Final Papers

Dan Blazo

1. SOCIAL MOVEMENTS, REVOLUTIONS, AND CIVIL SOCIETY IN COMPARATIVE POLITICS

Although the subfield of Comparative Politics claims an extraordinarily broad scope of analysis, certain sociopolitical issues are generally regarded as more intriguing or more important than others. Political revolution is one such issue. In some ways, revolution is the most fundamental research topic for Comparative Politics, for a domestic transition of regime-types typically affects entire populations in more salient ways than any other political phenomenon. Moreover, the particularities of each revolution invite the comparativist to search for patterns of social behavior. Theoretical constructs from other social sciences have been borrowed, compared, and refined to fit the particular research questions. Classical democratic theories are continually modeled, quantified, tested against one another, and remodeled accordingly. Thousands of case studies and hundreds of cross-national comparisons have been the product of comparativists’ dire yearning to answer the question: Why revolution? Indications of progress in solving this timeless riddle are evident in the elucidated parameters and integration of the relevant research. As scholars have cultivated diverse approaches to solving the riddle, a collaboration of their findings has shed light upon certain macro-theoretical deficiencies as well as measurement issues that must be addressed in order to maintain the scientific integrity of comparative solutions.

In his 1965 seminal work on the dilemma between collective action and rational choice, Mancur Olson suggested that “only a separate and ‘selective’ incentive will stimulate a rational individual in a latent group to act in a group-oriented way” (2). This dilemma became known as “Olson’s problem,” the “free rider problem,” or “the Collective Action Problem” among subsequent researchers, and remains a major logistical obstacle in comparative studies on revolution despite more than four decades of research. What social forces and/or personal incentives compel individuals to risk their lives in revolutionary attempts? Given that collective action is necessary for any successful revolutionary effort, the inclination to “free ride” must be either dismantled by social “virtues” or perceived as disadvantageous for the personal aspirations of revolutionaries. Most scholars in the comparative subfield have approached the issue from the latter hypothesis, for which individuals’ incentives for participating in the revolutionary effort are preferable over the free rider alternative. However, scholars are quick to point out that the appeal of a “free ride” is inextinguishable when the alternative involves risking one’s life. This theoretical dilemma has led scholars to overlook the fundamental link between individuals’ instigations of revolutionary efforts and the social constructs that help determine the outcomes of such efforts.

Jeffrey Berejikian’s 1992 article, "Revolutionary Collective Action and the Agent-Structure Problem,” reveals the ongoing theoretical difficulties in accounting for the free rider problem. Berejikian explains how two distinct approaches have come to dominate the comparative research on revolutions: Methodological Structuralism and Methodological Individualism. The former is essentially the application of neo-institutionalism to the study of revolution, as epitomized in Theda Skocpol's States and Social Revolutions (1979). Methodological Individualism, on the other hand, is a rational choice approach exemplified in Samuel Popkin’s book, The Rational Peasant (1979). Both approaches fail to draw a link between agents and structures, or, between individuals’ motivations to revolt and collective revolutionary efforts. Popkin concentrates exclusively on the agent, while Skocpol’s focus is almost entirely on the structure. As a consequence of implicitly denying the other’s legitimacy, both methodologies fail to provide solutions to the free rider problem.

At the heart of the agent-structure problem is the fact that agents and structures are ontologically inseparable (at least in the context of this research topic). Social structures are made up of individuals, who are concurrently agents of social activity. Both methodological structuralism and methodological individualism approach revolution by means of conceptual reductionism, focusing on either the agent or the structure. The interaction between agents and structures is fundamental for the outcomes of each: individuals affect social structures and vice versa. Agents, structures, and their interactions are all embodied in a single human being, just as they are in a widespread social movement.

