REVIEW ESSAY

ASSESSING THE THIRD WAVE DEMOCRACIES: SLIPPERY CONCEPTS, UNCERTAIN PROPOSITIONS, AND PARTIAL INSIGHTS FOR LATIN AMERICA

Juan J. Linz and Alfred Stepan. 1996. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore, MD: Johns Hopkins University Press. Figures, tables, exhibits, index. 479 pp. Hardcover $55, Softcover $18.95.

Larry Diamond, Marc F. Plattner, Yun-han Chu, and Hung-mao Tien, eds. 1997. Consolidating the Third Wave Democracies: Themes and Perspectives. Baltimore, MD: Johns Hopkins University Press. Figures, tables, index. 343 pp. Softcover $14.95.

Larry Diamond, Marc F. Plattner, Yun-han Chu, and Hung-mao Tien, eds. 1997. Consolidating the Third Wave Democracies: Regional Challenges. Baltimore, MD: Johns Hopkins University Press. Figures, tables, index. 332 pp. Softcover $14.95.

Larry Diamond and Marc F. Plattner, eds. 1995. Economic Reform and Democracy. Baltimore, MD: Johns Hopkins University Press. Figures, tables, index. 252 pp. Hardcover $40.00, Softcover $14.95.

Beginning with the liberalization of authoritarian rule in Portugal in 1974 and extending into the 1990's, dozens of countries around the world have completed transitions to democracy. All four books under review represent the state-of-the-art of a significant body of work that assesses how these "third wave" democracies have fared in the post-transition phase.⁽¹⁾ The theoretical scope of the books is impressive, including such topics as electoral systems, civil-military relations, presidential versus parliamentary rule, and the effects of economic reform and development on democracy and vice versa.

The empirical domain of these books is equally stunning. Through broad comparative studies, these works attempt to facilitate theory-building by extending the range of empirical cases of democratization. Juan Linz and Alfred Stepan go beyond their work on South America and Southern Europe to include comparisons with Eastern Europe and Russia. The Larry Diamond et al. Themes, Regional Challenges, and Economic Reform and Democracy volumes are more ambitious, adding African, Middle Eastern, East and Southeast Asian countries and China to the range offered by Linz and Stepan.

Despite the theoretical and empirical complexity of their task, the authors make a number of important contributions to the study of democratization. First, these works through their comparative scope demonstrate that there is no singular path to democratic consolidation. Democracy-building is complex and, although important, elections are not sufficient for guaranteeing the survival of democracy. Second, these books take the reader beyond the limited nature/nurture debate over whether democracies are "crafted" by skilled leadership or the result of "structural" preconditions. As Samuel Huntington asserts in Themes, "clearly . . . both preconditions and crafting have roles to play, and certain preconditions can facilitate democratic crafting" (p. 4). Third, the authors respect the international dimensions of democratization. Foreign powers, international institutions, global markets, and transnational ideas are part of the "domestic" story of democratization. Finally, these works bring a sober and somewhat pessimistic view of the progress of the "third wave" that corrects pedestrian visions of democratic triumphalism emanating from international policy circles. For most of the authors, the once globally resurgent expansion of democracy has moved into a phase of consolidation or, worse, perpetual unconsolidation. New democracies are rare and most of the third wave democracies have stagnated in an intermediate category that mixes procedural democracy with "illiberal" aspects (e.g., presidential decree powers, curbs on political and civil rights, restrictions on the media, and civilian-led democracy with military prerogatives over key areas of policy).⁽²⁾ These countries are, as Philippe Schmitter has argued, "condemned to democracy."

The stagnation of the third wave is troubling, but it is on this point that the works under review falter. They fail to develop new concepts to understand why "unconsolidated democracies" endure and even improve in key aspects. The authors offer important insights into the complexity of democratization, but they fail to get past the obstacles created by their main organizing concepts: consolidation, institutionalization, and the "rule of law." Often, these concepts obscure the conditions under which elites become more accountable to the demos - the central principle of liberal democracy. The contribution of these works is also muddled by contrasting theoretical approaches and empirical conclusions that are only loosely tied together into coherent theoretical frameworks. In the end, the authors do not address larger questions - how democracy-building is related to state-building and market-building - that their cross-regional comparisons to Russia and Eastern Europe beg.

RETHINKING DEMOCRATIC CONSOLIDATION

It is odd, but also telling, that a term which implies definitiveness, is still so heavily contested among theorists. What is "democratic consolidation?" What should observers look for to know that a democracy has become "consolidated?" More important, how do studies of consolidation keep from sliding into the analytical taboo of tautology - defining "consolidation" in terms of the factors that allegedly foster it?⁽³⁾

Linz and Stepan and most of the Themes and Regional Challenges authors define "consolidation" as a point in which "democracy becomes routinized and deeply internalized in social, institutional, and even psychological life, as well as in political calculations for achieving success" (Themes, xvii). This common definition of consolidation includes behavioral, attitudinal and institutional dimensions. Behaviorally, consolidation occurs when significant actors in the polity avoid the use of nondemocratic, illegal, or unconstitutional means to achieve their respective ends.⁽⁴⁾ Attitudinally, consolidation exists when a "majority of public opinion holds the belief that democratic procedures and institutions are the most appropriate way to govern collective life in a society such as theirs."⁽⁵⁾ Antisystemic "disloyal oppositions" do not exist. Constitutionally, a democratic regime is consolidated when all significant actors in a polity are subjected and habituated to the rule of law, the procedures and institutions of the democratic process.⁽⁶⁾ In short, democracy is consolidated when it becomes "the only game in town."⁽⁷⁾

