neoliberalism that provided an ideological justification for growing income inequality and social self-segregation. In fact, it wouldn’t be hard to put the rise of “identity” politics under that rubric too.Our collective complacency about the well-being of our fellow citizens is just part of our complacency about our democracies. Perhaps seeing our institutions threatened so openly as we have recently will be the jolt we need to rise to their defense. Yascha Mounk passionately hopes so, and I hope he is right.But not everyone shares such hopes, and that includes at least one person among the handful publishing hot new takes on democracy. The most provocative and entertaining of this lot is without a doubt Georgetown professor Jason Brennan. His book, the pithily titled Against Democracy, contains not a single reference to Trump or Brexit, as it was published just before all that unpleasantness supercharged the conversation on democracy. Still, it’s safe to assume that nothing about those electoral traumas (or the ones that followed) has driven Brennan to rethink any of his conclusions.Democracy, for Brennan, is something of a tragedy. It asks people who are ignorant, misinformed, impulsive, and shortsighted to make fateful public decisions. It doesn’t actually empower people and doesn’t solve conflicts. If anything, it makes us hate each other more. These are bold and unfashionable claims but Brennan backs them up with a wealth of survey data and cognitive psychology.The key word running through all the various nuances of Brennan’s argument is “competence.” Voters lack competence to make political decisions, but citizens are entitled to something called “competent government.”But is any of this what democracy is about? Brennan never distinguishes between different kinds of competence. His book is full of amusing examples and hypothetical situations to illustrate his point. But it’s never clear in his examples if his arguments are about people as citizens, or authorized decision-makers as political actors, or institutions as such. He switches from masses to elites as the examples suit him. The book never adequately differentiates the competence to make important public decisions from the competence to participate in an election, but these surely are not the same skill set.There is an elementary distinction between democracy as a regime type or a social arrangement and democracy as the act of voting, but Brennan treats these two radically different (if occasionally coincident) practices and their different deficiencies, as interchangeable where it suits his argument. We might ask, for example, why nine Supreme Court justices vote on a decision. Is 5-4 really any way to make such essential interpretive decisions? Maybe they should deliberate to unanimity or be allowed to bundle unrelated decisions or, conversely, build majority coalitions for parts of decisions taken separately. All of these possibilities touch on democratic proceduralism but they do not fundamentally call into question democracy as a political practice. The rules of the Supreme Court could change in any number of ways but the underlying reality would remain that critical authority invested in unelected and unrepresentative judges.The extent to which citizens are competent to participate in general elections and to have their votes equally counted is a different discussion from how Supreme Court justices assemble a majority. Especially since the public at large is so rarely asked to actually decide on anything; it usually just appoints people by means of election.Voting is not the same as democracy. Ruling is not the same as electing. Competence is not the same as suffrage. Governing is not the same as lawmaking. And appointing, even by election, is not the same as representing. Yet, throughout his book, Brennan continually conflates or elides these differences. The last two are not unique to him. Mounk, too, never makes any allowance for the difference between public decision-making on routine everyday matters on the one hand, and the rather unique form of public decision-making that creates general norms on the other hand. Governments aren’t empowered to do whatever they see fit, just as the people at large in a functioning democracy are not. They are there to apply general norms to specific policy problems. To give just one dramatic example of the difference, the process for adding a new amendment to the constitution is considerably more difficult and involved than that which is necessary to pass new noise regulations in a public space. That governments and societies operate in conditions of law—even laws that are subject to change and revision—is a fundamental aspect of modern statehood that can’t just be chalked up to a straitened category like “undemocratic liberalism.”And if we are to understand what is special about law as such, we’ll need to understand what is special about the representative assemblies where laws are made. But this too barely features in Mounk and Brennan or any of the other contemporary discussions of democracy. Representation is rarely mentioned, and even then only as a practical convenience. No distinction is made between the process of appointing someone to a position of authority via election and sending someone on behalf of a voting public to a large plural body where he or she will deliberate and bargain and ultimately legislate (to say nothing of oversee the same authorized figures who are charged with governing).No theoretical engagement with democracy is complete without a starring role for both law and representation, yet both are absent from all four books under review here and from nearly all the rest I have encountered.It’s a deficiency that slips past the boundaries of political theory. Perhaps we have lost the capacity for thinking about the element of lawmaking in a representative assembly in our popular conceptions of democracy. It’s an alarming loss but recognizing it illuminates the source of some overwrought concerns as well as to some poorly thought out institutional changes.What happens when we lose this fundamental understanding of how democracy operates and what it is supposed to do? We might stop worrying about what laws can and can’t do. We