History rewards cautious judgments. Had this book been written a half-century earlier, at the height of the postwar consociationalist experiments, a certain triumphalism would no doubt have accompanied the description of negotiated power sharing among rival ethnic or religious groups. That enthusiasm would no doubt have waned as civil wars ravaged Cyprus, Sri Lanka, Ivory Coast, and other bastions of negotiated accommodation of coordinated political solutions.

Twenty-five years have now passed since the fall of the Berlin Wall and a renewed period of democratic euphoria. The historic end of the Cold War brought with it a heralding that the epochal wars of the twentieth century had at last been concluded. Democracy was triumphant. Its ideological challengers of fascism and communism were defeated. The market was ascendant in China and the few outliers in North Korea or Cuba were simply rogue states that were ill suited to resist the demands of their populations for freedom and improved material standards of living.

With the fall of the Soviet Union, the proxy wars of the great powers ended, leaving the client states of East and West vulnerable to popular demands for liberty and democratic rule. In short order, apartheid fell in South Africa, democracy took root in the Pacific Rim, and Mexico recovered competitive elections. Entire regions were transformed, as with the stabilization of civilian rule in Latin America. Even in sub-Saharan Africa, long the bastion of strongman rule, there was actual rotation in office for the first time in the postcolonial period.

History was on the march, and this was a glorious Whiggish account, with perhaps a touch of Dr. Pangloss thrown in for good measure. The Hegelian unfolding of events had revealed the “end of history,” as proclaimed by Francis Fukuyama: “What we may be witnessing is not just the end of the Cold War, or the passing of a particular period of post-war history, but the end of history as such: that is, the end point of mankind's ideological evolution and the universalization of Western liberal democracy as the final form of human government.”

What are the boundaries of democratic tolerance? Of necessity, elections require the engagement of ideas, disagreements over policy, mobilizations of adherents, and fraught exchanges spilling over into core challenges frequently beyond the border of insult. Even while proclaiming its wide-open quality, democratic engagement presupposes some baseline accords on mutual tolerance and respect for rights across the spectrum, as well as the ability of democratic majorities to assess the past, make new political commitments, and alter the course of government. What of those who reject the core tenets of democratic governance, those whose participation exploits the porousness? This question leads to the precarious state of what may be termed democratic intolerance; that is, the intolerance that democratic governments must exhibit toward antidemocratic actors in the name of preserving the fundamental democratic character of government.

While much of the discussion that follows draws upon the tragic inability of Weimar Germany to withstand the Nazi attack from within, we can begin with a more contemporary example. The 2006 controversy surrounding the Danish cartoons mocking Islam provides an illuminating window into the problem of challenges to core democratic values. At issue were cartoons appearing in a secondary Danish publication that were deemed sacrilegious by religious adherents. Although the political maneuverings and machinations surrounding the protests were no doubt multifaceted, the controversy centered on Islamic fundamentalist demands that Denmark be held responsible for its failure to censor the publication of a series of cartoons perceived as blasphemous attacks on the Prophet Mohammed.

In commenting on the publication of these cartoons, my late colleague Professor Ronald Dworkin provocatively asserted a right to insult as a precondition for political engagement. In so doing, he made a moral and instrumental argument requiring weak or unpopular minorities to tolerate social insult as a condition of making a claim on the majority for protective antidiscrimination legislation.

In conversations over the years with justices from the South African Constitutional Court, and with South African scholars, I always come back to a question of historical uncertainty. By the time the South African court entered its famous decision in the Certification case, the transition to post-apartheid governance had been largely completed. Now secure in his role as head of state, President Mandela had control of the police and the military, had a formidable legislative majority, and possessed an overwhelming sense of authority as the unquestionably dominant figure in South African politics. The constitution that emerged from the constituent assembly process may have had defects, but at bottom it was just not that bad. It more or less honored the initial agreements, and where it erred on the side of too much unconstrained political power, it did so mildly. In other words, the initial constitutional effort was by no means an affront to the basic values that animated the South African transition, nor would it have been challenged as fundamentally unacceptable in any democratic country.

