The inestimable E.J. Dionne Jr. describes How Democracies Die as an “urgent book” that offers “a brilliant diagnosis of the most important issue facing our world.” Certainly, the topic is an urgent one, in this age of Trump, and the book is definitely interesting and provocative. But it is also deeply disappointing. It might be compared to picking up a recommended book on Alzheimer’s disease in anticipation of reading about an announcement of a genuine cure and finding instead a checklist of symptoms and some hortatory advice on how to stave off further decline.

Levitsky and Ziblatt teach at Harvard, where Levitsky specializes in Latin American politics and Ziblatt in European. That explains why they draw their central examples of democratic decline (or even death) from that region. We read a great deal about Hugo Chavez in Venezuela and Alberto Fujimori in Peru, both of whom created havoc in what had been reasonably democratic countries. Juan Peron’s Argentina also comes in for some discussion, as do current leaders in Ecuador and Bolivia; there is relatively little discussion of Brazil, perhaps the most important country in South America, which is certainly undergoing its own turmoil. The central point, though, is to offer lessons that can be drawn from looking at these (and some other) countries. They concentrate not on autocrats who come to power through military coups or revolutionary takeovers of existing governments, but, rather, on leaders who successfully navigated the existing political system in order to achieve the highest offices available via democratic means. Subsequent to election, the autocratic leaders in question consolidated their power and fundamentally changed the tenor of the constitutional order, distinctly for the worse. The most important lesson, perhaps, is that “institutions alone are not enough to rein in elected autocrats.”

Lurking beneath their argument, though unfortunately undeveloped, is the possibility that presidentialism itself, in contrast to parliamentary systems, generates a tendency toward autocracy. This theory is associated with Juan Linz, the distinguished Yale political scientist who died in 2013. Linz compared prime ministers to presidents. Prime ministers, however powerful, are always vulnerable to the discontent of the coalition that put them into power. Think only of the unceremonious firing of Margaret Thatcher, perhaps the most important peacetime British Prime Minister of the past century, by members of her own party in 1990 when they decided she was no longer an electoral asset.

Presidents, however, are on an entirely separate track. With fixed terms, they are relatively invulnerable to legislative unhappiness, save for the prospect of impeachment, which, at least in the United States, has been resorted to very rarely. Moreover, presidents across the world are often able to take advantage of modern communications techniques to make themselves the center of political attention. If coupled with inchoate “charisma,” every president thus becomes a potential autocrat in a way that is far less true even of powerful prime ministers.

If one accepts Linz’s view, the answer might require quite drastic changes in the American constitutional system and not, for example, simply the election of a “better” President in 2020. Trump’s own autocratic tendencies can be shown to build on many of his predecessors, both Democratic and Republican. Levitsky and Ziblatt, for example, are generally quite critical of Franklin Roosevelt on the occasions they mention him, for violating the unwritten norm of the two-term limit and then for engaging in highly controversial uses of presidential authority in the realm of both domestic and foreign affairs. It has become a notorious truth that recent presidents especially, in often justified frustration about an ever more dysfunctional Congress, have pushed the envelope (and then some) of executive power. But the “autocracy” that so worries Levitsky and Ziblatt is more than the simple pushing of envelopes and consequent potential overreach. Presidential character and psychology count for a great deal.

Early on the authors present a checklist of actions that should serve as “behavioral warning signs” of autocratic tendencies: “1) rejecting, in words or action, the democratic rules of the game, 2) denying the legitimacy of opponents, 3) tolerating or encouraging violence, and 4) indicating a willingness to curtail civil liberties of opponents, including the media.” What makes their book so important—and widely discussed—is the question of whether these warning signs are present with regard to the current President of the United States. Their answer, as one can easily imagine, is “yes.” “A politician who meets even one of these criteria is cause for concern,” they assert, but Donald Trump in fact meets all four. We should therefore be very concerned. But their critique goes well beyond personalities. Like Thomas Mann and Norman Ornstein in their 2012 book It’s Even Worse than It Looks, Levitsky and Ziblatt are scathing in their description of an accommodationist Republican Party that has basically rejected what they describe as the two most fundamental requirements of a functioning democratic political order: first, a genuine willingness to tolerate oppositional views and their proponents and second and crucially, an equal willingness to forbear from using all of one’s legal power when that would predictably generate widespread opposition and resentment.

