|
|
|
|
 | Borrowing Constitutional Designs: |
This file is also available in Adobe Acrobat PDF format Introduction AFTER THE COLLAPSE of communism, some thirty countries scrambled to craft democratic constitutions. Surprisingly, the modal constitution chosen in these countries was neither the pure parliamentary model found in most of Western Europe at the time, nor the presidential model found in the Americas. Rather, the modal constitution chosen after communism's collapse was semi-presidentialism--an unknown model known more generally as the "French type."1 This constitution combined elements of pure presidentialism and pure parliamentarism in one type. Specifically, semi-presidentialism combined a popularly elected head of state with a head of government who was responsible to a legislature. Throughout the 1990s, scholars and practitioners considered semi-presidentialism, in spite of its awkward name, its confusing structure, and its rareness as a constitutional type, a prudent choice for the new democracies. The constitutional specialist Giovanni Sartori, for example, suggested that some troubled democracies might benefit from "radically switching to semi-presidentialism."2 Gianfranco Pasquino, a noted scholar of European government, similarly suggested that, "on the whole, under most circumstances, semi-presidential systems appear endowed with both more governmental capabilities and more institutional flexibility than parliamentary and presidential systems."3 And along with these scholarly endorsements, semi-presidentialism moved rapidly across Europe, Asia, Africa, and even to Latin America.4 In fact, in addition to the eleven former Soviet Republics that are now semi-presidential, three of the newest members of the European Union--Poland, Lithuania, and Slovenia--are also semi-presidential, as are two of the candidate countries for European enlargement--Bulgaria and Romania. Eight of the twenty-six NATO member countries are now semi-presidential as well. Constitutional drafters in Brussels, led by the French, have been recommending a semi-presidential constitution for the European Union. And eleven African states, and five Asian and Middle Eastern states, are now among the forty-two semi-presidential systems in the world.5 Why the recent and increasing attraction to this model? The influential constitutional lawyer Maurice Duverger first sparked this attraction to semi-presidentialism decades ago. He pointed to France's positive semi-presidential experience during the Fifth Republic, differentiating it from France's turbulent and unstable parliamentary experience during the Fourth Republic. Using the example of France's volte-face, Duverger maintained that semi-presidentialism was "a viable" constitutional system with built-in mechanisms for ensuring political stability in hard times.6 The key to this constitution's success, according to Duverger and those who followed him, was the French model's flexible ability to alternate between pure "parliamentary and presidential phases" of government. By the time the Berlin Wall had come down in 1989, this conception of semi-presidentialism as an alternation between presidential and parliamentary phases had become, in the words of Arend Lijphart, "the most widely accepted concept of semi-presidential democracy."7 But analytically, as Lijphart also noted, this acceptance of semi-presidentialism as a simple alternation between pure presidential and pure parliamentary phases meant that "the semi-presidential category becomes a nearly empty cell."8 Concretely, with the semi-presidential category "empty," the numerous countries with a constitution like that of the French Fifth Republic can at best be analyzed as, and compared to, either pure presidential systems or pure parliamentary systems. Undeniably, the new institutionalist literature has developed along these very lines, classifying France and other semi-presidential countries as either "presidential" or "parliamentary," and analyzing political phenomena, such as the divided party control of national government, the frequency of assembly dissolution, and the performance of coalition government, from these traditional, alternate lenses.9 One recent example is the influential work on veto points by George Tsebelis, who states that "France is a case of a parliamentary system."10 At the same time, recent, important work in political economy by Alberto Alesina and Howard Rosenthal states the opposite, namely "that the similarities between the French and the American cases . . . are more striking than the differences," and thus, these authors lump France with the presidential, not the parliamentary, democracies.11 Some scholars have attempted to redefine the semi-presidential type, and bring it back to the debate over institutional performance. For example, Matthew Soberg Shugart and John M. Carey divide semi-presidential countries into two separate types of their own creation: premier-presidentialism, and president-parliamentarism. For Shugart and Carey, the key difference between these two types is the president's constitutional power vis-à-vis the prime minister. Shugart and Carey's analysis has contributed to our thinking about these systems by encouraging us to pay more attention to the amount of constitutional power held by executives. Yet their analysis has one stunning limitation, since they miss the crucial fact that a president in semi-presidentialism who does not enjoy a legislative majority