Skocpol’s focus on structures leads her to extrapolate class preferences across individuals of the particular social class. This leads to some theoretical problems. Although peasants may collectively benefit from a successful revolution, individual peasants are unlikely to affect the outcome of the revolutionary effort. In addition, peasants are likely to benefit from a class revolution even if they were not participants in the revolutionary struggle. Free riding is arguably the most rational option for the individual peasant. Berejikian addresses the contradiction that arises from Skocpol’s theoretical gap:

“The assumption of rational maximization as a decision rule on the part of individual peasants does not logically lead to the conclusion Skocpol's analysis assumes, namely, that peasants will collectively act to address their class-defined goals and aspirations. Thus, while the structural hypothesis directs itself to explaining the timing and scope of peasant revolts, the individualist assumptions implicit throughout would seem to argue for a nonevent” (649)

To provide further evidence of the “missing link” in Skocpol’s analysis, Berejikian mentions several historical examples for which Skocpol’s enumerated structural conditions for revolutions have not led to a peasant revolt. These findings imply that social class interests are best understood as overlapping individual interests. However, comparativists must keep in mind the fact that individuals will always have incentives that are not class-oriented, such as personal safety during a time of war.

Methodological Individualism, as portrayed in Popkin’s Rational Peasant, also fails to provide a coherent theoretical explanation for a revolution’s surmounting of the free-riding inclination. Popkin’s defense of rational choice theory in the context of revolutions relies upon the rationality of maintaining a positive reputation among members of a community. In other words, a peasant will join a revolutionary effort if refusal to join would result in his ostracization among fellow villagers. However, this concept of rational participation in a revolutionary effort is unable to explain the incentives for those individuals who joined the revolution before it became so socially obligatory that refusal incited ostracization. Social pressure to revolt derives from individuals, and its effectiveness depends, in part, on the number of individuals applying the pressure.

Unlike Skocpol, Popkin explicitly confronts the inclination to free ride. His use of social pressure to justify the rationality of revolting, however, implies that structure does indeed affect the agent’s incentives. Conversely, according to Berejikian, Popkin frames revolutionary efforts as, “the aggregation of intentional individual actions, rather than permissive structures, accounts for revolution” (650). In contradiction to the Popkin’s focus on the individual, social pressure is a structural component that Popkin views as influential upon individual rationality. In addition, Popkin’s complementary justifications for rational revolt rely upon the manipulation of private incentives regarding collective goods. For a peasant revolution to gain sufficient support, all peasants must accordingly recognize their direct participation in the revolution as necessary for the final outcome of collective good distribution. Not only is this consensus improbable, it is incomplete, for it emphasizes economic rationality without acknowledgement of the uselessness of economic outcomes for those who lose their lives in the struggle. The real dangers of revolution alter the dynamics of incentive structures typically employed in rational choice models. The free riding tendency is especially appealing (or “rational”) when participation in collective revolt sharply increases the likelihood of violent death.

The conceptual obstacles exposed by Berejikian can essentially be described as the effects of utilizing a narrow approach without regard for the insight of alternative approaches. The result of using a narrow approach is a failure to account for the interactions between agents and structures. This failure is revealed in the theoretical deficiencies of both Methodological Structuralism and Methodological Individualism, specifically, in their failure to adequately address the free rider problem in the constructs of their theories. While the structural approach overlooks individuals’ incentives to revolt in spite of the likely insignificance and high level of personal risk, the individualist approach is unsuited for explaining the original collective formation of a revolutionary effort, as it fails to address the structural context for which revolution is amenable.

The conceptual obstacle of explaining collective action and the free rider problem has been the topic of much scholarly literature over the past two decades. Despite the vast amount of theory and research intended to model collective action, however, Mark Lichbach concludes that “…most scholars recognize that the marriage between the CA (Collective Action) research program and conflict studies has largely failed” (1994 9). The obstacle that has yet to be surmounted, according to Lichbach, is the formulation of a “useful modification of a solution to the free rider problem” (30). Finding a solution to the free rider problem is not the obstacle, as around two dozen different solutions have been developed and refined in theoretical works. The difficulty lies in modeling a solution for use in testable comparative models. The mere presence of so many different solutions has caused conceptual dilemmas for any research seeking to analyze the causality chain of social occurrences.