Every practicing lawyer knows that insistence on exercising all of one’s legal rights is almost a sure sign of a precarious relationship, whether one is thinking of a marriage, a business partnership, or, in fact, governments in two- or multi-party states. For example, as a constitutional lawyer, I think that the Republican majority in the Senate had the legal authority to behave like louts and refuse even a hearing to President Obama’s nominee to succeed Antonin Scalia on the Supreme Court. But I equally believe that it was one more sign of the decay of the American constitutional order, which does, as Levitsky and Ziblatt argue, depend on compromise and accommodation even if, by stipulation, one would have a right to do otherwise. Yale Law School professor Jack Balkin has coined the term “constitutional rot” to refer to a number of recent developments in American politics. He is referring most basically to the willingness to shatter long-established norms of what Levitsky and Ziblatt call the “unwritten rules of American politics” in order to achieve what James Madison would have denounced as the narrow and selfish goals of one’s own “faction.” The GOP treatment of Judge Merrick Garland—independent of whether they had a duty to vote for his confirmation after holding traditional hearings and listening to his supporters—exemplifies such rot, even if was not illegal or “unconstitutional.”

There are valuable things to be learned in How Democracies Die. Yet I came away feeling disappointed. Some of it derives, no doubt, from the fact that this is a relatively slim book, only 231 pages of text before footnotes. Although written by academics drawing on their deep expertise in comparative politics, it has a tendency toward the superficial, leaving a significant number of unanswered questions in its wake.

Consider a basic problem presented by the title itself. What exactly constitutes a “democracy”? Every President from Woodrow Wilson to Barack Obama has committed the United States to some version of a worldwide “democracy project.” That is after all why, roughly a century ago, the United States entered the catastrophic war in Europe under the rubric of making “the world safe for democracy.” But what is entailed in supporting “democracy” and, therefore, being able to describe a given country as becoming more, or less, “democratic”? Consider Winston Churchill’s oft-cited quote that “democracy is the worst form of government except for all those other forms that have been tried from time to time.” The problem is not that Churchill is necessarily wrong, but, rather, that there are so many different forms of government that can, under one plausible theory or another, be described as “democratic.” This means that they can be compared with one another, favorably or unfavorably, depending on the metrics that one focuses on.

Is the United States a democracy? Yes, we’re taught; but as a matter of fact, democracy was not a word favored by those who framed our own Constitution in 1787; many were explicitly critical of the term, which for them represented rule by the mob. Benjamin Franklin famously wondered if we could keep the “republic” that had been envisioned in Philadelphia, and Article IV of the Constitution purports to guarantee every state within the Union a “Republican Form of Government.” A “republic,” argued Madison, would ultimately trace its provenance back to “We the People,” but, crucially, they (or we) would play an astonishingly small role in actual governance, which would be done by “representatives.” As he wrote in Federalist 63, “The true distinction” of the new system of government created in Philadelphia “lies IN THE TOTAL EXCLUSION OF THE PEOPLE, IN THEIR COLLECTIVE CAPACITY, from any share” in actually making specific decisions of policy. (The capitalization, incidentally, is Madison’s.) The Constitution, though “ordained” in the name of “We the People,” unlike almost all subsequent state constitutions written in America (and many new constitutions written abroad), deprives the people of the ability to engage in even a scintilla of what is often called “direct democracy.” Consider only the initiative and referendum found in many countries and in at least 26 of the states within the United States (the most famous, or infamous, being California). Or think of what American politics might be like at this very moment if, as in California or Wisconsin, the electorate could initiate a recall election of the President. There are 51 constitutions within the United States alone, and it is easily possible to describe the United States Constitution as the least democratic of all of them. Does this matter?