finds his actual power severely limited, especially vis-à-vis a prime minister who does enjoy a majority. Thus, their classification, based on static measures of presidential power, fails to capture the dynamics of actual countries, such as France, where power shifts over time have led to varying patterns of governance.12 Consequently, the current default has been to return to Duverger's "presidential and parliamentary phases" conception. That the semi-presidential type has eluded scholarship is no mystery, for even Duverger's own thinking on this constitutional type changed dramatically as he watched the Fifth Republic evolve. Writing at the beginning of the Fifth Republic, for example, Duverger was originally concerned about the democratic deficit of the new constitution. Just after its adoption in 1958, Duverger cautioned against it in an opinion piece in Le Monde, claiming that "a great opportunity has just been lost."13 One of Duverger's early concerns with the constitution was that, of the five major institutions of the Fifth Republic's new governing system in 1958, only one (the National Assembly) was elected by direct universal suffrage. This changed, and after 1962, presidents of the Republic were also elected by direct universal suffrage. Writing then in 1978, once the Fifth Republic had been working well under semi-presidentialism for two decades, Duverger became more optimistic. He coined the term "semi-presidential," placed France in this category, and defined it simply as "the coexistence of a parliamentary-type government with a presidential-type head of state."14 Duverger selected two criteria according to which he organized his classic analysis of semi-presidentialism: the "consistency" of the parliamentary majority, and the president's relationship to this majority. This elegant (if vague) division notwithstanding, Duverger then suggested that from these two criteria, seventeen subtypes were possible. With this scheme, almost no semi-presidential country needed to approximate another, and probable statements regarding semi-presidentialism and democracy became nearly impossible. By 1980 (in English) and 1986 (in French), Duverger then moved in a more parsimonious direction and simplified his analysis by suggesting that all countries with semi-presidential constitutions would function politically, depending upon the president's relationship to the parliamentary majority, and the nature of that majority, simply like presidential or parliamentary systems of government.15 Writing now at the beginning of the twenty-first century, not only do we lack clarity regarding the classification of an important member of our transatlantic alliance, France, we also have many more cases of semi-presidentialism in the world--as many as forty-two countries by one recent counting--and we still lack a clear, agreed-upon, up-to-date conceptual framework for analyzing these countries.16 In spite of semi-presidentialism's recent and growing popularity in the world, there is still very little theoretical work and almost no comparative historical work concerning the structure of semi-presidentialism, the reasons for its varied performance across time and space, or its implications for constitutional law and democracy.17 To make matters worse, the variation within each of these countries over time is even greater than the dichotomous "presidential or parliamentary phases" shading that Duverger found in the French Fifth Republic.18 In the six East Central European countries that chose semi-presidentialism after the fall of Communism, for example--Poland, Slovenia, Bulgaria, Romania, Croatia, and Macedonia--majority cabinets, in which the prime minister headed a coalition supported by a legislative majority, occurred only half the time during the 1990s. A substantial portion of these majorities were not full majority governments, as in pure parliamentary democracies, but rather, complex "cohabitation" periods, in which the presidents of these countries and their prime ministers came from different (often opposing) political parties. Consequently, the heads of state and heads of government in these new democracies, each with a separate political agenda and substantial executive power, were often in conflict with one another. Most important, these countries spent the other half of their first democratic decade in situations in which neither the president nor the prime minister controlled a legislative majority. Practically, this meant that countries spent at least half of their first democratic decade in the hands of leaders representing a minority of the population. This tendency for there to be minority presidents and minority prime ministers marks a drastic difference with the "model" in the French Fifth Republic, which enjoyed majority governments for most of its life span. Moreover, France did not have to confront a solid decade of "minority politics" while it was reequilibrating its democracy, as much of the postcommunist world did. The question thus remains, how should we best analyze these semi-presidential systems? Can we move toward probabilistic statements concerning semi-presidentialism--as the world's third governing framework--and democracy? For example, under what conditions is semi-presidentialism most (or least) likely to function in a way that is supportive of democracy within a particular country? When does semi-presidentialism enable leaders to govern with efficacy (the capacity to find answers