Each of these dimensions, however, presents inherent problems for the study of democracy. The behavioral aspect of consolidation fails to define the degree of nondemocratic action that is necessary to "unconsolidate" a democratic regime. Moments of "disloyal" opposition occur in even the most consolidated democracies, including the United States and Spain. Democratic procedures meant to ferret out these acts have themselves been subverted in these countries. Does the existence of such disloyal behavior threaten the consolidated nature of U. S. or Spanish democracy? Surely the frequency and severity of such acts are greater in new democracies of questionable consolidation such as Brazil's or Russia's. Therefore, some measure of degree in the activity of the disloyal opposition and in the context of such behavior must be assessed.⁽⁸⁾ Yet none of the authors provide such a universal yardstick for measuring disloyal action.

The attitudinal dimension fails to delimit the time frames for democratic consolidation. By stressing the beliefs of elites and the public, this dimension focuses not on short-term cultural events, but longer-term societal transformations. An accurate assessment of beliefs requires time for these cognitive constructs to mature. Despite this, the authors use various time frames to discuss the consolidation of democratic beliefs. For Dahl in Themes, the emergence of a "democratic culture" in South Korea and Taiwan took 40 years (p. 38). Yet Linz and Stepan study beliefs by employing public opinion survey research done only years after regime transitions. If Dahl is right, these assessments would only register "blips on the screen" - minute changes in the long-term evolution of the polity. If the consolidation of democratic beliefs takes decades, why speak of "consolidation" - a term denoting definitiveness - so soon after transition?

The real danger of the attitudinal dimension of consolidation is the tendency to see beliefs both as a cause and as a consequence. Dahl sees the existence of prodemocratic beliefs as an essential precursor to consolidation but, for Linz and Stepan, these emerge as indicators of consolidation that embody aspects of their independent variables. This confusion between independent and dependent variables invites circular and tautological arguments seen by many critics as all too common in the "consolidation" literature.

Another problem with the analysis of "consolidation" is the tendency to see the process in "teleological" terms. According to the authors and organizers of all of the volumes, democratization occurs in "phases," "paths," and "sequences" and then political change becomes consolidated at a certain point in time or it stagnates, reverses, and "deconsolidates." Although not teleological in the sense that the authors have in mind a natural movement toward an a priori end called "consolidation," such arguments are no less certain that the process of consolidation either progresses or regresses and that the indicators of these movements are knowable.⁽⁹⁾ Such a view, however, unnecessarily obscures several important aspects of democracy-building.

Linz and Stepan suggest at different times that democratization is circular. Consolidating democracies may still be "risk prone" to break down into authoritarian regimes (e.g., Greece and Uruguay). Such a "do or die" approach to democratization makes less sense if the process is governed by "learning by doing" experimentation in which setbacks are assets that teach democrats how to build a more workable democratic compact. Many of the authors argue that "learning" was an important factor in building South European democracy.⁽¹⁰⁾ Wouldn't learning also be important in making Latin American democracy better? Such an evolutionary approach applied to Brazil would not register the faults of that democracy in the mid-1990's as symptoms of a failing, out of sequence democracy or of an unchanging process "condemned to democracy," but of an evolving democracy that is building upon its experiences. In that regard, the impeachment of Fernando Collor in 1992 may be read both as a low point for Brazilian politics but also as an event that strengthened Brazilian democracy for the long haul.⁽¹¹⁾ Governed by its tendency to see a one-dimensional process of progression and regression, the study of "consolidation" is too unforgiving during such low points; too eager to see the setbacks as evidence of potential breakdown. A more useful view of democratization would accept its inherent sloppiness and open-endedness.

An evolutionary approach to democratization makes more intuitive sense considering how older democracies have emerged historically. Would the term "consolidated" be applied to a democracy that only gave the vote to white, male landowners (U. S. democracy in 1789)? Such a democracy might be consolidated on all three dimensions - attitudinal, behavioral, and institutional - but we would struggle with the fact that it would exclude a large percentage of the population. If, however, we treat such a system not as "consolidated" - which implies definitiveness - but as "evolving," then we can appreciate the dynamic process of building U. S. democracy over decades.

Another major problem with the application of the term "consolidation" to democracy is the tendency to assess when democracies consolidate rather than to assess the quality of democracy. The analysis of most of the authors suggests that what is really important is the quality of democracy. Linz and Stepan in their contribution to Themes argue that "no regime should be called a democracy unless its rulers govern democratically" and "only democracies can become consolidated democracies" (p. 15). Yet ruling democratically for Linz and Stepan assumes that violations of rights, infringements of constitutional rules, and breakdowns in the law - all issues central to the accountability of elites to the demos - must be extirpated before "consolidation" is even considered. But if all of these guarantees must be made beforehand, what is the unique value of the term "consolidation?"