Obviously, there is a vast academic literature on this question, though one gets no hint of its existence—and of the controversies embedded in the literature—in How Democracies Die (or, perhaps, transform themselves). What we get, instead, is an almost treacly concluding encomium to “democracy” written by the essayist E.B. White. I have my own dog in this hunt inasmuch as I published a book in 2006, Our Undemocratic Constitution, that set out the significant deviations of the U.S. Constitution from any plausible twenty-first century notion of democracy. Similarly, Robert Dahl, the greatest political scientist of the late twentieth century, published an important book in 2001, How Democratic is the American Constitution? His answer was not very; Dahl especially stressed the almost bizarre Electoral College and, just as importantly, the totally indefensible United States Senate, the most malapportioned legislative chamber in the entire world. Moreover, and more importantly, one can argue that these are not only offenses against democratic political theory, but also at least partial explanations of the widespread alienation that most Americans, of all political persuasions, feel from the national government.

We have Trump, just as we got George W. Bush, because of the Electoral College. And one cannot possibly understand the recent tax bill without grasping the importance of the fact that the majority of Americans, who live in the largest nine states, have a total of 18 senators, while the minority that live in the 41 remaining states have 82 senators. Some of the features in the tax bill can be explained in terms of standard plutocracy, support for constituency interests, or even, possibly, sensible revisions of tax policy. But the decision to go after deductibility of state income and property taxes, for example, has nothing to do with protecting constituents or even lining the pockets of contributors. Instead, it represents a new twist in the American culture wars and polarization, in which those who basically hate the cosmopolitan states where most Americans live actually seek to cripple their ability to fund important social services. This return to a raw form of sectionalism is yet another way in which one can see similarities between our contemporary polity and that of the 1850s. But the main point is that the Senate, unlike the House, whatever its own problems, provides no real protection for the majority of Americans who live in a small number of states. The House obviously has its own problems, most often analyzed in terms of partisan gerrymandering, but in fact derived from the requirement, thanks to a congressional statute of 1842 (and not the Constitution), that representatives be elected in single-member districts. This often ensures that millions of Americans simply feel unrepresented because the winners in these single-member districts are concerned only with their fellow political partisans—their “base”—and not at all the opponents who may well present no serious electoral threat. We would be far better off adopting some form of multi-member districts with proportional representation.

Perhaps most Americans can ignore the indefensible Senate, the consequences of our electoral system for the composition of the House, the Electoral College, or the near impossibility of amending the Constitution and be indifferent to the fact that the constitution of the state within which they live is considerably more democratic. Most defenders of what we think of as “American democracy” would focus on the fact that we conduct regular elections and enjoy the peaceful transfer of leadership. Some would point to the surmounting of the Great Depression in the 1930s and then the successful prosecution of World War II as added reasons simply to celebrate the Constitution. Those are no small matters, but do they suffice in the twenty-first century? Trump and his Republican minions are not threatening to cancel the upcoming midterm elections, about which Democrats feel a considerable optimism. In fact, one can argue that the GOP zeal to pass some of their most reactionary programs is triggered by justified fear that they will return to the state of a minority party in 2019. For most of us, though, that is scarcely enough to overcome our foreboding.

Is it possible, for example, that one reason for the rise of Trumpismo is the justified belief by millions of Americans that the national Constitution designed in 1787 creates so many veto points—which we admiringly and often unthinkingly label as “checks and balances”—that it regularly generates legislative gridlock with regard to addressing any of what one might believe to be our most challenging problems? Wherever one is located on the political spectrum—left, right, or center—one cannot possibly believe that Congress will adequately meet the challenges we face today. There is a good reason, for example, that the overwhelming majority of Americans have little regard for Congress and believe that the country is headed in the wrong direction. And a Democratic victory in 2018 will guarantee only the ability to block the worst GOP programs; in no way will it portend the ability of even a Democratic House and Senate to pass their own legislation and surmount a Trump (or Pence) veto.