Each solution to the “Rebel’s Dilemma” is somewhat predictable, as they require strategic planning on behalf of the rebelling parties. This predictability ensures that the regime in power—whose interests must lie in direct opposition to the rebels—will attempt to counteract the strategies (or “solutions”) employed by the rebels. Therefore, gauging the usage and effectiveness of such strategies is extremely difficult, especially for empirical and cross-national research. Moreover, the solutions are not mutually exclusive, but rather mutually dependent, in some regards. Lichbach explains: “Any one solution must be an incomplete explanation of the CA… The major difficulty is that each solution presupposes the existence of at least one of the other solutions… While each type of solution might be necessary to either create, maintain, or transform CA, taken independently, none is sufficient” (23). The vast amount of research addressing the collective action problem has clarified some of the ways in which the dilemma may be overcome in practice, but has not resulted in an applicable component for a comparative research model.

Nonetheless, the findings of the CA research program should not be disregarded in comparative studies of revolution. The first measure that should be taken by comparativists wishing to implement these findings is the acquisition of thorough knowledge about the specific societal circumstances of the revolution or revolutions under analysis. Upon acquiring comprehensive information on the relevant societal conditions, certain solutions are likely to emerge as more accurate depictions of the particular rebels’ strategies than others. Researchers can reduce the number of potential solutions in regards to social context and further predict which ones were most influential over the course of particular rebellions. However, they must also keep in mind that “in practice, CA problems are overcome by combinations, mixtures, adaptations, and modifications of the two dozen sets of pure solutions” (Lichbach 27). The social context is crucial in determining which combinations or hybrids of solutions are likely to be effective. In fact, the effectiveness of different solutions relative to the social context is, to Lichbach, more important than the “missing link” between rational choice theory and collective action. He explains:

“…the problematique of the CA research program, as I see it, is to determine the conditions under which a particular CA solution is adopted and the conditions under which a particular solution is effective. In short, we bypass the initial problem of why rational people rebel and never reach the ultimate problem of aggregate levels and outbreaks of rebellion. A midrange approach best elaborates CA ideas” (30)

This approach is not only better suited for integrating CA research into comparative studies, it also provides a more conducive medium for testing CA research in empirical models. Necessary for the advancement of integration between CA research and comparative politics is systematic evidence directed at gauging the specific components of the many CA solutions. This evidence has not been a top priority among the rivaling structural and rational choice approaches, for “…those who have applied CA theories to conflict have focused almost exclusively on the initial problem of whether anyone who is rational will actually participate in protest and rebellion” (9). A more productive course would include an explicit assumption that rational people do rebel, and would aim to identify the most effective strategies used to garner support by providing evidence, rather than mere logic. The evidence is currently trailing far behind the theories regarding conceptual solutions to the free rider problem.

Modeling the solutions to the free rider problem is clearly a difficult task for comparativists. Measurement issues add to the difficulties, as traditional empirical data generally obscures personal motivations and group strategies. Although survey evidence may be appealing for certain research questions on the topic, such data is generally unavailable and unreliable, due to the psychological tendency of misinterpreting personal motives. Any quantitative approach to the issue must be grounded upon intensive qualitative observations of particular societies in order to ensure the validity of the models. Cross-national studies on revolutions are especially prone to yield spurious findings, given the importance of the social context for this particular subject. Such comparisons should not be avoided altogether, however, for the scientific integrity of any comparative research demands the falsifiability of theories. The specific methodological caveats of this research topic are discussed below.

