The World of Political Science— The development of the discipline
Book series edited by Michael Stein and John Trent
Professors Michael B. Stein and John E. Trent are the co-editors of the book series "The World of Political Science". The former is professor of Political Science at McMaster University in Hamilton, Ontario, Canada. The latter is a Fellow in the Center of Governance of the University of Ottawa, in Ottawa, Ontario, Canada, and a former professor in its Department of Political science. Dr.Tim Heinmiller is the coordinator of the series, and Assistant Professor of Political Science at Brock University in St. Catherines, Ontario, Canada.
Dirk Berg-Schlosser (ed.)
Democratization
The State of the Art Second revised edition
Barbara Budrich Publishers Opladen & Farmington Hills 2007
line grande revolution denuicratique s'opere parmi nous."
Alexix de Toeqiieville
1. Introduction Dirk Berg-Schlosser
Tocqueville's prescient statement today applies more than ever. According to the latest counts, roughly 120 out of the world's more than 190 states presently claim to be a "democracy" of some kind (see, for example, Freedom House 1999). About half of them emerged after the crest of the latest wave of democratization (1989-90) (see also Huntington 1991), At a closer look, however, these new democracies show a bewildering variety of specific subtypes and concrete defects when compared with their more established counterparts, especially those in Western countries (see also O'Donnell 1994, Linz and Stepan 1996, Merkel and Puhle 1999, Lijphart 1999, Diamond 1999, Schmidt 2000). With this enormous enlargement of political science's field of study, the problems of appropriately conceptualizing, precisely measuring and adequately theorizing these developments are all the more urgent. This is not least because this also implies concerns of practical politics vis-avis these new democracies (for example, with regard to the "political condi-tionality" of the World Bank, the International Monetary Fund and similar institutions) (World Bank 1997).
Among tlie questions raised are: How are the previous regimes in these countries to be classified, and what legacies have they left? At what point can the new systems be called "democratic"? How democratic are they? Are there significant differences in their degree of democratization? Are there distinct democratic sub-types? What are their characteristic features and possible defects? How durable and stable (consolidated) have they become? How can their overall quality be further improved (in what areas, and according to what criteria)? (This question also applies to the "established" democracies.) Since we are all involved, as national and cosmopolitan citizens, what can and should we do about these developments? What perspectives and concrete policy advice can political scientists provide? What actions should we take? These and similar questions are of particular urgence and relevance, even though political science's concepts and methodological tools to cope with them remain limited.
Political science is a rapidly expanding and ever-changing field. An assessment of its state of the art is, therefore, limited. The authors can present
15
only a snapshot of the ever more universal processes that have continued since Tocquevilie's time. As he put it: "In question que j'ni soulevee n 'interesse pas seulemenl les Etats-ifnis, metis ie moncte entier; mm pas line nation, mais tons les homines." (Tocqueville 1963: 175)
1.1. Conditions of Democracy
Any assessment of contemporary processes of democratization has to begin with an overview of the background conditions of modern democracies and some of their historical, regional and cultural specificities. This includes a look at the processes of state formation and nation-building, which often have pre-democratic or external origins. Sovereign states are the most important geopolitical units today, and they are the most influential actors in international politics. Their identities and perceived legitimacy have dimensions both objective (in terms of concrete boundaries and specific institutions) and subjective. Discrepancies between these two dimensions may appear during processes of democratization because the participatory aspirations of citizens and their respective identities will not necessarily coincide with the existing political boundaries and the institutional framework. This may lead to a more or less peaceful redrawing of boundaries and attempts of internal democratic reforms. But it can also result in attempts of secession, wars with neighboring countries, or internal civil strife, together with more abrupt and sometimes revolutionary and violent regime changes. These processes and possible conflicts cannot be resolved by democratic standards and procedures because the rule of law pre-supposes an existing political unit, and procedures such as majority decisions may exclude and possibly suppress important segments of the population. If democracy, in a broad and simple sense, means "rule of the people," it first has to be decided who die people are and which boundaries should be respected. In this sense, state formation and nation-building must be considered as prerequisites of any meaningful democratization. As such they are, however, only rarely addressed by many works of democratic theory. Rather, they constitute, in Dahl's (1989) terms, a "shadow theory" of democracy.
In modern times, large nation-states were first formed in Europe, in particular after the "Westphalian Peace" treaty in 1648 and the agreements of the Vienna Congress in 1815. Both coercive military interventions and commercial-capitalist interests were most instrumental in that process (see also Tilly 1990). In the 20th century, the re-drawing of boundaries after the two world wars and the dissolution of die Soviet Union and former Yugoslavia changed the political landscape and gave it its present shape.
In other parts of the world, the colonialism and imperialism of the major European powers determined most present-day boundaries. This applies to
16
Latin America and the Caribbean, most parts of Africa and large parts of Asia. Exceptions include Ethiopia. Iran. Afghanistan, Thailand and, most significantly, Japan and China. In the Middle East, the dissolution of the Ottoman Empire after World War I and League of Nations mandates for Great Britain and France shaped most of die present political landscape (for a more detailed account, see Berg-Schlosser, 1999b).