Separating consolidation from the quality of democracy must be done carefully so that the former concept is not left vacuous. In justifying the distinction between the two concepts in their book, Linz and Stepan argue that no democracy can guarantee the quality of other public institutions (e.g., the central bank, the armed forces, the courts, etc.). It is possible then to have a consolidated democracy in a socially unjust and corrupt state and society. Surely, though, the "quality" of these other institutions affect the accountability of elites and, therefore, the very existence of a consolidated democracy. That leaders are elected is stripped of meaning if these officials cannot improve the professionalism of a police force and court system that is rampantly killing citizens (Brazil) or brutalizing youths (Argentina).

Making elites more accountable to the electorate assumes that both the process that puts them in power and the rule of their government are democratic. In this regard, Philippe Schmitter in Themes offers a superior framework by specifying the short-term and long-term aspects of democratic consolidation. In the short-term, consolidation means that elites solve their intrinsic conflicts over rules. In the long-term, democratic consolidation will depend upon the extrinsic effects that policies made under these rules will have on civil society (p. 241). In this sense, no democracy can be considered "consolidated" until the quality of democracy is assured.

CONSOLIDATION AS INSTITUTIONALIZATION

The final dimension of "consolidation" - institutionalization - presents additional problems for analysis. The equation of "consolidation" with "institutionalization" ignores certain qualities of the "unconsolidated" democracies that may: 1) suggest that these are really "consolidated" democracies after all, and 2) prove that these cases are more deserving of the label than many countries included under the "consolidated" rubric.

Many South American democracies, and particularly Argentina, Brazil, Bolivia, and Ecuador, are routinely labeled as "unconsolidated." Although these countries have endured as democracies for over a decade, they are considered "unconsolidated" due to the fragility of their democratic institutions (e.g., weak party systems, the persistence of clientelism and corruption, etc.). In comparison to the "consolidated" South European cases, however, the "unconsolidated" nature of these South American democracies is not apparent. O'Donnell in Themes shows that democracies in both regions pass all of the crucial "tests" of consolidation including: governments that have alternated in power peacefully for several years; popular support for democracy that has persisted on balance (despite periodic ups and downs); and the absence of "antisystem" parties (pp. 48-49).

It might even be argued that the South American countries are more consolidated than the much-vaunted democracies of Southern Europe in that the former set of cases underwent more significant political restructuring and economic hardship while maintaining support for democracy. By surviving more brutal "tests," the South American democracies are fundamentally stronger than the South European democracies.

The shortcomings in the definition of "consolidation" revealed by O'Donnell suggest that the term lumps and splits too broadly. The study of consolidation shades over important distinctions among "consolidated" and "unconsolidated" cases. It also ignores regional and historical contexts that matter for interpreting the strength of democratic experiments.

Another way in which the authors treat consolidation as institutionalization is by seeing widespread support for the "rule of law" as a precursor to making democracy "the only game in town." Yet arguments about the rule of law are often ambiguous about what the law is for. While the study of democratization must focus on increasing the accountability of elites to the demos, arguments about the rule of law tend to center on an entirely distinct, yet related, concept: order. Many works in the Themes volume are primarily concerned with order, or what one title in the work calls, "designing coherent government."⁽¹²⁾ Linz and Stepan are intuitively correct to argue that democracy needs order, but order, as Samuel Huntington suggested some years ago, does not necessarily need democracy.⁽¹³⁾ What is the importance of the "rule of law" for democracy beyond the foundational need for governability? Most important, what are the tradeoffs between creating order and increasing the accountability of elites?

Establishing the "rule of law" is, at least in part, a euphemism for building the state. According to Linz and Stepan, the absence of a "usable state" reduces social governability and thus the lifespan of democracy. Of course, all of the authors do not envision just any state, but a certain type of state. For example, in transitions from military authoritarianism, the state must develop "proper governmental structures" that allow civilians to exert "ministerial control" over the armed forces.⁽¹⁴⁾ The democratic state, therefore, must give democratic leaders the capacity to exert authority over society and the economy. In this sense, democracy-building and state-building are mutually reinforcing.

Historically, however, state-building in both developed and developing countries was a decidedly illiberal, often bloody, process that was not at all concerned with increasing the accountability of elites to the demos. In their attempts to establish authority over territory, states have used civil wars, "dirty wars" against "disloyal" oppositions, and democratic usurpations in the name of "law and order" - all efforts designed to subject the demos to the will of rulers. Some of these tendencies are even apparent in the "consolidated" democracy of Spain. There, democratic leaders continue to struggle with Basque terrorists and the legacy of an illegal dirty war waged by the Socialist government against them during the 1980's. In Latin America, violence in the poor barriadas of Lima and the favelas of Rio is, in part, the result of the absence of the state, but it is also a result of schizophrenic attempts by state agents to exert their authority.⁽¹⁵⁾ These and other cases confirm that state-building, alas, even for democracies, is indeed "organized crime."⁽¹⁶⁾ Although the authors agree that the existence of a state is essential for democracy, they do not consistently analyze how democratic states emerge despite the basic tendencies to use illiberal means to consolidate state power. This can be seen as another time frame conceptualization problem since the undemocratic methods used to strengthen state authority in the short-term lay the groundwork for the "usable state" required of consolidated democracy in the long-term.

