An Examination of Select Works in Comparative Politics
The study of democratization has occupied a privileged position in social science research over the past few decades. According to Samuel Huntington, the study of democratization is important because “the correlation between the existence of democracy and the existence of individual liberty is extremely high” (Huntington 1984, 194). But what do social scientists really know about democratization? What is the best way to evaluate the extent of that knowledge? In this post, I summarize work by James Mahoney on knowledge accumulation in the social sciences and apply his insights to an examination of select literature on democratization in the field of comparative politics.
James Mahoney, in a 2003 book on comparative historical analysis, devoted a chapter to “Knowledge Accumulation in Comparative Historical Research” in which he outlines four elements of knowledge in social science research (Mahoney 2003). Mahoney uses these elements as a rubric to “assess the extent of knowledge accumulation that has taken place in the field of comparative historical analysis,” but narrows his discussion “to studies of the origins of democratic and authoritarian national regimes” (Id. at 131). My goal in what follows is to offer an outline of Mahoney’s elements of knowledge, summarize his application of these elements to the body of literature he examines, and then use Mahoney’s model as a lens through which to examine select literature on democratization.
Mahoney identifies four elements of knowledge for the purpose of social science research. The first two—descriptive findings and causal findings—involves, says Mahoney, specific knowledge claims. The next two, methodologies and meta-theories, are the tools of knowledge-generation. Descriptive findings are collections of data about social phenomena, but only become accumulated knowledge “when scholars classify existing information using well-specified concepts, typologies, and quantitative indexes” (Id. at 134). Only when new information can be integrated into the existing body of knowledge does knowledge accumulation occur.
Once scholars know something about the world, some may attempt to establish causal relationships between observed phenomena. When a social scientist makes a causal claim, that opens the door for iterated hypothesis testing, whereby “new researchers may attempt to replicate the original finding exactly, or they may retest the original hypothesis using new data and cases” (Id. at 135). Of course, this process may also lead to the introduction of “new independent variables that works in conjunction with previously identified ones to better explain the outcome under consideration” (Id.). This activity, which Mahoney terms hypothesis elaboration, can result in “a new hypothesis that yields additional information about causal patterns” (Id.).
Methodologies and meta-theories are the other two elements of knowledge identified by Mahoney. The former “encompasses the broad range of procedures analysts use to generate descriptive and causal inferences, and the latter “suggest orienting concepts, target certain kinds of variables as important, and point to styles of research and explanation” (Id. at 136). Taken together, Mahoney uses these four elements to evaluate three separate research programs in the field of comparative historical analysis on the causes of democratic and authoritarian regimes.
Maloney identifies and examines the research programs of Barrington Moore, Guillermo O’Donnell, and Juan Linz and Alfred Stepan. He illustrates how each research program lays out an initial causal argument which is then subjected to iterative hypothesis testing and hypothesis elaboration. New methodological tools may be developed to gain additional leverage on pressing questions and new meta-theories may emerge. Afterward, a new round of hypothesis testing begins and knowledge is accumulated as some hypotheses are rejected and others are substantiated. I will offer a brief sketch of Maloney’s treatment of the Moore research program and then proceed to examine a related body of literature on democratization using Maloney’s criteria.
For the sake of simplicity, Maloney reduces Moore’s argument in his seminal work, Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World (1966), to two theses regarding the causes of democracy and authoritarianism:
First, Moore argues that a strong bourgeoisie (understood mainly as town people engaged in commerce and industry) was important to the creation of democracy. In particular, democracy developed in France, England, and the United States because the bourgeoisie ultimately avoided becoming the subordinate partner in an alliance with landed elites against the peasantry….Second, in reactionary-authoritarian cases, labor-repressive landlords forged an alliance with the bourgeoisie against the peasants. Specifically, reactionary authoritarianism developed in Japan and Germany because labor-repressive landlords maintained the upper hand in an antipeasant coalition with a bourgeoisie of intermediate strength (Id. at 138-139).
Maloney offers a detailed sketch of subsequent iterative hypothesis testing of Moore’s theses by other scholars, who largely falsify Moore’s claims. For example, some scholars, using new data from Moore’s own cases, undermined the link between powerful landed elites and authoritarianism, and between strong bourgeoisie and democracy. Other scholars looked at others cases to test Moore’s claims, “reaching nuanced if decidedly mixed conclusions about their validity” (Id. at 143). This process led to “fruitful modifications of Moore’s arguments” and the development of new hypotheses which were subjected to another round of testing (Id. at 146).