Most modern states have relatively firm and undisputed boundaries (which in some regions are beginning to be transcended by "supra-national" arrangements and institutions). Nevertheless, some critical places have not resolved their territorial problems. These include Israel/Palestine, Cyprus, Lebanon, the Kurdish areas in the Middle East, war-torn states such as Afghanistan and Cambodia, and, in particular, the "collapsed states" in Africa: Somalia, Liberia, Sierra Leone, etc. (see, for example, Zartraan 1995). In others, severe internal conflicts between contending ethnic and other identities still predominate (see Horowitz 2000; a more general recent assessment of democracy's outer and inner edges can also be found in Shapiro and Hacker-Cordón 1999 a).
Over and above these basic historical pre-conditions of present-day democracies, a great number of other factors contributed to their emergence over time. If we take only formally established democratic regimes, as covered in the "Polity" time series data set (Jaggers and Gurr 1996; for a discussion of some of its limits, see also Chapter 2, below), the worldwide growth of the number of countries with democratic governments can be depicted as in Figure 1.
Democracies have emerged in increasing numbers since the beginning of the 19th century, mostly in Anglo-Saxon and Western European countries. This culminated in a more rapid expansion shortly after Word War I, still mostly in Europe but now including some parts of the former Tsarist, Habsburg and Ottoman Empires. This trend was then considerably reversed, leading to fascist or other types of audioritarian regimes, until the end of World War II (see also Linz and Stepan 1978, Berg-Schlosser and Mitchell 2000 and 2002). The breakdown of the colonial empires after World War II then led to the emergence of many more independent states, first in Asia and then in the Middle East and Africa. These included a number of new democracies. In addition to the latter, some civil-authoritarian or military regimes in Southern Europe and Latin America have also democratized or redemocra-tized since the middle of the 1970s. The most recent upsurge occurred after the democratic changes in Eastern Europe in 1989-90, the dissolution of the Soviet Union and its worldwide repercussions.
MASARYKOVA UNÍVE1U! i Fakulta inciálnich studii Joštovu 10 602 (M) B íl N O (^SD-V
17
Figure I: Emergence of Democracies
18(10
S820 S&40
2000
Year
Source: based on Jaggers and Gurr (1996), countries with 8 and more points on the "Polity 111" democracy scale.
To speak of these developments as three distinct waves with their respective "reverse waves," as does, for example, Huntington (1991), is an oversimplification, The causes and inter-relationships of this partem are far from being undisputed (see also Markoff 1996, Green 1999a and 1999b, Doorenspleet 2000). In empirical democratic theory a great number of contending approaches and perspectives to explain these developments can be distinguished. To these we now turn.
1.2. Contending approaches and perspectives
The analysis of conditions conducive to the emergence of democratic political systems has always been one of the central concerns of political science. From Aristotle through Locke, Rousseau and Tocqueville, up to the multitude of contemporary studies, this analysis has been attempted again and again. Under closer scrutiny, however, the results obtained are still controversial.
This is not surprising if one considers the complexity of the notion of "democracy." To begin with, there are, at the micro-level, aspects relating to "democratic personalities" (compare, for example, Sniderman 1975 and Berg-Schlosser 1982) and to the conditions of "rational choice" (including the "economic theory of democracy" by Downs 1957). Then there are questions relating to a more general participatory, tolerant, "civic" political culture with accepted "rules of the game" and "rational-legal" sources of legitimacy (Al-
mond and Verba 1963. 1980, Weber 1922). The more general bases include social-structural characteristics and their specific historical dynamics (see, for example, Moore 1966). These, of course, are related to-the respective modes of production, the economic mechanisms of distribution, aspects of class formation, social mobility, etc. (see also Schumpeter 1943).
At the level of intermediate structures, the "plurality" of interest groups and voluntary associations (cf., e.g. Dahl 1971), the formation of party systems (Lipset and Rokkan 1967) and "new social movements" (cf., e.g.. Brand 1985) are of special concern. Within the central political system, constitutional questions relating to the formal division of power, "presidential" or "prime-ministerial" executives, electoral laws, the independence of the judiciary, centralized or federative administrative structures, etc. become relevant (cf., e.g., Loewenstein 1957). In addition, "consociational" or majoritarian patterns of decision-making (Lijphart 1977) and the more general questions raised by organization theory (Etzioni 1968, Naschold 1969) require discussion. The tolerance of institutionalized forms of opposition, a political style that allows for compromise, the accountability of government decisions, an "open" administration, and the avoidance of nepotism and corruption constitute further attributes of such systems, at least in an "ideal" sense (Dahl 1956).
Between these various levels, there exist problems of "congruence" (Eckstein 1966), multiple feedback mechanisms, and dynamic adaptations over time. In many instances, external factors, such as questions of military security, economic dependence, population migration and the global context of the "world system" (cf. Thompson 1983), also require consideration. All this must be evaluated from a normative perspective and measured against the central concerns of political philosophy: the question of a "good" political order, a "humane" existence in both a material and nonmaterial sense, basic rights and freedoms, the realization of the emancipatory potential of human beings, the protection of minority rights, etc. (cf., for example, Pateman 1970, Scharpf 1972).