Most important, the authors (with the exception of James Cotton in Regional Challenges) do not consider how globalization constrains the utility of the state. Latin American states are not too different from East Asian states in terms of their increasing dependence on foreign capital and their sensitivity to international economic fluctuations. Under pressure from financial markets, social spending in both regions has been downsized in order to redress fiscal imbalances and respond to monetary crises. In Singapore and Malaysia, political leaders have voiced the concern that the redistributive policies associated with liberal democracy may make their countries less attractive to foreign investment.⁽¹⁷⁾ Linz and Stepan argue that democracy requires a "usable state," but they ignore whether the state usable by democrats is the same one required by foreign capitalists. If it is not, then the "international factor" of globalized capitalism may be a much more negative force on democracy than these authors assume.

INSTITUTIONALIST ARGUMENTS

As with much of the literature on democratic consolidation, these works emphasize the ways in which the interests of civil society are aggregated and represented by democratic institutions. Much of the discussion about formal rules in these works deals with the role of particular institutions: parliamentary v. presidential systems, electoral rules, party structures, etc. The authors are careful to point out that they are not describing a singular, guaranteed path to consolidated democracy. Despite this caveat, many specify particular institutional arrangements for democracy that they believe are superior to the alternatives.

Despite extensive discussions of the importance of institutions, overall the authors do not arrive at a consensus concerning which rules matter. Przeworski et al. in Themes and Linz and Stepan in their book argue forcefully that parliamentary systems are more stable and efficient in many capacities than presidential systems. By contrast, John Carey in Themes points out that these distinctions are not helpful analytically. In practice, there is a tremendous diversity among presidential and parliamentary regimes regarding the constitutional powers allocated to presidents and legislatures (p. 80). Many presidential systems, for example, have formal rules that grant legislatures control over cabinets. Such institutions promote the kind of linkages between the executive and the legislature that enhance the stability and efficiency of parliamentary systems. Similarly, Emerson Niou and Peter Ordeshook in Themes argue, with reference to Taiwan, that the degree of "integration" of self-interest and common action among legislators, presidents, prime ministers, local and provincial governments explains the efficacy of government. Since different institutions promote distinct degrees of integration within both presidential and parliamentary systems, the formal distinction between the two is unnecessary.

Beyond their differences concerning institutional prescriptions, the authors are not always clear about the more important question of how institutions and democracy are related. Most of the works only beg the question with incomplete or inconsistent analyses.

One example of the incompleteness of institutionalist arguments is John Carey's contribution. Carey's sophisticated arguments on behalf of electoral rules that encourage partisan as opposed to personal voting and strong leadership control over party labels explain the factors that directly affect party cohesion. Within this framework, the most cohesive parties emerged in Israel, Argentina, and Spain. Despite their strong parties, the accountability of political elites in these countries has faltered. In Israel, electoral rules have made parties cohesive, but they have also empowered a large number of extremist religious parties that have succeeded in limiting the range of political debate. "Delegative democracy" in Argentina has allowed President Carlos Menem to skirt legislative and judicial checks. Spain is embroiled in a scandal involving an illegal anti-terrorist campaign waged by otherwise democratic leaders. In these cases, party cohesion has not made people feel more represented by their leaders.

It is not obvious that party cohesion and democracy are even positively related. In theory, Michel's "iron rule of oligarchy" argues that strong parties are representative of the political elite, not the larger demos. The "strong, cohesive parties" many scholars of regimes embrace as good for democracy,⁽¹⁸⁾ may become unrepresentative if they ignore local political interests. This has been one prominent cause of public disaffection with "partyocracy" in Venezuela.⁽¹⁹⁾ Strong parties may also abuse power. Cotton makes this point clear with his description of the Koumintang's (KMT) strong-arm party tactics in Taiwan and similar experiences in Singapore and Malaysia.

Concerning the inconsistency of institutionalist arguments, many of the authors do not clearly define the importance of institutions even on questions that test the role of formal rules. Felipe Agüero's study of civil-military institutions in Themes is one example. Agüero argues that civilian governments that successfully reverse the institutional constraints placed on them by outgoing military governments, acquire civilian supremacy over the armed forces. The focus on institutions is essential in this case since civil-military relations are mediated primarily by "legal-formal norms" (p. 178, fn. 8). Nonetheless, in two essential examples - Spain and Argentina - Agüero credits ill-fated coup attempts in 1981 and 1990, respectively, as "decisive" in reorganizing civil-military relations. Suddenly, the major causal factors are serendipitous events over which neither civilians nor institutions have deliberate control (p. 183).⁽²⁰⁾

Formal institutions are important for the quality of democracy, but they are neither unique causes nor unqualified indicators of democratization (and they certainly cannot be both!). Institutionalist arguments must be able to specify how these rules affect democratic processes and outcomes. How do institutions make elites more accountable to the electorate? How do formal rules affect the interests of key democratic players? Do institutions create, to follow Diamond, prodemocratic "behaviorial patterns" in elites and the citizenry? To what extent do formal institutions set the stage and to what extent do they tell the story?