The question I then ask is why Mandela did not simply disregard the constitutional court. As the future architect of the U.S. Bill of Rights once rhetorically inquired, what reason do we have to think that “parchment barriers” are of any use “against the encroaching spirit of power?” Mandela might have echoed Joseph Stalin's rejoinder to disapproval from the Vatican: “How many divisions does the Pope of Rome have?” Perhaps more gently, Mandela could also have claimed shared constitutional authority with the court, a nascent form of the “popular constitutionalism” claims that briefly entertained American constitutional law at the time. Or, most simply, why not just thank the court for its efforts and get on with the fraught task of governing the still fractured republic?

To ask such a question is to invite a bewildered look of incomprehension. To begin with, Mandela's personal trajectory – notably his long legal battles against his incarceration on Robben Island – left him with a surprisingly deep respect for the rule of law.

The modern wave of democracies would have been inconceivable without the collapse of the Soviet Union. Most apparently, the removal of the Soviet armed presence allowed the mobilization of democratic forces without the threat of the invasions of Hungary in 1956 and Czechoslovakia in 1968. The impact of the demise of Soviet power reaches much further, however. The easy divide of the world along the Cold War lines of demarcation ended, with far-reaching consequences. The American patrons of right-wing regimes reexamined their commitment to nondemocratic rulers once freed from the need to maintain the anticommunist alliance. Similarly, client states of the Soviet Union from outside the Soviet bloc found their benefactor and, oftentimes, financier suddenly unresponsive. For the Cubas and El Salvadors of the world, one on each side, the realignment forced a recalibration of authoritarian rule in a seemingly post-authoritarian world.

Realignment was not limited to those holding state authority. The end of the Cold War also commanded the agenda of those contesting the lockhold on power of the client regimes of the major powers. For every military regime in Central America counting on American support there was at least one guerilla group drawing sustenance from the Soviet Union or its regional proxy, Cuba. For every Soviet satellite in Africa or Eastern Europe there was a corresponding challenge from some group funded directly or indirectly by Western sources. The prompt toward democracy after 1989 did not need to be triggered by the sudden withdrawal of Soviet troops. It could as easily have been the product of the collapse of the corresponding Cold War patrons and the need for national accommodation of competing political factions.

No country seems more removed geographically from the Cold War struggle than South Africa. The National Party (NP) government was no mere American puppet, and in its final stages, the apartheid state had to withstand increasing economic sanctions from its Western allies.

Extremist groups threaten democracy in terms of both what they might try to do through elections and governmental office and what they might provoke democratic societies to do in order to ward off the perceived danger. The threat is real, from both directions. That there are antidemocratic groups trying to worm their way into governmental positions so as to undermine tolerant, pluralistic, democratic societies is not a new development. What is perhaps new is the increasing likelihood that these groups will be clerically inspired rather than driven by the messianic social visions of communism or fascism. But there is the corresponding threat of excessive intolerance. Democracy thrives on contestation, and suffers if the ambit of democratic deliberation is drawn too narrowly and if the threat to social peace is used to drive out the uncomfortable voices of dissent.

In most circumstances, efforts to silence parties by prohibition are probably ill advised. As nettlesome as the Quebec independence movement may have been for Canada, the national government's ability to channel disputes over Quebec's status through the political process and even the Supreme Court is far preferable to any attempt to drive the party underground. The relative civility and tolerance of debate in Canada, however, is unfortunately not the global norm. So the question becomes what preconditions must exist for the banning of parties or for other restrictions on political expression in the electoral arena. Here I wish to leave to the side the parties alleged to be allied with insurrectionary or regional military forces. With respect to such parties, the directness of the organizational link to unlawful activity and the immediacy of the likely harm serve as workable responses to the problems posed, at least in theory.

In the absence of direct or indirect participation in illegal or violent activity, the starting point for any discussion of the banning of political parties, political participation, or political speech should be that the presumption is in favor of freedom of political expression and association.

On November 6, 1985, leftist guerrillas from the Colombian 19th of April Movement overran the Colombian Supreme Court building, taking the justices of the court as hostages. In the ensuing shoot-out with the military, twelve of the judges were killed, along with more than 100 other civilians. The shootings at the court were but the most visible signpost that the country's background of war and strife could overwhelm even the central institutions of state in Bogotá, the capital. Colombia was, for much of the last half of the twentieth century, a state struggling for control of its territory against powerful private militias, and the resulting loss of state authority left open the terrain for one of the highest murder rates in the world, pervasive extreme poverty, and a stubbornly flurishing drug trade. After a civil war that left more than 200,000 people dead during the Gran Violencia of the 1950s, and then the militant uprisings and drug wars of the following decades during which even more succumbed to violence, any prospect of peace and stability seemed nonexistent.