Among this multitude, some major emphases in the more recent literature are evident. The broadest is closely linked to what has become known as "modernization" theory. Based on studies by Lerner (1958), Lipset (1960), and Almond and Coleman 1960 (among others), this approach takes general trends of socio-economic development, urbanization, literacy, etc. and considers them as basic conditions for modern "political development," including democratization. This approach employs a number of indicators, such as the levels of GNP per capita and of literacy, as independent variables on which the resulting level of democratization (also measured with certain indices; see also Chapter 3) is seen to depend.
In a more extreme version, a high level of socio-economic development is seen as a requisite of democracy (this was expressed in the title of Lipset's
19
original article [1959]; his later, more '''probabilistic/^ view can be found in Lipset 1994).
There always have been a number of counter-factual examples guarding against an ali-too-simple interpretation of this thesis. These include the breakdown of democratic regimes io highly modern countries, as happened in Weimar Germany, and the continued existence of workable democracies in poor countries such as in India and some other Third World states (see also Berg-Schlosser 1989). In particular, some of the policy recommendations based on such perspectives advocating "development dictatorships" in the early stages of modernization (for example, Lovventhal 1963) have turned out to be false. The most comprehensive recent study of this kind clearly shows this: "Democracies can survive even in the poorest nations if they manage to generate development, if they reduce inequality, if the international climate is propitious, and if they have parliamentary institutions" (Frzeworski et al. 1996:49, our emphasis added). Rather than being a prerequisite, economic development can be a condition favoring the emergence of democracy and an associated factor that increases its sustainability.
Against these broad "macro-quantitative" statistical analyses based on the respective means and correlations of their major indicators, more specific "structuralist" approaches have been developed. These consider the specific emerging class structures and their dynamic interactions, rather than the overall level of economic development, to be decisive.
In a neo-Marxist sense, Moore's (1966) study distinguished three paths to modernity, one based on a successful bourgeois revolution and strong middle classes (as in the United States, the United Kingdom or France) leading to the contemporary democracies, another based on an alliance of the old landed oligarchy and the more recent capitalist class ending in fascism (as in Germany or Japan), and a third one emanating from a successful peasants' and workers' revolution establishing communist regimes (as in the Soviet Union and China).
In a more refined and extended version, which includes smaller European states and Latin American countries, Rueschemeyer, Huber Stephens and Stephens (1992) and Collier (1999) followed up this line of argument and pointed out the sometimes ambivalent role of the middle classes and the significance of workers' organizations, such as unions and socialist parties, in the process of democratization.
In many cases, the vertical ("class") dimension of social structures has to be supplemented by a horizontal one juxtaposing ethnic, religious and similar social cleavages that often have particular regional strongholds. These may interact with the vertical dimension, forming crosscutting or reinforcing patterns. They can also be ordered in hierarchical ("ranked") or parallel ways (see, for example, Horowitz 2000). In addition, ethnic or religious groups are usually also internally stratified, which complicates their potential for conflict even further (Waldmann 1989).
The most comprehensive integration of dominant vertical and horizontal cleavages and their consequences lor state formation, nation-building and democratization for a concrete region and period has been attempted by Stein Rokkan in his "Typological-Topological Model of (Western and Central) Europe" (see Flora 1999). There, he identifies the major social cleavages in Europe since about the I61'1 century concerning the relationship between church and state (in particular after the Protestant Reformation in Northern Europe), relations between the respective political center and the regional periphery/ peripheries in each country, conflicts of interest between the rural (often formerly feudal) and the urban (including the emergent bourgeoisie) classes, and, finally, modern conflicts between capital and labor in increasingly industrialized states. On this pattern, in his view, can be based many important political developments. These include trajectories of the respective countries towards audioritarianism, fascism or democracy in the 20* century and the major characteristics of their party systems up to the present time (see Lipset and Rokkan 1967 and the more recent assessments in Karvonen and Kuhnle 2000). Attempts to develop and apply similar models to other parts of the world have, however, remained very limited (see, for example, Shiratori 1997, Temelli 1999, Randall 2001).
These "objective" social-structural dimensions of the social bases of democratic development were also contrasted by more "subjectively" oriented political-cultural studies. Pioneering among these was Almond and Verba"s "Civic Culture" (1963). They showed that the democracies in the United States and Great Britain were also deeply rooted in the attitudes and values of the population at large, in contrast to the situation in (post-war) Germany, Italy and Mexico. More recent studies indicate that, in the meantime, democracy has become more anchored in the minds of (west) Germans and (northern) Italians as well, but with strong remaining regional sub-milieus (see, for example, Baker, Dalton and Hildebrandt 1981, Berg-Schlosser arid Rytlewski 1993, and Putnam et al. 1993).