might stop thinking of governments as complicated corporate entities that mix varying levels of expertise and accountability in their day-to-day operations, and we might forget that legislative assemblies are places where ritualized argument and collective decision-making on binding general norms take place. We might instead reduce our political lives to loud, sloganeering entertainment, and take complex fateful matters directly to the public for heedless one-off votes as though institutions, norms, negotiations, and compromise all have no place in politics. We might, in short, embark on something that would look like former Prime Minister David Cameron’s wild adventure with British democracy.Like every reckless gambler, Cameron started small and took the absence of immediate failure as a sign that he should keep raising the stakes. He unleashed on the British public three referendums in the space of five years each of which had the potential of upending the British constitution in dramatic fashion. Each vote arose as a way to appease a coalition partner or quell a potential internal schism. When, in 2011, voters rejected the Alternative Vote electoral reform referendum to change the rules for electing the House of Commons, Cameron was emboldened to try a referendum on Scottish independence in 2014. When that prospect was rejected as well, Cameron put all his chips on the Brexit referendum.Had he succeeded a third time in using a sloppily worded and hastily crafted referendum to tamp down a noisy domestic insurgency, Cameron might have bought the British establishment, including his own Conservative Party, a decade or more of quiet. Or maybe he would have kept gambling and held a referendum on the monarchy or nuclear disarmament or disestablishment of the Church of England or a switch to a presidential regime. We’ll never know.Nor, thankfully, will we know quite how risky the Scottish referendum was. A shift of 5% in voter preferences would have given Scottish independence a majority, rending the union with no clear roadmap. The U.K. would have committed itself to redrawing its borders and reconfiguring all of its international defense and treaty and trade relations based on a vote in a referendum in which 91% of U.K. citizens (those residing in England, Wales, and Northern Ireland) were ineligible to participate—and where, weirdly, Scottish 16-year-olds did have the vote. It’s not the case that the Scottish government and the U.K. government negotiated a treaty for independence or federalism or something in between and then brought it to the voters for approval. It would have been, like Brexit, a binding resolution to do something drastic where no one really knows what that something is.Instead Britain is tearing itself apart over the result of an equally ill-considered referendum, and every option from here on out is bad. Ignoring the result would be an assault against the people’s will. A new referendum is almost as insulting, and none of the people proposing such a course of action can agree on how to word it or how many options there should be (approve the deal, leave without a deal, remain after all, etc.). Crashing out would be an economic disaster. A soft departure from the EU satisfies no one, as it would necessarily involve the worst aspects of both remaining and leaving with the benefits of neither. And negotiation is impossible when Britain self-declared endgame leaves it with no leverage. Government ministers pursue a policy they know is wrong, opposed by an opposition just as committed to that disastrous policy. Forty-eight percent of the public may have opposed Brexit in the referendum, but that voice has very few options to express itself in parliamentary elections. How did we get to this?The Brexit referendum, in the view of Cambridge politics professor David Runciman, shows “how easily the popular demand for more democracy can end up having the opposite effect.” His new book, How Democracy Ends, is the only one of the recent crop to take institutions and not just procedures seriously. But even Runciman’s work never seriously engages with the twin problems of law and representation—the latter a particularly disappointing oversight as he co-authored a short book on the topic in 2008.It is a fun read, though, avoiding the alarmism that generally characterizes the genre. The exclamation marks on the chapter titles (“Coup!” “Catastrophe!” “Technological Takeover”) are the principal reminders that the issues at stake are deadly serious.Runciman never resists the temptation to make clever counterintuitive claims, but most of these are then backed up with reasonably plausible arguments. He argues, for example, that Trump’s electoral victory should be seen as a vote of confidence in the institutions of the American constitution because, if anyone was genuinely worried that Trump could actually get his way he could never have assembled a winning coalition. Either there is a “safety net … or the whole thing is a sham.” Either way, the risks are smaller than advertised.While everyone else is implicitly focusing their trepidation on comparisons with the 1930s, Runciman asks us to look back even further to the golden age of populist democracy in the 1890s. It was a decade of conspiracy theories, unpopular wars, immigration panic, and financial crashes that eventually gave way to a decade of democratic reform. In this reading of history, elected politicians were forced to confront and assimilate populist anger by expanding the franchise and erecting the rudiments of the welfare state. Populists came close to undermining constitutional government in the U.S. and France (and possibly Britain, too), but were beaten back when their anger was challenged into a progressive agenda. It’s one possible outcome, but clearly not the only one.Runciman begins his book at the H-hour of today’s democratic crisis, high noon on Jan. 20, 2017, with the inauguration of President Trump. But his sweep is broad as he takes us through all the conceivable ways democracy might (and maybe even should) end, whether by coup, catastrophe, or unplanned consequence of technology. Like Mounk, he sometimes adheres too rigidly to his own categories, and the metaphor of democracy’s “midlife