Moore’s research program, says Maloney, has spurred knowledge accumulation on several fronts. First, the falsification of his arguments led not only to knowledge in the negative sense, but also inspired elaborations which were able to stand up to scrutiny—such as the establishment of a strong relationship between capitalist development and democracy by Rueschemeyer et al. Second, Moore was able to establish new descriptive class categories which many scholars still find useful. Third, Moore contributed to meta-theory by “reinvigorat[ing] an entire structural tradition” (Id. at 151). Finally, Moore contributed to methodology by exploring the importance of the temporal sequence of key variables to political outcomes.
Samuel Huntington occupies a similar position to Moore in the democratization literature. In his 1984 article, “Will More Countries Become Democratic?” Huntington lays out the preconditions for democracy. He operationalizes democracy as a political system in which “the most powerful collective decision-makers are selected through periodic elections in which candidates freely compete for votes and in which virtually all the adult population is eligible to vote” (Huntington 1984, 195). Huntington’s preconditions can be summed up as: a minimal threshold of wealth and wealth equality, the presence of autonomous intermediate social groups, an external environment favorable to democracy, and internal cultural norms which allow a distinction between political goals and “ultimate ends” (Id. at 208).
Huntington’s list of preconditions is a convenient target for the iterative hypothesis testing and elaboration offered by other scholars in the field. Before considering the refinements offered by other authors, however, it is important to mention some qualifications Huntington made with respect to his “modest conclusions” (Id. at 214). First, Huntington insisted that none of his conditions were sufficient and only a market economy might be absolutely essential to democratization. Second, while some combination of the factors is required, the mix of factors may vary and the degree to which some factors are present may off-set the absence of others. Finally, process (especially sequence) is consequential. When elites buy into the process of democratization and increased contestation precedes a large-scale political participation, the probability of a successful and enduring transition to democracy is enhanced.
Many authors have suggested fruitful modifications of Huntington’s argument. I will briefly discuss three: Mary Gallagher, Michael Herb, and Eva Bellin. Gallagher, in “Reform and Openness: Why China’s Economic Reforms Have Delayed Democracy” (2002), attempts to explain why rapid economic growth in China over the past few decades has not produced the expected democratization. Herb, in “No Representation Without Taxation? Rents, Development, and Democracy” (2005), examines the putative role of natural resource dependency and rentierism in the largely unrealized democratization in Africa. Finally, Bellin, in “The Robustness of Authoritarianism in the Middle East: Exceptionalism in Comparative Perspective” (2004), suggests that the so-called prerequisites for democracy may not be the most useful way to understand democratization and offers an alternative approach.
Economic development, which both Moore and Huntington consider a core prerequisite for democratization, has been advancing rapidly in China for over two decades yet without the accompanying political liberalization one would expect. Mary Gallagher, in her paper, “Reform and Openness: Why China’s Economic Reforms Have Delayed Democracy” (2002), offers an explanation. Economic development does spur political liberalization, she allows, but only when the economic development is “indigenous” (Gallagher 2002, 354). The PRC, through its pursuit of economic growth through foreign direct investment, has been able to experience high rates of growth without the corollaries of growth which tend to produce “demands from society for political change” (Id. at 343).
What is missing from China’s growth, says Gallagher, is a strong domestic business class which serves as an alternative source of power and influence in China. Thus, two of Huntington’s prerequisites are closely related—economic wealth and autonomous intermediate social groups—but through FDI, China has been able to sever the connection. While China’s domestic private industry remains “small scale and dependent on local government support,” the PRC has pursued economic growth based substantially on FDI which “has become the substitute for domestic private industry in China” (Id. at 346).
The success of the PRC’s policy, according to Gallagher, has been its ability to achieve FDI-led growth prior to any such economic reforms as privatization of state industry and development of a domestic private sector. This is because the PRC has been able to forge a relationship with foreign capital investors that would not prove as stable with indigenous business interests. Such a strong and antecedent partnership with foreign investors strengthens the state vis-à-vis any emergent domestic forces. The importance of sequence in socio-political processes for democratization is indicated by both Moore and Huntington. Here, Gallagher provides a confirmatory case with respect to China as well as a fruitful modification of the economic growth condition.
In his article, “No Representation without Taxation? Rents, Development, and Democracy” (2005), Michael Herb addresses another major puzzle in the democratization literature: “why is the Middle East so resistant to democratization” (Herb 2005, 297). Like Gallagher, Herb also has a fruitful modification to make to the existing literature on democratization as he “tests the most prominent political claim, that rents harm a country’s chances of becoming democratic” (Id.). Using a new data set and new technique, Herb’s study is an attempt to resolve the dispute between the majority of scholars who believe the lack of democratization in the Middle East is explained by the so-called natural resource curse and skeptics who “argue that the negative effects are exaggerated” (Id.).