to problems facing the system) and effectiveness (the capacity to implement policies with the desired results)?19 When does this constitution enable leaders to democratically solve the distributive conflicts that affect new democracies, and when does it not? Is there more to the type than the "presidential or parliamentary phases" conception led us to believe? These questions are more important than ever, and a reconceptualization of the semi-presidential constitution, investigating semi-presidentialism's relationship to democracy, is therefore urgently needed.20 My book begins to provide this theory and evidence. The book has three parts. I begin with a conceptual cleaning of semi-presidentialism, establishing it as a single, unique constitutional type by showing that its incentive structure is radically different from both pure presidentialism and pure parliamentarism. I then present a simple typology of semi-presidentialism that builds upon, but also offers us greater analytic power than, the previous conceptions. Borrowing insights from American politics and comparative politics, I go on to show that the performance of semi-presidentialism is highly contingent upon the interaction effects between semi-presidentialism and a country's party system. In this first part of the book, I then detail which party system combinations work well, and which work poorly because they, together with the constitution, provide incentives for presidents (and other actors) to transgress democratic boundaries.21 Analyzing whether, and when, democratically elected leaders have incentives to govern democratically under a particular constitution is crucial, for "if freely elected executives . . . infringe the constitution, violate the rights of individuals and minorities, impinge upon the legitimate functions of the legislature, and thus fail to rule within the bounds of a state of law, their regimes are not democracies."22 More specifically, I distinguish three, qualitatively different, electorally generated, subtypes within semi-presidentialism: consolidated majority government, divided majority government, and divided minority government. I argue that with respect to the consolidation of democracy the most critical of these subtypes is divided minority government. Divided minority government is logically more likely to be characterized by legislative immobilism and cabinet instability, on the one hand, and simultaneously, by continuous presidential dominance, on the other hand. I suggest that divided minority government combines the gridlock of presidentialism's divided government with the cabinet instability of parliamentarism's minority government. This unique combination of legislative gridlock and governmental instability can put excessive pressure on governments, presenting strong incentives for independently elected presidents to govern alone. How is democracy then affected by the semi-presidential type, and particularly, by a country's placement in these various subtypes of semi-presidentialism? In order to examine this, one needs to observe how governments go about the day-to-day business of accomplishing their political programs, whether they do so in ways that suggest to their societies that the most urgent distributive conflicts are being regulated effectively and efficiently, and whether they convince society's current losers in these conflicts that there is a decent chance they will be tomorrow's winners.23 Adapting the work on American constitutionalism by Robert Dahl, I therefore ask how well semi-presidentialism--as a constitutional arrangement--helps a country to:
In this book, I am also concerned with the consolidation of democracy, which has attitudinal and behavioral dimensions in addition to procedural ones. A democracy is consolidated behaviorally when no significant political group tries to overthrow the regime or promote violence as a means of secession. Attitudinally, democracy is consolidated when the majority of citizens firmly believe that political change must take place according to democratic procedures, even in cases of severe crisis. Constitutionally, democracy is consolidated when the laws, procedures, and institutions governing the democratic process are respected by leaders and subjects alike.25 To actually begin to see these effects on democracy and democratic consolidation empirically, one needs to examine and compare governments under the three subtypes. But in addition, one must peer into the parliamentary debates on legislation, study the type and amount of conflict between and even within branches of government over this legislation, and also observe and try to understand how citizens perceive the quality of government over time. My method in this book is thus to compare what took place with respect to distributive conflicts during periods of consolidated majority government, divided majority government, and divided minority government across cases and over time. In order to do so, I have chosen two countries that for better or worse have been among the most influential countries for European politics in the twentieth century: the Weimar Republic from 1919 to 1933, and the French Fifth Republic from 1958 until 2002. Using the typology I present and discuss in the first chapter of this book, I classify all governments in the Weimar and French Republics as either consolidated majority, divided majority, or divided minority. The fourteen German years and forty-four French