Yet in 1991, in the face of overwhelming odds, a democratically elected constituent assembly promulgated a new constitution that served as part of the “profound constitutional moment throughout the Americas.” The fall of military dictatorships across much of South America, most notably in Argentina, Brazil, Chile, and Uruguay, unleashed a democratic revival in the region and a new commitment to limitations on the powers of government. In many ways, the democratic surge in South America paralleled the broader democratic expansion in Eastern Europe and Asia following the fall of the Soviet Union. In each case, central to the desired democratic restoration was the idea of constraint in the exercise of state authority. Authoritarian rule allowed the direct translation of political power into arbitrary and repressive governmental conduct. In response, the democratic revival sought both to restore civilian authority and to enshrine the primacy of a rule of law.

Banning political parties is not the desired end state of democracy. Rather, it is a prudent act to ward off an existential threat not yet realized. It is always possible that driving a party underground will harden its cadres, or gain them public sympathy as romantic figures of the Che Guevara sort. These are questions of prudence and tactics that can only be answered pragmatically. The question of principle is the justification for the ban, not whether it is wise under the particular circumstances.

Nor is it an objection to say that a party facing a ban is still weak. How much wiser to remove surgically a cancerous tumor when it is small rather than large. The wisdom of the incision depends on the proper diagnosis, not whether the patient is at death's door. Indeed, allowing an antidemocratic movement to mature into a mass party renders intervention quite unlikely to succeed.

The initial question of principle is therefore to identify the conditions that require democracies to assume a defensive posture against threats to their continued existence. In turn, this requires identifying the types of risk that democratic states consider to require removal from political participation. By and large, these party prohibitions fall into one of three categories, each of which raises a separate set of concerns.

First, there are prohibitions that result from groups having a terrorist or insurrectionary component based either domestically or abroad. The paramilitary side of these groups is independently subject to criminal prosecution or defensive military operations. Yet there is often a political wing that seeks to operate publicly and may serve as a legal or propagandistic front for terrorist or insurrectionary groups. A ready example is the complicated relationship between the Irish Republican Army's military wing and the Sinn Féin political branch.

When teaching a course on the constitutional law of democracy, it is my custom to open the first day of class with what I describe as a modern-day parable. As presented to the students, the events concern a precipitous and heated confrontation some twenty years ago that I suggest may have escaped their notice. The conflict pitted the head of a large and powerful state against the leader of a small and poorly armed state. As the events unfolded, the tension between the two leaders mounted dramatically. The partisans of the two camps became increasingly divided and braced for what seemed the inevitable showdown. After months of skirmishing, the denouement arrived and the leader of the small state proved victorious. Perhaps most remarkably, the leader of the state of crushingly superior military force turned over his entire arsenal to the newly victorious head of the small state and quietly left office with little concern for incarceration or assassination, contented with nothing more than the equivalent of the palace guards to usher him away.

Now the events in question are, of course, a cartoon rendition of the 1992 U.S. presidential election between the incumbent President George Bush and the governor of Arkansas, Bill Clinton. The shaggy-dog quality of the telling aims to convey just how exceptional it has been, in the course of human history, for power to pass peacefully across a heated partisan divide, particularly when the usually decisive weight of military force rests exclusively in the hands of the displaced incumbent power. The question for the students then becomes to discern the necessary conditions for power to pass in this fashion.

In turn, the chapters that follow examine just this question from the perspective of new democracies. Our modern conception of democracy assumes what Abraham Lincoln termed a “government of the people, by the people, for the people,” in which each election presents a meaningful alternative between rival sets of candidates, and in which the electors make the choice of who shall govern.