Similar studies in the behavioralist tradition have also been extended to other parts of the world, including the "Latino-", the "Afro-", and the "New Democracies" barometers (see, for example, Laiinobarometro 2001, Afro-Barometer 2005, Plasser, Ulram and Waldrauch 1998, Rose, Mishler and Haerpfer 1998) and the three waves of the World Values Survey (see, for example, Inglehart 1997). The scope of political attitudes and (potential) political actions has also been considerably widened, including the "conventional" and "unconventional," legal and illegal, and peaceful and violent forms in the panels of the Political Action study (see Barnes, Kaase et al. 1979, Jennings and van Deth 1990). This "subjective dimension" of politics is an important factor for the long-term consolidation of democracy. It interacts with the objective social-structural and institutional aspects, but only rarely can it be considered as an independent variable in early processes of democratization (see also Elkins and Simeon 1979).
20
21
The impact of these objective and subjective social bases of politics depends on their interactions and tonus of aggregation at the intermediate ("meso-") level. There, certain cleavages and their cultural expressions often harden into particular "sub-milieus," which can reproduce themselves over long periods (see also Lepsius 1966). The party system may also reflect such structural or cultural strongholds. If the party system is mainly based" on strong horizontal affinities and identities among contending (such as ethnic or religious) groups, then no "floating vote" from one election to another can be expected. Elections then become just another form of a population census. This may lead, depending on the respective number of such groups and their relative sizes, to permanent majorities of one or a few groups. This seriously endangers, if no "consociational" agreement can be found, the long-run stability of any democratic system (see also Lijphart 1977, 1999). In addition, various forms of (economic and other) interest organizations and (often more temporary) social movements shape this sphere. Taken together, they constitute the most important collective actors (see also Olson 1965).
In recent times, the importance of this intermediate realm between the "micro" (individual) and "macro" (state) level for ongoing processes of democratization has also been emphasized by proponents of civil society (see, for example, ICeane 1988 and Hall 1995). These included all kinds of nongovernmental organizations that act in the public sphere. The major modes of transmission of interests can be pluralist, emanating from the more-or-less open competition of a multitude of social groups; corporatist or neo-corpora-tist, involving the major economic interest groups of employers and unions in conjunction with the state authorities (see e.g. Schmitter and Lehmbruch 1979); or clientelist, based on personal vertical relationships of "unequal exchanges" (for example material benefits in exchange for political support) (see, for example, Eisenstadt and Roniger 1984, Kitschelt and Wilkinson 2007).
The major link between social structural cleavages, particular sub-milieus, the party system and the representative institutions at the macro-level of democratic systems is provided by the electoral system. To some extent, electoral systems, such as majoritarian or proportional ones, exert a certain influence of their own. This may work in different directions. For example majority systems in single-member constituencies may lead to two-party systems in horizontally relatively homogenous but vertically stratified societies and, consequently, to clear-cut majorities in representative institutions (Duverger 1962). Conversely, simple majority systems in societies that are highly fragmented horizontally tend to reproduce this pattern in the party system and parliament. By contrast, strong proportional systems can lead to highly fragmented party systems and parliaments in economically stratified societies, but tend to create more crosscutting parties in countries with strong ethnic or regional divisions (see, for example, Grofman and Lijphart 1986,
Nohlen 2000). The choice of an electoral system, therefore, forms an important part of the constitutional engineering and crafting of new democracies (see also Di Palma 1990 and Sartori 1994). It often -involves, however, a trade-off between the exigencies of a more "just" political representation and democratic stability and efficiency.
The other major institutional choices concern a centralized or federal setup of the state structure. This also depends on the size and'on the degree of regional fragmentation of a country and on the separation of powers at the central level, such as in parliamentary or presidential systems or some of their variations (for discussions of their advantages and disadvantages, see Shugart and Carey 1992, Linz and Valenzuela 1994). Presidential systems may, for example, be "culturally" more suited in countries with strong personalistic traditions, as in Latin America, but they also tend to reinforce clientelistic relationships, with their often concomitant high level of nepotism and corruption. This makes constitutionally limited terms of office in such systems all the more necessary. In any case, the full independence of the judiciary must be safeguarded in all democratic systems to ensure the rule of law and, together with independent media and a well-informed public, the accountability of all public leaders and organizations, which also enhances their efficiency and effectiveness.
Within a given institutional set-up. the matters and decisions of major actors also play an important role. These can be analyzed (with the advantage of hindsight) by historians looking at specific events, but also, in a more general sense, through psychological or socio-cultural approaches (see, for example, Elms 1976, Furnham and Heaven 1999), In recent years, "rational choice" and "game-theoretical" models and arguments have been employed in this context. This was, for example, the case with the strategies and decisions that crucial actors made in the various modes of transition from authoritarian to democratic forms of government. Rational pacts of this kind thus have been concluded by softliners and moderates in the authoritarian and democratic camps respectively in a number of cases (see, for example, Co-lomer 2000). Games such as "Battle of the Sexes" and "Staghunt" can also be modeled for such transitions (see, for example, Geddes 1999). In a more general sense, an "actor-centered institutionalism" (see Scharpf 1997) can help explain such developments.