A prime example of the importance of thorough case-level analysis is portrayed in E. Spencer Wellhofer’s 2003 article, “Democracy and Fascism.” Rather than accept the traditional reasoning for the rise of fascist regimes, Wellhofer examines the particular circumstances surrounding the rise of Italian Fascism in 1919-21. The extensive research on the fall of Weimar Germany and rise of the Third Reich has led to the overgeneralization of conclusions, with the German case as the basis for explanation. For example, the formation of particular interclass alliances was generally accepted as a universal requirement for the rise of a fascist regime, due in large part to the interclass coalitions of the Weimar Republic. Wellhofer’s analysis reveals that the rise of Italian Fascism occurred under much different societal circumstances than the rise of the Third Reich. As part of his in-depth analysis of the Italian case, Wellhofer tests the most prevalent theories addressing a link between civil society and democratic sustainability. Exhibiting the importance of cross-approach analysis, Wellhofer includes in his model a test of the rational choice explanation, which is measured according to voting demographics and elections. The format of this piece is intellectually satisfying because it includes the necessary thoroughness of case-specifics as well as tests of multiple alternative explanations. Moreover, the results vindicate the case-level analysis in that they reveal a unique case that does not “fit the mold” of Weimar Germany. Wellhofer is careful not to generalize his conclusions, aside from the encouragement of case-level analysis preceding cross-national research models of revolution.

Wellhofer’s article also emphasizes the particularly rich diversity of civil society across nations. For the same reasons that make the topic of revolution unfavorable to cross-national generalizations, civil society is impossible to measure without a keen knowledge of specific societal attributes. Robert Putnam’s article, “Bowling Alone,” is latent with some of the basic conceptual and measurement problems that scholars should avoid. This article lacks both a research design and a systematic depiction of evidence. Ad hoc conglomerations of evidence are presented to support the author’ claims, yet the sources of evidence are never validated as being representative of the implied trends. Moreover, Putnam’s reliance on a few very specific indicators of civil society strength (Elks Club membership, Boy Scout membership, etc) makes his conclusions vulnerable to social transitions over time, which are not necessarily indicative of fluctuations in the strength of civil society. A more inclusive, systematic approach would produce more reliable findings.

Norris et. al.’s article, “Who Demonstrates?” takes such an approach in examining the strength of civil society. The indicator of civic fortitude is participation in demonstrations. This variable’s strength is its compatibility with cross-national comparisons and time-series analyses. Norris et. al. do not provide the in-depth case study analysis that most research on civil society demands. The authors’ dependent variable, however, depicts interesting trends across nations and other independent variables, such as economic factors. The particular social context certainly affects the extent of participation in demonstrations, and this article helps expose some of these structural and economic influences. As the focus of this article is not civil society per se, demonstration participation may be an imperfect indicator of civic engagement. However, the research design is clearly stated, the evidence used is systematically acquired, and the results certainly cast doubt upon Putnam’s contention: that civil society is atrophying in the U.S.

Comparative research on revolution and civil society is as methodologically problematic as it is intellectually appealing. The empirical standards of social science, as expounded by King, Keohane and Verba in Designing Social Inquiry, suggest some especially difficult standards for research on revolutions. These standards are tremendously important, however, and should be followed as closely as possible. One particularly stringent DSI standard that poses unique challenges to revolution research is falsifiability. KKV suggest that “…to make sure that a theory is falsifiable, choose one that is capable of generating as many observable implications as possible. This choice will allow more tests of the theory with more data and a greater variety of data, will put the theory at risk of being falsified more times, and will make it possible to collect data so as to build strong evidence for the theory” (19). Since revolutions can only be studied after they occur, and research is highly dependent upon the particular social context, researchers face a particularly difficult challenge in formulating conclusions that can be tested against other cases. Nonetheless, this challenge is necessary for progress to ensue, so creative ways of ensuring falsifiability are beneficial to the subfield. For example, Lichbach explains that CA theories are not considered to be either right or wrong, but rather better or worse than one another for explaining social phenomena. Better theories essentially falsify worse ones because they reveal that the worse theory is not the best one in existence. This process leads the conditioning and refining of theories in a competitive environment.

Unit homogeneity, as described by KKV, is another measurement obstacle for cross-national research on revolutions (KKV 91). The importance of social context inhibits the generalization of findings due to unit heterogeneity. Even case study research must take caution when extrapolating its conclusions across time, as social transformations may alter the effects of different variables within the same nation. Another important issue of consideration is the fact that correlation does not imply causation, despite the appeal of drawing such a link. The mere presence of two dozen solutions to the CA problem reveals that a spurious link can easily go undetected when the variables do not specifically gauge the factors for which they are intended. In addition, KKV’s warning: “…do not select observations based on the dependent variable so that the dependent variable is constant,” is often violated in revolution research. When the occurrence of a revolution is the topic of inquiry, researchers often compare nations that witnessed a successful revolution against those who did not. Such comparisons should only be made after the selection of cases and testing of hypotheses; otherwise the case selection will be biased.