In 2010, retired Justice Albie Sachs of the South African Constitutional Court was giving a series of lectures on his recently published autobiography, The Strange Alchemy of Life and Law. As part of that series, Justice Sachs came to New York University School of Law, where I teach, for a public event on the role of a constitutional court with my colleague Jeremy Waldron and myself. Our inquiry was on the role of a court in the transition from an unjust to a just regime, and on how a court addresses the question of the frailties of an emerging democratic society. Jeremy Waldron is famously protective of the dignity of the legislative process and critical of judicial review of legislative enactments. My views, as will be evident from this book, are notably different, and I defend the importance of judicial oversight of some of the characteristic vulnerabilities of democratic rule.

The discussion with Justice Sachs grounded the divide in the role of courts in the South African experience. The South African Constitutional Court has actively engaged the political process during the transition process and beyond, a theme I develop in subsequent chapters. Alone among courts anywhere in the world, the constitutional court was tasked with assessing the first draft of the constitution for its conformity with core democratic principles, and indeed found the draft wanting. This was the world's first unconstitutional constitution.

While Justice Sachs strongly defended the active role of the South African court in the transition from apartheid to multiethnic democratic rule, he was cautious about the relation between the emerging South African jurisprudence and American constitutional doctrines. According to Justice Sachs, it was difficult in any new democracy to resist the gravitational pull of American constitutional law, particularly for an English-language constitutional court. It was not simply the longevity of American democracy and the U.S. Constitution, but the commanding language and arguments honed by the U.S. Supreme Court over centuries of constitutional debate.

Until quite recently, most scholarly discussions of the restrictions on antidemocratic groups began (and many of them ended) with the question of whether a democracy ever has the right to impose viewpoint constraints on extreme dissident views. Professors Gregory Fox and Georg Nolte, for example, in their important contribution to the debate, primarily focused on the possibility of restricting political participation consistent with international law, particularly the guarantees of the 1966 International Covenant on Civil and Political Rights. The responses to Professors Fox and Nolte did not question their analytic framework; instead, they simply challenged the capacity of any society to police the boundaries of something as nebulous as “democracy” and questioned whether the remaining product was worthy of the name:

Although these critiques take a back seat to claims that suppression does not work, all of these arguments tend to lump together the different sorts of responses that might be deployed against antidemocratic threats. More recently, however, “consensus has emerged on the fundamental principle underlying the theory and practice of militant democracy: Democracies have a right … to limit fundamental rights of free expression and participation – albeit with various qualifications and caveats – for reasons of self-preservation.” The idea that every political viewpoint is entitled to participate in the political arena “has virtually no supporters today.” Even staunch advocates of protecting political parties such as Nancy Rosenblum acknowledge that there are times when, quite simply, “the facts justify banning” a party.

Rather than questioning whether prohibitions of antidemocratic forces are ever possible, it is more fruitful to begin by focusing on what kinds of prohibitions are permissible and under what circumstances.

The argument of this book is unreservedly instrumental. There is no claim of “Herculean” wisdom in the judiciary, as per Ronald Dworkin, nor any attempt to engage at a first-order level the arguments for greater democratic legitimacy of the political branches, as per Jeremy Waldron. Instead the argument is that courts have emerged as the stopgap protections to two of the classic disabilities of democracy, particularly for young and untested regimes. The first challenge comes from without in the form of enemies of democracy who use the inherent porousness of democratic politics to undermine the core value of electoral legitimacy that underlies any form of democratic rule. The second comes from within through the suffocating command over the instrumentalities of government by a strong party or strong leader, no longer accountable to meaningful electoral contestation.

Despite the provenance for the new form of court-enforced constitutionalism in the need to watch over the democratic process, the reality is more complicated. In many instances courts back away from this role, fearing wisely or not that intervening in the name of democratic legitimacy threatens direct conflict with political powers. Wojciech Sadurski notes that, despite their strong constitutional mandate, “the dominant justification for the robust position of the constitutional courts of post-communist states in [Central and Eastern Europe] is based on the role of these courts in the protection of individual rights – in particular, those explicitly entrenched in the respective constitutions.” Sadurski attributes this reluctance to engage the preconditions for democracy as such to, at least in part, “the existence of reasonable disagreement about the processes and devices of democracy.”

But the reluctance to embrace democracy as the metric for court oversight can hardly be explained by the complexity of the ensuing inquiry. Constitutional law abounds with topics such as abortion, affirmative action, hate speech, or same-sex marriage that are no less subject to disagreement for being framed in the language of rights.