Complex theories of democracy must look at more than the general historical and social conditions—the "input" side of the political system and the central institutions and actors. They must also take account of the respective "outputs" and the more general performance over time. In this respect, a number of studies have compared the results of democratic regimes with different types of authoritarian ones (see, for example, Berg-Schlosser and Kersting 1996, Schmidt 1998, Przeworski 2000). This applies not only to the common economic indicators (GDP per capita growth, etc.), but also the
more differentiated social and quality-of-life criteria (see, for example. UNDP 2002) and normative aspects as reported by organizations such as Amnesty International, Human Rights Watch or Freedom House. Such criteria are also increasingly taken into account by the major international development agencies, which have become concerned about good governance (see, For example, World Bank 1992). The extent of public waste, corruption and private enrichment from public sources is now also regularly ; monitored by organizations such as Transparency International. In this re- : gard, a critical public, independent media, and a well-functioning judiciary, which are characteristic of the more democratic states, contribute to the performance of democracies in the longer run. Amartya Sen, winner of the 1998 -Nobel Prize for economics, stated: "A country does not have to be deemed fit for democracy, rather it has to become fit through democracy" (Sen 1999, emphasis in the original). :
All this takes place, of course, in an international environment that may or may not be favorable to such developments. During the Cold War, for example, superpowers or camps often gave external (including military and financial) support to their "friends" without talcing into account the internal conditions of those regimes. Then, after 1989-90, a number of authoritarian regimes collapsed when this external support (or threat of intervention) was withdrawn. This occurred in a number of East European and African coun-tries, for example. Events in one country may also have significant | "demonstration" (chain reaction and domino) effects as media and other g contacts transmit news to neighboring states in a similar situation as well as J to countries further afield (Whitehead 1996). The international political cli- g mate may also be less (as in the inter-war period in Europe 1919-1939) or | more (as in present times) favorable for democratic regimes. Furthermore, external support can help to stabilize and consolidate new democracies, as the | European Union is currently doing in Eastern Europe. ;|
Over longer periods of time, all these factors interact and may form par- | ticular patterns, sequences, or "path-dependent" effects, such as Huntington's j "waves." But apart from such metaphors, the actual causal links and their | dynamics and the subsequent feedback mechanisms must be analyzed more J closely. As the following chapters show, a lot remains to be done. This re- J quires both better information and data, but also, and even more importantly, J clearer conceptional and theoretical tools. Some recent attempts to integrate J different levels of analysis and to put them together in a more comprehensive | "systems" framework can serve as a starting point. |
1.3. Possibilities of theoretical synthesis; remaining issues
James S. Coleman (1990) made the most elaborate attempt to coherently link macro-, meso- and micro-levels of analysis. He linked a given objective ("structural") situation at the macro-level with the subjective perceptions and motivations of individuals at the micro-level, then took into account their possibilities of aggregating their interests and activities at the meso-level in order to "explain" as much as possible die final outcome ("the explanan-dum") at the macro-level (see Figure 2).
Figure 2: Linking Levels of Analysis
macro-level explanandum
actor _ behavior
logic of selection
Source: Adapted from Coleman (1990) and Esser(1993)
For our purposes, this model can integrate social structural conditions, in Moore's or Rokkan's sense, in the upper left-hand side, with the resulting political culture at the micro-level and the differing forms of group aggregation at the meso-level (interest groups, party systems, etc.) in order to more clearly establish the broader social bases of democratic regimes, again at the macro-level (on the upper right-hand side). At the macro-level, then, the more specific political institutions are established, in which the major individual actors (presidents, prime ministers, leaders of the opposition, etc.) are involved. The concrete situation at each level can be filled by empirical data for the respective cases under investigation. From the systematic comparative observation of several similar cases, certain generalizations may result in an "analytically inductive" manner (for the use of this term see, for example, Blalock 1984). These generalizations may, in turn, serve as hypotheses to be
24
25
r
tested across a more diverse selection of cases in order to establish their theoretical "range" in space and time (for the use of such macro-qualitative comparisons, see also Ragin, Berg-Schlosser and De Meur !996, Berg-Schlosser and Mitchell 2002). Most of the concepts and hypotheses of the "transitology" and "consolidology" literature mentioned above are of this kind.
At some points, but not in the sense of an overarching unified theory (Wallerstein 2001), even more general hypotheses concerning the reasoning and actions of individuals and groups can be introduced. An example is the concept of "restricted, resourceful, evaluating, expecting, maximizing men" (RREEMM) or women at the micro-level (see Esser 1993: 231 ffj. But even such general assumptions in the "rational choice" and "collective action" traditions are embedded in a more specific structure ("restricted") and cultural ("evaluating") context. Similarly, at the level of individual actions within certain institutions and their respective strategies, such assumptions and various specific models concerning their concrete "games" may be applied.
In an even more comprehensive manner, such models can be embedded in a general "systems" framework (see, for example, Easton 1965, Almond and Powell 1978). In this way, the major social sub-systems and their interactions can be meaningfully integrated (see Figure 3).