The vast array of comparative research on revolutions and civil society has led scholars to endorse and enforce the rules of comparative research. These rules are necessary for the integration of research, and thus the advancement of the subfield. While revolution is an especially intriguing topic of comparative politics, it presents particular obstacles that must be addressed in the research design. The high level of interest in the topic of revolution has most likely encouraged the strict enforcement of social science “rules” upon the relevant literature. This is a positive feature of the research topic, albeit an additional obstacle for scholars to surmount.

4. REGIME TYPES, REGIME TRANSITIONS, AND DEMOCRATIZATION.

One of the most highly developed topics of research in comparative politics is that of regime transition. Within this body of research, the lion’s share of scholarship is focused on the process of democratization. This should come as no surprise to those living in the twenty-first century, for recent history has witnessed the most extensive “wave” of democracy the world has ever witnessed. Today there are more people living under democratic rule than during any previous time in history (Geddes). These developments have essentially given the comparative subfield a unique and important duty, for no other academic school of thought is better equipped to answer the most pressing questions on the topic. What social forces are imperative for democratization? What inhibits the process? Why do some democratic regimes collapse during economic crises while others sustain? Comparativists have made substantial progress in providing solutions to these riddles of history. In the process, they have specified some important conceptual and empirical considerations for future research.

One of the most prominent works on the subject of regime transition and democratization is Samuel Huntington’s 1993 book, The Third Wave: Democratization in the late Twentieth Century. By the time this book was published, it was already clear that the basic concept of “democracy” required a consensus among scholars. While political philosophy has witnessed more than two thousand years of the concept’s evolution, contemporary social scientists require a concise definition for the integration of research. Converting concepts to empirically testable criteria, as is necessary for large n quantitative studies, has proven to be quite difficult, for generic indicators inherently negate some of the unique features of each case. The challenge is to account for the most generalizable variables that denote the features of the concept in mind.

Although it is always difficult to achieve a consensus among social scientists, Huntington makes an impressive effort for developing a definition of democracy that is suitable for the entire subfield of comparative research. This definition, however, is not independent of historical eras, but rather specifically constructed to address the “third wave” of democracy that began in 1974. Timeframe is important to consider for studies of democracy because a static definition will prove irrelevant across historical eras. For example, minority groups and women have historically been unable to participate in elections, yet the exclusion of such governments from the “democracy” category would make the term inapplicable to all governments of earlier times.

At its core, Huntington’s definition is very similar to Joseph Schumpeter’s 1942 explanation, which reads: “The democratic method is that institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote” (6).

Anticipating scholarly debate over the applicability of this definition, Huntington provides five specifications at the onset of The Third Wave, most of which explain the parameters of the “minimal” or “procedural” approach as derived from Schumpeter. Huntington’s second specification is often overlooked in the research. This stipulation concerns the limitations on the power of democratically elected governments, especially in regards to other groups within society. Huntington reminds researchers, “If those democratically elected decision makers become… simply a façade for the exercise of much greater power by a nondemocratically chosen group, then clearly that political system is not democratic” (10). The author later suggests that “such allegations are often made, and they may be true. But they should not be judged to be true until they have been demonstrated to be true. That may be difficult, but it is not impossible” (10). This caveat reveals that even Huntington’s definition lacks the conciseness that further research is likely to demand, and that a more extensive definition of “democracy” will develop as the research progresses. For now, however, very little attention is granted to this specific stipulation. The minimal definition advocated by Huntington is generally the basis of empirical modeling, which compels researchers to forego the assessment of a democratic government’s power limitations and adhere only to the positive screening for “democratic” criteria. In other words, most comparative approaches are unable to detect any “façades.”