Herb explains the natural resource curse or rentier theory thusly:
First, rentier states need not tax (or need not tax much)…winning popular acquiescence through distribution rather than support through taxation and representation. Second, rentierism increases the capacity of the state to both buy off and to repress opposition….[Third], democracy is stymied when oil revenues prevent changes in class structure that usually lead to democracy (Id. at 298).
So a state which receives a substantial portion of its revenue from the international sale of oil under its control is less likely to have a strong independent business class to check it, and is less likely to develop in ways which would produce such autonomous intermediate social groups. Analogous to the role played by FDI in Gallagher’s China study, the rentier argument suggests that if a state has means of acquiring wealth through external partnerships, it lacks incentive to develop its own domestic human capital in ways that could undermine it later.
Herb tests the effect of rentierism and natural resource dependency using a counterfactual technique designed to isolate the effects of rent wealth from other forms of wealth in a society. Herb also gathers what he considers less noisy data gathered from multiple sources. Having so done, he reports a lack of “support for the thesis that rentierism has a harmful net effect on democracy scores” (Id. at 310). However, “the effect of other variables emerge with much more clarity” (Id.). Herb found that rentierism “has a smaller substantive impact [on democratization] than region, Muslim share of the population, and income,” each of which can be found on Huntington’s list of preconditions (Id.).
Eva Bellin, on the other hand, suggests that the pre-conditions or prerequisites of democracy model may be the wrong way of thinking about democratization. In her article, “The Robustness of Authoritarianism in the Middle East: Exceptionalism in Comparative Perspective” (2004), she points out that a failure to meet the prerequisites for democracy is not exceptional to the Middle East, but a failure to make the democratic leap is. She offers an alternative approach to understanding democratization: “robustness of the coercive apparatus” (Bellin 2004, 144). The robustness of a state’s coercive apparatus is the regime’s will and capacity to repress democratic forces. This is the approach to democratization studies, says Bellin, that makes the most sense of the observable data.
A state’s robustness can be analyzed by reference to four variables. The first is the fiscal health of the state. A critical look at sub-Saharan Africa shows that those countries’ successful democratization was less about strong democratic forces within the society and more about a state which lacked fiscal resources to resist. Second, an international support network makes a state coercive apparatus more robust against democratic forces. This can be seen in the democratization that swept the Eastern European satellite states once those regimes could no longer depend on the Soviet Union or the collapse of Latin American authoritarian regimes after the withdrawal of U.S. support.
The third and fourth variables of a state coercive apparatus are the institutionalization of the military and the level of popular mobilization. To the extent that the military is “rule governed, predictable, and meritocratic” (versus corrupt and patrimonial) it will be more amenable to reform and less likely to repress a democratic movement or use such a movement opportunistically (Id. at 145). Bellin points to the successful democratic reforms under South Korea’s General Roe Tae Woo as an example. The fourth and final variable, the level of popular mobilization, is all about the regime’s cost-benefit analysis with respect to repression. As in the 1987 case of Roe Tae Woo, “the elite is forced to ask if the cost of repression is worth the benefit” (Id. at 147).
Bellin tests her model on the case of the Middle East and North Africa and finds greater explanatory power:
The Exceptionalism of the Middle East and North Africa lies not so much in absent prerequisites as in present conditions that foster robust authoritarianism and especially a robust and politically tenacious coercive apparatus….In this context, regime elites possess both the will and the capacity to suppress democratic initiative. And where international support and financing is forthcoming to the authoritarian regime, rapid regime change is unlikely (Id. at 152).
Much of Bellin’s analysis seems to be supported by the recent Arab Spring movement in the Middle East and North Africa, but as this movement is quite recent and on-going it would be premature to draw too many conclusions. Bellin, with a hedge that brings the discussion full-circle, notes that “while the removal of democracy-suppressing coercive apparatuses is a necessary condition for democratic transition and consolidation, it is not sufficient.”
From Huntington’s initial hypothesis of preconditions for democracy to each of the fruitful modifications offered by Gallagher, Herb, and Bellin, we seemed to have come full circle indeed. Ultimately, what social scientists know about democratization is well-summed up by Bellin: “A host of conditions, including a minimal level of elite commitment, a minimal level of national solidarity, a minimal level of per capita GNP, and…impartial and effective state institutions must be present” (Id.).
A.K.