crisis” is fun the first time the reader encounters it, but loses some of its spark with each repeated use.If democracy is having a midlife crisis, then John Dunn didn’t wait for the flashy sports car and hair piece to appear before diagnosing it. Dunn, an emeritus professor of political theory at Cambridge, adapted a lecture series he gave at Yale into a short and alarming book on democracy back in 2014. If that seems hopelessly out of date, think again. Almost everything about our present crisis is spelled out patiently and urgently in Dunn’s Democracy: A History.Where the other books want us to look at the new authoritarians in the democratic countries, Dunn broadens his geographical and historical scope. If we want to unpack the connection between democracy and good government—“happy accident” or “magic formula”—we need to look at evolving institutional practices in decidedly nondemocratic China and in Asia’s democratic behemoth India.Dunn’s book is very much a pre-2016 work, but a post-2011 one, too. His concerns about democracy’s direction don’t have the bitterness of political disappointment and aren’t obsessively steeped in the latest tweets or crises. Dunn fretted about democracy before it was cool. Though China and India get a lot of attention in the book, it is the Arab Spring, only casually mentioned, which casts a dark shadow that lingers over nearly every page.Dunn asks us to decouple our notions of good government and democracy and consider that much of what we positively attribute to democracy may be little more than a historical coincidence, bound in place and time.His is an elite project with no catchy subtitles or mass audience in mind. His anger, in the book’s final pages, is not directed at Twitter but at universities. It’s the universities, in Dunn’s reckoning, that have failed to ask the difficult questions about our political arrangements and their evolution.If that is the case, as I strongly believe it to be, then maybe our trauma as scholars of democracy isn’t 2016 or even 2011, it’s 1989 when the Cold War ended. A shocking failure against the Soviet Union might have caused some soul-searching but, instead, success bred a smug complacency that dulled any interest in democracy and set political theorists off on a hundred other pursuits (global justice and rights being the biggest). It all seemed so easy at the time but history has made a mockery of such pretensions. By the early ’90s, Liberal democracy was destined to take over the whole world, but in reality failed to take hold in most of the post-Soviet space, while the few central European success stories today constitute the go-to examples of illiberalism riding into power on democratic horses.When regimes fell or reformed in other theaters, there were very few political theorists on hand to offer lessons on what to avoid. To take perhaps the most glaring example, no one seems to have seen fit to warn of the dangers of a presidential regime as it careened toward its first free election. The idea of handing over all executive power to a party that wins just over half the votes in a country almost evenly split between ardent supporters of Islamist fundamentalism and terrified opponents of it would have been bad enough in a country that had a tradition of repeated electoral contests. In a country that had none, this was a guarantee that the first election would be the last.Nor were we as alert as we should have been to the insidious encirclement of our best practices in the established liberal democracies, in particular the hollowing out of the powers of representative assemblies and the concomitant rise of executive and judiciary power. Without a firm grasp on what laws are supposed to do and what governments are supposed to do (not the same thing!) we couldn’t see the problem with handing legislative prerogatives to civil society and international institutions and subjecting crucial public decision-making to increasingly authoritarian leaders or the frenzies of privately mobilized public passions.Political thinkers have come to see governance as yet another market mechanism where preferences can be aggregated and efficient outcomes determined. If people disagree, they must be ill-informed or prejudiced. But disagreement is a fundamental condition—the fundamental condition—of politics. And the most important practice of democracy isn’t voting, but rather proscribing norms for the habit of legitimate disagreement. Representative assemblies, with their large numbers of diverse members and their ritualized speech acts and decision-making rules engender these habits, particularly when their proceedings and decisions are at the center of public attention. Twitter can’t replace that; cable news can’t replace that; referendums can’t replace that; liberal high courts, international organizations, pious human rights groups, and the free market can’t replace that; and authoritarian populists certainly can’t replace that.To measure democracy against some Athenian ideal, or to criticize it in those terms, is to miss the point. We are not Athenians, not because we can’t be, but because we don’t want to be. Our smartphones, our social media apps, our almost unlimited access to information and platforms have all given us the means to subject our politics to a daily public decision-making process free of all gatekeepers and constraints. And yet we still, most of us at least, prefer to live under the rule of laws. We prefer it even as we lose our grip on the democratic institutions that have historically yoked the most powerful members of society to the same laws that the rest of us must abide. Law can’t just be a nondemocratic means for imposing a liberal agenda, if only because it will eventually become a nondemocratic means for imposing an illiberal agenda.As long as our discussion of democracy treats representation as though it were some kind of rounding error or approximation of the real thing, and law as though it were just another policy outcome, we won’t have a fully formed sense of what this democracy thing really is—or, as the subtitle of Mounk’s book would have, how to save it.***Like this article? Sign up for our Daily Digest to get Tablet magazine’s new content in your inbox each morning.