An important measurement issue embedded within Huntington’s five stipulations is the empirical treatment of democracy and nondemocracy as either a dichotomous or continuous variable. Huntington is justified in noting the inherent complication of using a continuous variable: that the weighting of indicators is inherently subjective or arbitrary. He asserts that a dichotomous approach is better suited for his particular research project, since the topic is the transition from nondemocratic to democratic government (11). However, this assertion faces legitimate criticism in subsequent comparative research.

Rather than employ a dichotomous variable for democracies and nondemocracies, Epstein et. al.’s 2006 article, “Democratic Transitions” introduce a trichotomous scale to include an intermediate category for “partial democracies.” If the transition from one regime-type to another is the issue of consideration, they argue, the regimes in the process of such transitions ought to be examined. The authors’ empirical research suggests that “partial democracies” are unique, in many ways, from both authoritarian regimes and democratic ones. These governments are defined as having some, but not all components of democracy. They are demonstrated to be far more unstable than either of the other two regime-types. In addition, they are most likely to go to war. Most importantly, however, is Epstein et. al.’s finding that dispels the general wisdom of the subfield: “that higher incomes per capita significantly increased the likelihood of democratic regimes, both by enhancing the consolidation of existing democracies and by promoting transitions from authoritarian to democratic systems” (566).

An important contribution of Epstein et. al.’s article is the methodical deconstruction PACL’s prominent work on democratization. This work serves as the means of comparison for use of the trichotomous variable, and one of the fundamental of PACL’s conclusions is discredited on several accounts. First of all, Epstein et. al. reveal that PACL reported the wrong p values in their original analysis, and that the correct p values lend support for the opposite conclusion reached by PACL. The broader lesson here is that empirical analysis must be meticulously proofread and understood by those who employ it. Moreover, it is always a good idea to retest the models of other research before utilizing findings in one’s own research. By the time PACL’s error term misrepresentation was discovered—several years after the original publication—the misinterpretation had been accepted as common knowledge among comparativists. While sophisticated statistical models provide new opportunities to comparative research on regime transformation, a thorough comprehension of the statistical techniques must precede publication of findings. This is a measurement issue that applies to all social science research.

Epstein et. al.’s analysis exhibits the advantages of a trichotomous variable for regime-type over a dichotomous one. The discoveries of partial democracy trends lead the authors to conclude that this new category is “critical to the understanding of democratic transitions… upon which future research should focus” (566). This finding represents progress in the field, as Huntington’s prescription has given way to more refined, empirically sound systems of measurement that are able to provide more leverage in providing solutions to the most salient comparative riddles.

On the opposite side of Huntington’s dichotomy, a diverse array of authoritarian regimes are combined into a single category. In fact, any government that does not satisfy the minimum criteria of democracy is automatically labeled “authoritarian” in Huntington’s analysis. However, these regimes are very diverse in their systems of governance, and therefore violate KKV’s rule of “unit homogeneity.” Of course, some level of overgeneralization is necessary for large N comparisons such as these. But what could come of a further specification of authoritarian regimes? Preceding Epstein et. al.’s clarification of democratic regime-types, Barbara Geddes explores the potentials of specified measure of authoritarian regimes in her 1999 article, “What Do We Know About Democratization After Twenty Years?”

Noting the lack of scholarly concern for the ongoing authoritarian regimes, Geddes seeks to illuminate this dark corner of the political climate. In fact, she is quite critical of the relevant comparative work on the topic thus far, stating:

“It seems as though there should be a parsimonious and compelling explanation of the transitions, but the explanations proposed thus far have been confusingly complicated, careless about basic methodological details, often more useful as description than explanation, and surprisingly inconsistent with each other. The basic problem faced by analysts is that the process of democratization varies enormously from case to case and region to region. Generalizations proposed have failed either to accommodate all the real-world variation or to explain it” (117)

This critique of the literature underlies an important conceptual dilemma for democratization research. The tendency to overgeneralize is rampant, and comes at the cost of reproducibility. The importance of case study information has been increasingly underappreciated during recent decades, and Geddes has long been a critic of this trend. In her 1999 article, Geddes explains that an analysis of specific types of authoritarian regimes is important for understanding the process of democratization, as well as the many unique components of authoritarian regimes themselves. Most nondemocratic regimes can be accurately described as either personalistic, single-party, or military regimes. These are the subcategories employed by Geddes in her analysis.