Here, the cultural, social and economic sub-systems, originally derived from Parsons' (1951) AGIL scheme (see also Munch 1984), interact and set the scene for the political system, which derives its main inputs from these spheres. At the meso-level, again, these inputs are aggregated and mediated in specific ways and then "ruled" upon by the central political system in a narrower sense of the term as the overall decision-making unit. Its outputs are fed back to the respective sub-systems by the usual administrative structures and complete the cycle, which, of course, will be repeated over and over again in a dynamic sense. At the same time, all this is also embedded in the encompassing international system, with which all sub-systems interact at different levels. Within this framework current problems of "governance" can also be considered (March and Olsen 1995).
This is not the place to elaborate further on this framework (see also Berg-Schlosser and Giegel 1999), but it must be emphasized that this framework does not assume a priori any far-reaching implications in a more ambitious "systems theoretical" sense such as Easton's (1965) or Luhmann's (1984, 2000) concerning, for example, certain equilibria or the long-term stability of such systems. But it is helpful to locate the emphases of different approaches and to assess their potential overall contribution to a more general but still empirically founded theory of democracies concerning their stability, their potential centrifugal or centripetal tendencies, and their effectiveness and overall quality in both a functional and a normative sense.
-a o
s
3
o
/
0)
"—ID —
tem	5 c _D	
a>	"in	
>t	"u <U	JE tZ i/j
<n		
ical	n ^_ o	O) c E .2 W ut
	(/]	'— *U itj OJ
0	c	0- Q
a.	utio	
	"[ft	it cn > c
	c	^ -9
		Exect Deels
.5 5 U
m p a. m
.3 TT, xj - ~
Internalionaf Syslem
g E B'
a.
jl 1
□
o
•4—►
i II 1
u
E o c o
■2 E
to
□ M
5 °
"in to
•I !=§
g SB
r- —   □  c- o   o ™
^    üj ^ q. o to a. S
a. ^ rn   ^ c
Q_ aj    (J _i:
£s 5 if
I sfS to
'S
o w
3*5
ü S 3 a n o x x -a rtT m 0) o aj c
_ in C
s g-s
|EJS
I as
£ J e e
ID C
1 8
— c
a. to
— Ufa)
J3 u
w - D
to
3  5 ojS E ä j S " -=
u
gl
UIS)S/(S |BUQ!)BUJa)U|
w
E?
o
CO
In any case, such a framework must be "filled" by operationalized concepts )
and empirical data that are historically and culturally grounded. These pose if.
many more problems of appropriate conceptualization, temporal and regional |
range, adequate measurement, resulting sub-types and categories, and better |
overall grasp and fit of our theories. |
In Chapter 2, Dirk Berg-Schlosser first turns to problems of an appropri- J
ate conceptualization and empirical measurement of democracy. He discusses 1
the basic dimensions, operationalization, and the validity, reliability and I
availability of a number of indices that have been developed in this respect. 1
From such a perspective, criteria concerning certain sub-types of democracy j
and their characteristics as well as defects (actual or potential) can also be I
established. This also has concrete practical implications and may lead to I
some institutional proposals or policy advice for the major internal and exter- I
nal actors involved. |
Chapter 3, by Gerardo L. Munck, examines recent studies in the field of 1
democratization and organizes them according to three major agendas: demo- |
cratic transitions, democratic stability, and democratic quality. Munck shows I
the increasing richness of this literature and points out some of the strengths I
and weaknesses of various approaches. His balanced assessment con- I
structively maps out some fruitful avenues for future research. f
In Chapter 4, Jan Teorell and Axel Hadenius take a critical look at much !
of the "large N" literature of a statistical nature and develop their own model |
which helps to explain significant developments over a longer period of time. I
In Chapter 5, Axel Hadenius and Dirk Berg-Schlosser assess, on a re- §
gional basis, the actual spread of democratic regimes after the last "wave," ^
and they provide an overview of the strengths and weaknesses of these re- j
gimes with regard to their longer-term consolidation. J
In Chapter 6, Laurence Whitehead critically examines the state of schol- j
arly debate and actual experiences in the field of democratization since the I
beginning of the new century. In particular, he emphasizes the uneasy (and 1 j
often unresolved) relationship between processes of state formation and de- |
mocratization. These may be more blurred and intertwined than much of the !
current literature suggests. This he illustrates with the paired concrete exam- ; j
pies of Indonesia and Nigeria, Colombia and Sri Lanka, and Brazil and India. j
Furthermore, even stronger weight must be given to international factors in j
present processes of democratization (or failed attempts imposed from the !
outside). The events after "9/11" and the subsequent reactions constitute an- J
other watershed for international relations and the prospects of democracy. i Again, as Whitehead emphasizes, these have to be perceived in a differentiated manner, which balances demands of longer-term analytical rigour and
concrete political relevance. Finally, he discusses the role of international j
norms and (potentially) universal values, such as basic human and political : j rights, in the present world: His conclusions point to the continuing chal-
lenges of scholarly endeavours and real world politics in an ever-changing and ever-closer international environment.