Schmitter in Themes provides a format that helps move this debate forward. In bringing civil society back into the discussion of democratic consolidation, the author disaggregates democracy into a variety of different "systems of [interest] intermediation." All are potentially compatible with democratic consolidation but each has distinct implications for the "performance, distribution of benefits, and 'quality' of the democracy that emerges" (p. 242). Political parties, electoral institutions and constitutions are important, but they are unsuited to represent the multifarious interests of civil societies in modern democracies. Schmitter argues that contemporary democracies have multiple channels of representation and various points of decision making, many outside the purview of political parties and electoral politics. Democracy is not a singular regime but a set of "partial regimes." Each partial regime has a set of institutions and conflict-resolution mechanisms. Schmitter's framework serves as an alternative means for addressing the connections between institutions and democracy; one that illustrates the limits of institutionalist arguments that focus exclusively on national, formal rules. Schmitter's approach also facilitates empirical research by disaggregating complex regime systems. Despite this author's inclusion in the Themes book and his role as a critic at the conference that led to the two-volume set (which is also published as a single, hardback book), few of the authors move on his suggestions. Although we cannot fully explore it here, Schmitter's nuanced approach deserves much more attention.⁽²¹⁾

THE POLITICAL ECONOMY OF BUILDING DEMOCRACY

The relationship among economic development, reform, and democracy adds an additional level of complexity to the study of the "third wave." Economic Reform and Democracy and some of the authors of the other volumes attempt to cull propositions from this complexity on two levels: 1) does economic development enhance democratization broadly? and, 2) is the success of economic reform and democratization mutually reinforcing?

On the first question, many of the authors agree with the pioneering work of Seymour Martin Lipset that "the more well-to-do a nation, the greater the chances that it will sustain democracy" (1959, 56). In their cross-national quantitative study of 224 regimes, Przeworski et al. in Themes conclude that the risk of a breakdown of democracy falls as annual per capita income increases. Beyond $6,000, "democracies are impregnable and can be expected to live forever" (p. 297).⁽²²⁾ Democracies in poorer regions will probably endure if their economies grow, but they are not guaranteed immortality.

The macro correlations between level of development and the persistence of democracy are evocative, but they do not explain why, in some cases, development does not foster democracy. Indeed, such correlations do not reveal any of the causal mechanisms between development and democracy. Comparative historical studies such as Barrington Moore's (1966) show that not every path to development is paved with democratic intentions. The "Junker route" in Germany, for example, led to industrial development but also set the stage for authoritarianism. By comparison, Britain achieved a parliamentary democracy, but the route it took was decidedly different from that of the United States or France. None of these pathways, however, will be repeated so precisely since each was the product of a particular historical context. This concern for historical context and timing is rarely seen at the center of the macropolitical theories in the works under review. Instead, the cross-national statistical contributions tend to treat the causal factors underlying the link between development and democracy as a black box.⁽²³⁾

In their book, Linz and Stepan use comparative historical methods to provide a rationale for the relationship between economic development and democratization. They argue that modern consolidated democracies require a set of sociopolitically crafted and accepted norms, institutions, and regulations - "economic society" - that mediate between the state and the market. Pure command and market economies without these regulations produce undemocratic outcomes. Democracies persist where economies are mixed and economic societies are strong.

Like the macropolitical arguments, however, the comparative historical approach suffers from a similar tendency to create causal arguments around black boxes. The authors' study of the East European cases demonstrates that the development of an economic society is an important precursor to democratization. How such an economic society is built, however, remains a mystery. The assumption is that the mediations of an economic society are outgrowths of socioeconomic development and "effective states." Yet Schmitter in Themes provides a powerful corrective to this assumption by arguing that civil societies (and economic societies that emerge from them) are, in part, the creation of public policies that extend legal recognition, rights of assembly, fiscal resources (to name only a few possible policy elements) (pp. 248-249). Private practices such as the creation of class-based or professional associations and the social capacity to engage in other forms of collective action, are also precursors to the rise of economic societies. Far from being the natural offspring of economic change or indeterminate state intervention, the institutions of economic society are constructed by a mix of state and social action.

Despite suggestive case studies, most of the works under review take these processes as given without analysis of their historical origins, context, or timing. The authors do not generalize consistently about the conditions under which particular class actors (the bourgeoisie; labor; the middle class) may, by themselves or in alliance with other class actors and the state, lead a route to democracy.⁽²⁴⁾ Indeed, some of the historical case studies call out for such theorizing as they suggest that class actors typically associated with democracy may act against it under certain circumstances. Hsin-Huang Michael Hsiao and Hagen Koo in Themes argue that, in South Korea, economic development created a politically mobilized middle class which favored democratization during the 1980's. Yet South Korea's middle class also proved to be a conservative ally of business and, consequently, a supporter of antilabor legislation during the 1990's (pp. 320-321). Under what conditions does the middle class act in the interests of democracy by supporting an alliance with workers, and when does it act in its own economic interests and against those of labor? Mere reference to the existence of "mixed economies" does not provide an answer to such questions.

Economic Reform and Democracy makes important contributions to the study of how regimes and economic policy affect each other. The essays by José María Maravall and Barbara Geddes, for example, add significantly to the emerging consensus that authoritarianism does not make successful market-oriented reform more likely.⁽²⁵⁾ In fact, democracies may be better at economic reform due, in part, to their stronger base of diverse support and their capacities for reducing the transaction costs of reform for their populations.

Having established that in theory, the volume presents a somewhat mixed view of exactly how democracy affects economic reform and vice versa. For example, on the question of "sequencing" - which should go first: political or economic reform - the authors are unsure. Linz and Stepan and a number of the Economic Reform and Democracy authors argue that political reform should precede economic reform as it did in Spain. Once the rules of the political system are established, economic conflicts can be put to rest without having them upset democratization. Economic reforms may also precede political reforms as they did in Chile, but that assumes that authoritarian regimes are better than democracies at managing economic restructuring. The main point is that "simultaneity" - enacting political and economic reforms at the same time - is the worst possible option.