Geddes examines 163 authoritarian regimes in 94 countries and finds that the patterns she predicts hold true under the microscope. Each subcategory is shown to exhibit unique tendencies regarding durability, resilience to exogenous shock, and the source of regime-dissolution (where applicable). Personalistic regimes, for example, are the most unstable among the three, while single-party regimes are the most stable. Military regimes are peculiar in that they are far more likely to initiate democratic governments to replace their own authority. These stark differences offer an incredible source of new insight into the research on the democratization process. Furthermore, Geddes’ findings suggest that Huntington’s dichotomous measure of regime-type obscures some imperative components of the transition process.

The brutally simplified dichotomy of regime-types has probably served its purpose in the development of democratization research. It was a starting point for refinement and scholarly expansion by subsequent researchers. To use such a blunt classification system today, however, would severely impair one’s potential for explaining cross-national regime transitions. Epstein et. al. and Geddes have collectively exhibited the advantages of using intermediate categories for research on the democratization process. Evidently, scholars must stay up-to-date on the developments of research in their particular subfield.

Huntington’s contribution to this particular body of research remains influential, despite the academic evolution of some of his proposals. The Third Wave is valuable to researchers today in part because of its cautious approach to claims of causality. Huntington notes that the modeling of the causal chain of democratization is rife with conceptual difficulties. This point is made clear with his list of twenty-seven different variables that have been defended as crucial components of the transition from authoritarianism to democracy (37). Causality is a fundamental issue in the research, but also one of the most difficult ones to assess. In part, this is because of the contemporary academic focus on large n quantitative analyses. Regime change is one of the least generalizable topics in comparative politics, as cross-national comparisons are bound to obfuscate the intricate social context of each regime transition. The process of democratization is a different process for each country that experiences it. Therefore, “the search for a common, universally present independent variable that might play a significant role in explaining political development in such different countries is almost certain to be unsuccessful if it is not tautological. The causes of democratization differ substantially from one place to another and from one time to another” (38).

Comparativists should take note of some simple conceptual issues which Huntington underscores in his book. First, the concept of democracy is not necessarily the catch-all word for everything good about government. A “procedural” definition is far more appropriate for the purposes of comparative politics, thus the acceptance of a minimum criteria approach must precede any comparative analysis of democratization. Democracies may be unstable, ineffective, or otherwise flawed. Second, the causes of the breakdown of authoritarian regimes are not necessarily the same causes of democratization. In fact, authoritarian regimes are more likely to be replaced by new authoritarian regimes than by democracies. This point is expanded by Epstein et al in the passage:

“To this fruitful, ongoing debate we add a reminder that leaving autocracy is not the same as entering democracy. Between these two lie partial democracies, which often act in a manner distinct from those countries either more or less democratic than they, and whose dynamics, while shaping contemporary politics, remain poorly understood” (567)

Third, the problem of overdetermination, or “a multiplicity of plausible theories to explain an event and the consequent problem of establishing the relative validity of those theories,” (36) should not necessarily deter a course of research. Huntington explains, “to occur historically, an event almost has to be overdetermined theoretically. Such is clearly the case with democratization” (37). Excessive efforts to avoid overdetermination are likely to inhibit the development of new theories and approaches. For explanatory purposes (as opposed to inferential ones), overdetermination is not an important concern.