■ The final chapter, by Juan J. Linz, reflects on democracy's recent "victory" and considers its. future. Linz discusses some of democracy's remaining shortcomings and risks, as well as the continuing theoretical debates. This, of course, is not only an empirical or comparative question, but also a profoundly normative and, in this sense, political-philosophical one (for such concerns and problems, see also Shapiro 2001). In an addendum, he reflects about the current international political situation and some of the consequences of U.S. foreign policy after "9/11".
Attempts to impose "democracy" from the outside by force in cases where the internal conditions are very unfavorable for such systems, as the American-led invasion of Irak in 2003 again has shown, are mostly doomed to failure. A simple check of the conditions favorable to democracy as summarized by Dahl (1989:264) would have demonstrated that Irak fulfilled none of them:
- means of coercion (and their legitimate use) controlled by the state;
- a modern, dynamic and pluralist society,
- a political culture supportive of democracy;
- no strong and distinctive subcultures;
- and no interventions by foreign powers hostile to democracy.
The often-heard argument that Germany and Japan after World War II were successful examples of establishing a democracy from outside is quite misleading. First of all, the preceding regimes in these countries had been thoroughly de-legitimized by the defeat in the (self-inflicted!) war. Secondly, and even more importantly, these were countries with a secured statehood, a relatively homogeneous population, a high level of "modernization", and, in the case of Germany, significant pre-war democratic traditions and social forces, even if they had succumbed to the National Socialists in the Weimar period (see, e.g., Berg-Schlosser/Mitchell 2002).
Only in relatively rare cases of post-conflict societies, as in present-day Bosnia, may initially very unfavorable conditions be overcome by external intervention and long lasting outside support (see Schneckener 2002, Gromes forth.). In such instances, it is all the more important that more generally agreed upon principles of human rights and self-determination, as laid down in the United Nations and similar charters, are adhered to. Again, in this respect, the situation in Irak, as demonstrated by the creation of "law-free" zones for the detention of suspected terrorists and blatant human rights abuses, is different.
Altogether, the international "climate" for democracy has again deteriorated somewhat after "9/11" and the subsequent events. Existing democracies may be threatened by terrorist acts from the outside, but they must also be
28
29
2. Concepts, Measurements and Sub-Types in Democratization Research
Dirk Berg-Schlosser
ingful notion of democracy.
In this chapter, I will first address some of the conceptual issues of democratization and empirical democratic theory. I will then turn to some of the concrete measurements and operationalizations that have been proposed and are currently applied in this regard, pointing out some of their advantages and disadvantages. The third section will look at some of the more specific subtypes which have emerged, followed by the broader methodological and theoretical and, to some extent, even "paradigmatic" and metatheoretical implications of this kind of research. I will then draw some preliminary conclusions.
2,1. Conceptual background
The starting point for all these considerations must be a sufficiently complex, consensual, and workable notion of democracy that can capture the differing forms of contemporary appearances of this land of rule. At the same time, it should be sufficiently distinct to draw meaningful boundaries between democracy and other types of political systems, and sufficiently open to be linked to existing sub-types and to future developments. We thus need a "root concept" in Collier and Levitsky's (1997) sense, which satisfies these demands and allows the further differentiation and characterization of present and possible future sub-types by adding the respective attributes ("democracy with adjectives") up or down the "ladder of generality."
Such a root concept is Robert DahFs notion of polyarchy (Dahl 1956. 1971, 1989), which has become the most frequently cited referent of empirically oriented democratization studies in the last decades (see also, for example, O'Donnell, Schmitter and Whitehead 1986, Hadenius 1992, Seren-sen 1993, Diamond 1999, Berg-Schlosser 1999a). He explicitly distinguishes two dimensions of this more modest characterization of contemporary democracies, the amount of regular and open competition in a political system, and the extent of different forms of participation in the process of political decision-making by the population of a given society. Implicit in his notion is
30
31
a third (normative) dimension that concerns basic civil liberties, such as freedom of information and organization, and a political order that guarantees and maintains the rule of law to make regular political contestation and participation possible and meaningful. Even though there are some variations between different authors' formulations and interpretations, these three dimensions of the root concept of democracy that emphasize the "input" side of political systems and the necessary institutional and legal framework have become largely accepted.
To be distinguished from such a definition are the respective bases (historical, economic, cultural, etc.) and conditions of democracy, which often have been seen as requisites (see, for example, Lipset 1960, 1994), and the actual performance and effectiveness of democratic systems, which comprise the "output" side and various distinct policy areas,
Over and above such broad conceptual classifications, we must employ some further distinctions and criteria when discussing concrete problems of die "measurement" of democracy. These are, first of all, related to the purpose of a particular study, and the kind of method and research instrument used. Such a purpose may be, for example, the development of a comprehensive typology of political systems and the existence of more or less "democratic" forms among them. A second purpose, which must be distinguished from the first one, may be the development and measurement of concrete sub-types of democratic political systems: for example, presidential or parliamentary, majoritarian or consensual (see, for example, Ujphart 1977, 1984, Powell 1982, Linz 1994, Sartori 1994). A third purpose can be the further improvement and qualification of democratic systems in a functional or normative sense, which also identifies deficiencies or articulates desired further options. At this point, the wider field of normatively and philosophically motivated as well as empirically oriented theories of democracy is reached (see also Sartori 1987, Habermas 1992, Shapiro and Hacker-Cordon 1999b, Berg-Schlosser and Giegel 1999).