Haggard and Kaufman, however, note that the intensity of the economic crisis may make simultaneity inevitable. Only where economic conditions are good can reformers "give the right-of-way to the building of a democratic consensus" (p. 8). Therefore, the "choice" of sequence is not a choice at all, but a structural characteristic of countries engaged in "dual transitions."

Moving beyond the facile "state versus market" debate, many of the volume's contributors argue that "neoliberal" reforms require a "strong state." Indeed, many of these reforms are designed to strengthen the state's regulatory, public goods, and taxation functions.⁽²⁶⁾

However, it is not clear that the elements of "state strength" are entirely consistent with democratization. Joan Nelson, echoing the arguments of other contributors, asserts that reformist governments must be open enough to consult societal actors on the content of economic reform, but closed to ("autonomous of") "rent-seeking networks" looking to reestablish their privileges. Evidence from the East European and Latin American cases, however, shows that reformist states have created coalitions of support for reform through the redistribution of privileges and costs. Consultation has not been broad but concentrated on large conglomerates and business associations.⁽²⁷⁾ Therefore, reformist states have tended to build up their "strength" through a process of picking privileged "winners," not through democratic consensus-building.

It is also not clear how the state in developing countries can become "strong" in a world economy that favors state acquiescence to the interests of global capital. In this context, expanding the regulatory power of the state confronts the interests of business in reducing "state interference" and deregulating the private economy.⁽²⁸⁾ Hungry for foreign finance, developing country states will reduce controls on the economy in order to attract investors.⁽²⁹⁾ As with democratic consolidation, the "strong state" required by reformers may not be the same one required by capital. The conflict between the two can produce weakened states and hollow reforms.

Finally, references to regime type do not fully reveal the immense economic problems facing developing countries. Discussion of Africa's plight by Thomas Callaghy and Adebayo Adedeji in Economic Reform demonstrates that most of the continent's problems have to do with the weakness of states and the rising "specter of anarchy." As Linz and Stepan would argue, these questions must be resolved before democracy is even considered. Similarly, the exigencies placed on states in Latin America for improving fiscal management and economic competitiveness, work independently of regime type. Both authoritarian and democratic countries face these constraints as long as they are engaged in the international political economy. Not surprisingly, the responses of middle class and labor organizations to the social dislocations produced by reforms have focused more on differences in policy approaches than in regime type.

COMPARATIVE LESSONS FROM AND FOR LATIN AMERICA

The works under review expand the empirical study of democratization to cases in East Asia, Eastern Europe, and Russia. The attempt, most apparent in the Linz and Stepan book, is to create a universal set of propositions applicable to all empirical cases of democratization. Yet, as Michael McFaul notes in Regional Challenges, "the farther east one goes, the less fitting the democratic transition and consolidation models become" (p. 64).

The biggest problem of comparison is that the issues that China, Russia, and the East European countries face are far more fundamental than those addressed by South American and South European polities during their democratizations. Post-communist countries lacked (for the most part) autonomous civil societies, private property, and capitalist states. Democratization in Russia, for example, was affected by the need to construct "a new state, a new political system, and a new economic system simultaneously."⁽³⁰⁾ Larger issues such as creating capitalist markets and building trust over ethnic hatred became essential preconditions for analyzing the prospects for democracy-building. The institutional and elite-centered variables of the more conventional studies of regime transitions were only ancillary in China, Russia, and Eastern Europe.

Such cross-regional differences usefully clarify certain issues by highlighting the historical advantages of democracy-builders in Latin America. For example, the existence of "flattened" civil societies in some of the East European cases facilitated the political recovery of authoritarian leaders. In these cases, communist parties either retained control (Bulgaria) or exerted prerogatives over the post-transition regime (Hungary, Poland, and Czechoslovakia). Although these same authoritarian prerogatives persisted in some of the Latin American cases (e.g., Chile), the existence of a functioning and experienced civil society prior to regime liberalization made the Latin American countries comparatively better prepared for limiting the threat of an authoritarian "reverse wave." By contrast, this danger persists in Russia and some East European countries.

On other questions, the cross-regional comparisons do not develop new insights into the study of Latin American democratization. For example, the "divided nationhood" of many of the postcommunist states added a level of complexity to these transitions to democracy that was not present in the Latin American cases. In none of the Latin American countries were sophisticated consociational and federal institutions necessary to avoid ethnic-based and nationalist attacks upon the legitimacy of democracy as they were in the Baltics, Balkans, and the post-Soviet republics.

To be useful, such cross-regional comparisons must focus on the more basic questions common to both sets of cases: democracy and its relationship to the market and the state. By spending much effort on the conventional issues of regime transitions and consolidation (e.g., electoral rules, elite pacts, etc.), the works under review pay high opportunity costs. They fail to ask larger questions about the relationship between democracy-building and state- and market-building.⁽³¹⁾ Eastern Europe and the post-Soviet countries provide a useful laboratory for addressing these fundamental issues which are also essential to any analysis of the quality of democracy in Latin America. For example, how reformers in Russia deal with corporatist state-business relations may inform how democratic reformers in Brazil deal with patrimonial business ties to the state.