years fall into one of the subtypes of semi-presidentialism. This provides a total of fifty-eight years of semi-presidential governments to analyze. Granted, the pre-World War II and post-World War II units, from which these countries are chosen, are significantly different from one another in many respects. However, should we see the patterns associated with certain subtypes in one period then recur across these space and time units, the generalizations made here would accordingly gain strength.26 Therefore, one case from each of these historical periods seems necessary. Why these two countries? Weimar was one of the first historical experiments with semi-presidentialism, a fact that has been almost completely absent from the minds of semi-presidentialism's advocates. Interestingly, many of the newer semi-presidential democracies have institutional and social characteristics that bear a relatively strong resemblance to Weimar--not France. And Weimar, unlike France, also spent more than half its life span in the subtype of semi-presidentialism I call divided minority government. Shouldn't Weimar's experience with the various patterns of semi-presidentialism be revisited and brought back to the discussion and attention of constitutional engineers? Duverger claims that without semi-presidentialism, "with a conventional parliamentary government, the Weimar Republic would probably have collapsed sooner than it did."27 Sartori concurs, noting that, "had Germany adopted in 1919 a parliamentary type of constitution, I believe that Weimar would have collapsed much earlier than it did."28 In chapters 2 and 3 of this book, I challenge Duverger's and Sartori's analyses of Weimar, and bring this early and quite devastating case back to bear on the analysis of semi-presidentialism as a potential governing framework for fragile democracies. In chapters 4 and 5, my typology of semi-presidentialism, and the insights from the Weimar case, are then used to structure a new exploration of the French Fifth Republic. Duverger's classic work on semi-presidentialism was motivated by the experience of the Fifth Republic. Almost every new democracy choosing semi-presidentialism has claimed to take its inspiration from the "French model." Newspapers and radio programs, as well as constituent assembly debates, in Russia, Poland, Taiwan, Argentina, Nigeria, the EU, and many other countries, document this widespread perception. And yet, what is really known about the French experience? The assumption is that French semi-presidentialism has never been problematic, and rather, has helped solve the problems that plagued the institutional turbulence of the Fourth Republic. But a recent revision of the French constitution confirms that semi-presidentialism undeniably has been a source of strain. Duverger's landmark English-language piece, published in 1980, did not take into account any of the real challenges to democracy posed by French semi-presidentialism. One of the most crucial of these challenges, as I detail in this book, surfaced between 1958 and 1962, when French democracy was nascent and the Algerian question presented serious shocks to the entire democratic system. What was the critical push from this turbulent period to uninterrupted decades of consolidated majority governments in France? Did these consolidated majorities help French democracy weather the turbulent period of 1968 without a constitutional crisis? What were the negative, exclusionary aspects of this long period of consolidated majorities? This crucial story needs to be told in order to shed greater light on, and nuance, the Fifth Republic's constitutional experience, which the existing literature simply casts as positive. During the respective periods studied, both countries in this book were facing "end of empire" situations that exacerbated the problems of democratic stability. And yet each country had a qualitatively different experience with constitutional law, and with democracy, over time. Together these countries, as analyzed by this new framework, offer us an analytically useful picture of the possible, varied patterns of semi-presidentialism, and their relationship to democracy in the modern and contemporary world. One of my central points in this book is that semi-presidentialism, qua type, is conceptually and analytically distinct from presidentialism and parliamentarism, and thus must be brought into the debate on constitutional law, constitutional frameworks, and democracy in its own right. So far, our failure to do this has impeded our understanding of this increasingly important constitutional framework, a framework that already structures political and economic outcomes across the European continent. One of my other, more general points in this book is that comparative historical analysis must be brought back to the theory and practice of constitutional law. For, as J. M. Balkin and Sanford Levinson caution, the current study of constitutional law in the United States lacks comparative and historical perspective. Importantly, "this tendency is revealed whenever American constitutional lawyers eagerly offer advice to emergent political systems struggling with the task of constitution writing. Like all too many of their professors, students rarely grapple with the possibility that there are other ways of dealing with the problems of democratic institutions."29
File created: 8/7/2007 Questions and comments to: webmaster@pupress.princeton.edu |