Huntington’s in-depth examination of the “third wave” of democracy leads him to several important conclusions regarding controversial explanations of democratization. Economic factors are often seen as the most influential variables in the course of regime transition. The macroeconomic conditions of the 1970s and early 80s, Huntington finds, “provided the economic impetus and context for democratization” (59). The economic dynamics that served to weaken authoritarian regimes were, for most countries, either the oil crisis that caused prices to soar or the “Marxist-Leninist constraints” that instigated economic downturns (59). By the early 1970s, many countries under authoritarian rule had already developed economic infrastructures dynamic enough to facilitate the initiation of democracy. Economic prosperity typically increased the likelihood of democratization, especially in countries with internationally resilient economic infrastructures. Furthermore, very rapid economic growth appeared in some countries, which weakened the grip of authoritarian regimes. This destabilizing effect on government rule led some authoritarian leaders to liberalize accordingly. Of course, the process of regime transition was not constant across nations, but rather, “the relative significance of these general causes varied from region to region and from one type of authoritarian system to another, as well as from one country to another… also varied over time during the third wave” (107).

Democratization cannot be reduced to merely economic factors. An influential social norm that pervaded the third wave was the issue of legitimacy and regime-type. The international community had clearly adopted the democratic conception of legitimacy, which led to pressure on authoritarian regimes from both foreign and domestic sources. The term “democracy” has consequently been exploited by dictators in their quests for social legitimacy. Authoritarian leaders have rigged elections or systematically intimidated voters, issued single-party ballots, censored political opponents, and otherwise sought an image of democracy without truly committing themselves to such a system. Fortunately, comparativists use their own definition of democracy and do not accept the self-identification of national governments. This tendency to exploit the international consensus on the legitimacy of democratic rule should lead the comparativist to employ a close examination of each national government before categorization.

Probabilism is the most basic practical assumption that comparative models must employ. Deterministic theories, especially in a cross-national context, are sure to fail tests of reliability. There are far too many factors at-play for comparativists to accurately predict societal events. Individuals play a crucial role in the outcomes of political developments. Nonetheless, the likelihood of general outcomes or developments can be assessed with the tools of comparative politics. This point is made clear in the second chapter of The Third Wave:

“General factors create conditions favorable to democratization. They do not make democratization necessary… A democratic regime is installed not by trends but by people. Democracies are created not by causes but by causers… The emergence of social, economic, and external conditions favorable to democracy is never enough to produce democracy… Political leaders cannot through will and skill create democracy where preconditions are absent” (108)

The research on regime transition and democratization has certainly progressed over the decades, but the potentials for comparative insight remain bountiful. Less than a decade ago, Geddes stated that the research project is still in its “infancy,” despite the thousands of publications on the subject. It is important for future researchers to pay close attention to the findings of their predecessors, as the first segment of the trail has already been blazed. Progress within the subfield is the goal of all good research, and thus the collaboration of discoveries must occupy a top priority among scholars. Conceptual and measurement issues have been revealed over the course of academic developments. These lessons and caveats must be recognized in order for progress to ensue.

References:

• Berejikian, Jeffrey. 1992. "Revolutionary Collective Action and the Agent-Structure Problem." American Political Science Review 86(3): 647-657.
• Epstein, David L., Robert Bates, Jack Goldstone, Ida Kristensen, and Sharyn O'Halloran. 2006. "Democratic Transitions." American Journal of Political Science, Volume 50 Issue 3, Pages 551-569.
• Geddes, Barbara. 1999. "What Do We Know About Democratization After Twenty Years?" Annual Review of Political Science 2:115-144.
• Huntingdon, Samuel P. 1993. The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press. (chapters 1&2: 1-108)
• King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton: Princeton University Press. (Chapters 1-3)
• Lichbach, Mark Irving. 1994. "Rethinking Rationality and Rebellion: Theories of Collective Action and Problems of Collective Dissent." Rationality and Society 6(1): 8-39.
• Norris, Pippa, Stefaan Walgrave, and Peter Van Aelst. 2005. "Who Demonstrates? Antistate Rebels, Conventional Participants, or Everyone?" Comparative Politics, 37(2): 189-205.
• Putnam, Robert D. 1996. "Bowling Alone: America's Declining Social Capital." Journal of Democracy 6:65-78.
• Wellhofer, E. Spencer. 2003. "Democracy and Fascism: Class, Civil Society, and Rational Choice in Italy." American Political Science Review 97(1): 91-106.