"Going east" in the comparative analysis of democratization does produce some important propositions for Latin America if the focus is on East Asia. Constitutionalist formulas for strengthening the National People's Congress in China (the subject of Andrew Nathan's essay in Regional Challenges) may be a model for smoothing the transition to a post-Fidel Cuba. The strategies of the tangwai (literally, non-KMT) opposition in Taiwan present useful experiences for Mexico's non-PRI parties. In both countries, political societies are making not so much a transition from authoritarianism to democracy as from a hegemonic party system to a truly multiparty democracy. Finally, the East Asian cases remind us that liberalization still matters. Recent experiences with liberalization in East Asia provide valuable insights for Cuba and China where further liberalization may be all that is possible for the medium-term if not longer.

CONCLUSION

All four of the books under review represent important contributions to the larger debate on the continued democratization of the "third wave." Nevertheless, they also illustrate some of the frustrating limitations of the research program. Central concepts such as "consolidation" continue to be heavily contested, even by scholars such as O'Donnell who initially advocated their use. The ambiguities of this concept have long been an obstacle to understanding the evolution of "unconsolidated" democracies. Allied concepts such as "the rule of law" and "institutionalization" are conceptually less muddled but are inconsistently related to democratization. Deepening the rule of law and building cohesive institutions are often insufficient for and even counterproductive to efforts to expand the accountability of elites.

The study of "partial regimes" is a useful alternative to the failures of the sweeping concepts of the third wave literature. By disaggregating complex political processes, this approach allows for more precise specification of which institutions, interests, and processes advance democratic procedures and outcomes. The parsimony that is lost by jettisoning generalizable concepts such as "consolidation" is replaced with the greater explanatory power of arguments that bring the "quality of democracy" back into the analysis.

The works under review do a good job of highlighting the major debates in the political economy of the third wave democracies, but they do not organize these views into a coherent theoretical framework that advances new insights. On issues such as the use of comparative methods, the importance of the sequencing of reform, and how to think about the role of the state, the books provide conflicting and incomplete treatments.

Applied to Latin America, the comparative case studies of Eastern Europe, Russia, China, and East Asia produce some partial insights, but they ask larger questions that go unaddressed by the works under review. Assessing the third wave will require more attempts like these to link empirical evidence to workable propositions that can be tested across regions and national cases. The persisting problems of democracy in Latin America and other third wave regions require that new perspectives emerge from the new propositions.

1. The terms "third wave democracies" comes from Samuel Huntington's (1991) study. The authors of the volumes under review include the East Asian cases in this set, although Huntington could not at the time his study was done.

2. These arguments build on recent work by a couple of the main contributors in the Journal of Democracy. See Diamond (1996) and O'Donnell (1994).

3. This is a recognized limitation of the many uses of "consolidation." See Schneider (1995, 219).

4. Linz and Stepan (1996, 6).

5. Ibid.

6. Ibid.

7. The felicitous phrase, "only game in town," comes from Linz (1990, 156) and Przeworski (1991, 26).

8. Several scholars who defend the use of the term "consolidation" admit the difficulty of assessing a "cutting point" for disloyal action. See Gunther et al. (1996, 153).

9. Linz and Stepan accept that their arguments regarding an end state of consolidated democracy are teleological. But they give weight to both structural and agent-based causes and are wary of "overdetermining" democracy. See Linz and Stepan, Themes, fn 3. For an assessment of "consolidation" as inherently teleological, see O'Donnell (1996).

10. See, for example, Diamandouros, Regional Challenges, pp. 8-10.

11. Shortly after the impeachment of Collor, opinion polls registered the highest support for democracy since 1989. Some observers see these trends as evidence of a new level of civic awareness and regard for democracy following successful fights against corruption. See Pinheiro (1997, 269).

12. Niou and Ordeshook, 160.

13. See Huntington (1968). For analysis of Huntington on this point, see Kesselman (1973).

14. Agüero in Themes, 178.

15. See Burt (1997) and Pinheiro (1997). I thank Desmond Arias for this thought.

16. For a theoretical discussion of this argument, see Tilly (1985).

17. Cotton, Regional Challenges, 116. Chu, Hu, and Moon in Regional Challenges note that the governments of the East Asian NICs have often seen democracy as the seed of political instability and, therefore, as undesirable (p. 272).

18. One example is Edgardo Boeninger's observation in Regional Challenges that the only democracies to survive the "reverse wave" to authoritarianism in the 1960's and 1970's in Latin America were those with strong party systems and little ideological polarization (pp. 30-31).

19. See Coppedge (1994).

20. The role of Fortuna as a deus ex machina in democratic consolidation is a persistent problem in this literature. See Remmer (1991, 483-485).

21. See Schneider (1995, 221).

22. The $6,000 figure is expressed in purchasing-power-parity U.S. dollars in 1985 prices.

23. Cross-national quantitative and comparative historical approaches are contrasted in Rueschemeyer, Stephens, and Stephens (1992, 12-36), who come to similar conclusions about the comparative advantages of approaches such as Moore's.

24. Useful alternative analyses which do formulate such propositions are Collier & Collier (1991) and the aforementioned Rueschemeyer, Stephens, and Stephens (1992) and Moore (1966).