The authors state that the real protection of American democracy, such as it is, lies not in “Americans’ firm commitment to democracy, but, rather, the gatekeepers—our political parties.” It is the “normalization” of political conflict between two large-tent parties that has allowed our history—with the obvious and all-important exception of 1861—to feature the peaceful transfers of power among those who would govern us. But, as already suggested, something is now rotten in the state of our party system, particularly within the Republican Party. I write this review on the first day of the first government shutdown in American history to occur when the same political party notionally controls both Houses of Congress and the presidency. Perhaps one wishes to blame the Democrats for the failure to compromise, but that failure gains purchase only because the Senate operates under the rule that the votes of 60 senators are, with few exceptions, necessary to move a bill forward to a vote. Levitsky and Ziblatt discuss this and take note of the fact that recourse to filibusters has now become a standard part of the way the Senate operates, but what they don’t do is offer a systematic discussion of whether we are overall well or ill served by this particular aspect of “American exceptionalism” that disallows majorities in the Senate from actually governing. Of course, one has to recognize that there may often be no correlation between majorities in the Senate and majorities in the national population, given the fact that Wyoming counts the same as California and Vermont is the equal of Texas.

Moreover, one would like considerably more discussion of the limits of “toleration and forbearance.” The authors admirably recognize that the victims of such “forbearance” in the past have often been vulnerable minorities, most notably African Americans, who were asked to do a lot of forebearing for a very long time. One reason for contemporary “polarization,” in contrast with the 1950s and even ’60s, is the fact that the electorate is now remarkably more inclusive than was the case then, prior to the Voting Rights Act of 1965. But we are also living with the consequences of the 1965 repeal of the 1924 immigration legislation. That earlier legislation was designed to preserve America along what we might today describe as “Trumpian” lines; northern Europeans received far more favorable treatment than other potential immigrants. Thankfully, that is no longer the case, but it is foolhardy to deny that one reason for our increasing polarization is the consequences of having become far more heterogeneous along lines of race, ethnicity, religion, and language.

We have always operated under a regime of “identity politics,” but in the past there was thought to be a single hegemonic identity, summarized in the language of my youth under the acronym WASP. As we have become more truly pluralistic and diverse, the nature of American identity is a subject of increasing contention, revealed most dramatically, perhaps, in the sometimes violent disputes about public monuments and what they represent. To their credit, Levitsky and Ziblatt caution against a critique, like that identified with Mark Lilla, that would reject “identity politics” in the name of bread-and-butter economics inasmuch as this would often have the practical consequence of averting one’s gaze from continuing discrimination, injustice, and indignity. But the recognition that “forbearance” might have its limits is fleeting. One awful possibility, left undiscussed, is that the United States, which now reaches from Maine to Hawaii and contains 320 million people, is just too large and multi-faceted to be effectively governed by any reasonably democratic political system.

Levitsky and Ziblatt certainly capture very well the mood of the moment, especially among those likely to identify with Dionne and the general politics of this journal (as I do). How Democracies Die is a quick read with some valuable analysis. The idea of what might be termed “autocratic character” is important, and other countries offer cautionary warnings about the ability of skilled demagogues to be transformative in almost entirely negative ways. I have no doubt that they are correct in their basic diagnosis of the dangers facing the United States as a democratic society.

But I think they end up all too American, as it were, by a refusal to address the possibility that at least part of our dilemma is traceable to what might be called “fault lines in the Constitution” that, just as their geological analogs, can wreak havoc. Even if we do not yet have effective ways of predicting earthquakes, we are aware of their possibility and can engage in a variety of important preventive measure designed to limit their consequences. But that requires becoming aware of the potential dangers. Our reverence for the Constitution, unfortunately, works to infantilize our political dialogue and therefore to limit the contribution even of such a heartfelt warning as How Democracies Die.