25. Other significant works on this question are Haggard & Kaufman (1995); Remmer (1990); Maravall (1995).

26. Haggard and Kaufman, 5; Naím. This point is also made by Kahler (1990, 47) and Schamis (1994).

27. See Silva (1996) and Acuña (1994).

28. See Schmidt (1995, 78-81).

29. This is particularly true of the demise of capital controls. See Haggard and Maxfield (1996).

30. McFaul, p. 65.

31. An excellent example of just such a work which employs comparisons between Latin America and Eastern Europe is Przeworski (1991).

ACUÑA, Carlos. (1994) "Politics and Economic in the Argentina of the Nineties (Or Why the Future No Longer Is What It Used to Be)," pp. 31-73 in William C. Smith, Carlos H. Acuña & Eduardo A. Gamarra (eds.) Democracy, Markets, and Structural Reform in Latin America. New Brunswick, NJ: Transaction Publishers.

BURT, Jo-Marie. (1997) "Political Violence and the Grassroots in Lima, Peru," pp. 281-309 in Douglas Chalmers, Carlos Vilas, Katherine Hite, Scott B. Martin, Kerianne Piester, and Monique Segarra (eds.) The New Politics of Inequality in Latin America: Rethinking Participation and Representation. New York, NY: Oxford University Press.

COLLIER, Ruth Berins and David COLLIER (1991) Shaping the Political Arena: Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Princeton, NJ: Princeton University Press.

COPPEDGE, Michael (1994) Strong Parties and Lame Ducks: Presidential Partyarchy and Factionalism in Venezuela. Stanford, CA: Stanford University Press.

DIAMOND, Larry (1996) "Is the Third Wave Over?" Journal of Democracy 7, 3 (July): 20-37.

GUNTHER, Richard, P. Nikiforos DIAMANDOUROS, Hans-Jürgen PUHLE (1996) "O'Donnell's 'Illusions': A Rejoinder." Journal of Democracy 7, 4 (October): 151-159.

HAGGARD, Stephan and Robert R. KAUFMAN (1995) The Political Economy of Democratic Transitions. Princeton, NJ: Princeton University Press.

HAGGARD, Stephan and Sylvia MAXFIELD (1996) "The Political Economy of Financial Internationalization in the Developing World," pp. 209-239 in Robert O. Keohane and Helen V. Milner (eds.) Internationalization and Domestic Politics. New York, NY: Cambridge University Press.

HUNTINGTON, Samuel P. (1991) The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press.

______. (1968) Political Order in Changing Societies. New Haven, CT: Yale University Press.

KAHLER, Miles (1990) "Orthodoxy and Its Alternatives: Explaining Approaches to Stabilization and Adjustment," pp. 33-61 in Joan M. Nelson (ed.) Economic Crisis and Policy Choice: The Politics of Adjustment in the Third World. Princeton, NJ: Princeton University Press.

KESSELMAN, Mark (1973) "Order or Movement? The Literature of Political Development as Ideology." World Politics 26, 1 (October): 139-154.

LINZ, Juan J. (1990) "Transitions to Democracy." Washington Quarterly 13 (Summer).

LIPSET, Seymour Martin (1959) "Some Social Requisites of Democracy: Economic Development and Political Legitimacy." American Political Science Review 53 (March): 69-105.

MARAVALL, José María (1995) Los resultados de la democracia. Madrid, Spain: Alianza Editorial.

MOORE, Barrington (1966) Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston, MA: Beacon Press.

O'DONNELL, Guillermo (1994) "Delegative Democracy." Journal of Democracy 5, 1 (January): 55-69.

______. (1996) "Illusions About Consolidation." Journal of Democracy 7, 2 (April): 34-51.

PINHEIRO, Paulo Sergio (1997) "Popular Responses to State-Sponsored Violence in Brazil," pp. 261-280 in Douglas Chalmers, Carlos Vilas, Katherine Hite, Scott B. Martin, Kerianne Piester, and Monique Segarra (eds.) The New Politics of Inequality in Latin America: Rethinking Participation and Representation. New York, NY: Oxford University Press.

PRZEWORSKI, Adam (1991) Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. New York, NY: Cambridge University Press.

REMMER, Karen (1991) "New Wine or Old Bottlenecks? The Study of Latin American Democracy." Comparative Politics (July): 479-495.

______. (1990) "Democracy and Economic Crisis: The Latin American Experience." World Politics 42, 3: 315-335.

RUESCHEMEYER, Dietrich, Evelyn Huber STEPHENS, and John D. STEPHENS (1992) Capitalist Development and Democracy. Chicago, IL: University of Chicago Press.

SCHAMIS, Hector E. (1994) Re-forming the State: The Politics of Privatization in Chile and Britain. Ph. D. Dissertation, Columbia University.

SCHMIDT, Vivian A. (1995) "The New World Order, Incorporated: The Rise of Business and the Decline of the Nation-State." Dædalus 124, 2 (Spring): 75-106.

SCHNEIDER, Ben Ross (1995) "Democratic Consolidations: Some Broad Comparisons and Sweeping Arguments." Latin American Research Review 30, 2: 215-234.

SILVA, Eduardo (1996) The State and Capital in Chile: Business Elites, Technocrats, and Market Economics. Boulder, CO: Westview Press.

TILLY, Charles (1985) "War Making and State Making as Organized Crime," pp. 169-191 in Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol (eds.) Bringing the State Back In. New York, NY: Cambridge University Press.