Furthermore, these purposes are related to different levels and "scales" of political maps, which have to be measured, and varying historical periods and regions. Thus, it makes a considerable difference, for example, whether you want to measure the long-term development trends of democracies and then-major features, as in the time-series data established by Vanhanen, Gurr and associates or Freedom House, or whether you want to supervise and "audit" all major features of a contemporary democracy and develop specific criticisms and policy advice, as done by Beetham and Weir (all of these will be discussed below).
In addition, of course, the usual "quality" criteria of empirical research, such as the validity and reliability of the respective measures and indices, will also be employed. Some authors, such as Cook and Campbell (1979), also distinguish in this regard between the "internal" validity and consistency
of individual findings and their "external" validity, which may lie in their possible generalization and broader theoretical relevance.
2.2. Longitudinal quantitative measurements
In the past, there have been frequent attempts to measure the degree of "de-mocraticness" of political systems using quantitative indicators (see, for example, Cutright 1963, Coulter 1975, Bollen 19S0, Coppedge and Reinicke 1990, Hadenius 1992). These and similar measures are also discussed in Inkeles (3 991), Beetham (1994) and, in a very detailed and constructive way, in Munck and Verkuilen (2002). Today the most frequently employed indices of democratization, which are also the only ones reaching back over a longer period, are the ones mentioned by Vanhanen (19S4, 1990, 1997), Gurr (Gurr, Jaggers and Moore 1990, jaggers and Gurr 1996), and Freedom House (Gas-Lil 1978 ft"., Freedom House 1990 ff.). Each has its particular merits, but also a number of problems and deficiencies.
The index of democracy (ID) developed by Tatu Vanhanen takes as its point of departure the two basic dimensions considered by Dahl— competitiveness and political participation—and operationalises them in a relatively simple, straightforward and more easily objectifiable manner, taking more generally available electoral data as its base. In his latest version (Vanhanen 1997), he covers 172 countries with data reaching back in some cases to the middle of the 19th century.
The index of democracy works this way: the degree of participation (P) is assessed by the voter turnout in consecutive elections in terms of the share of persons voting of the total population of a country. The competitiveness of elections (C) is measured by the share of the largest party in national parlia-' mentary elections subtracted from 100. Both measures are then multiplied by each other and divided by 100 to result in a scale ranging from 0 to 100 (P«C/ 100).
The aggregation by multiplication of these two dimensions is important because it avoids the possibility that one registers only sham elections with the usual 99.9% turnout, as in many of the former Communist states, but with zero competition, as has been done, for example, by purely additive procedures in previous indices (see, for example, Coulter 1975). Taking the share of voters of the total population, instead of the adult population only, distorts this index somewhat, disadvantaging "young" nations with a high population growth. But the author defends this on pragmatic grounds: total population data are more easily available than differentiated demographic census results. The index also necessarily disregards the fact that some countries, such as Brazil and Belgium, have compulsory voter registration and/or voting. This gives them a higher score on this index. The share of the votes of the largest
32
33
i .
I
■ > I \ f
parry in parliament also may produce certain distortions favoring highly frag- ; niented party systems with many small parties, which may. be a result of the electoral systems (for example, highly proportional ones). Therefore, beyond a \ certain threshold, a higher score on this count does not necessarily mean that a country has a "better" democracy in a functional or normative sense.
Ted Gurr and his associates have assembled another large historical data set (Polity I, II, m, and IV), which in its latest version comprises 157 contemporary countries that had populations of more than 500,000 inhabitants in the early 1990s. The data go back, in the oldest states, to about 1800, and to the year of independence of the more recently created states, Based on a variety of documents for each country, including, where they exist, the respective constitutions, Gurr and his associates identify three major dimensions of an "institutionalized democracy": the competitiveness of political participation, the openness of executive recruitment, and constraints on the chief executive. Five major indicators of these—the regulation and competitiveness of participation, the regulation, competitiveness and openness of the recruitment of the chief executive, and constraints on the chief executive—were coded separately, weighted and added up to a 11-point scale (which ranges from 0 to 10). In this way, they provide a wealth of information on most countries in the world on an annual basis in modem times. This can and has been used for a number of comparative and longitudinal studies (see also Harmel 1980, Jaggers and Gurr 1996).
Nevertheless, this data set has a number of important limitations. Its focus on the institutional side of democracy neglects certain broader aspects of so- J cial and political reality, such as the extent and kind of actual participation or the observance of civil liberties and human rights. It also tends to take some of the coded features of the "institutional democracies" at their face value without being able to assess their substance and actual performance. A certain coding bias favoring an American type of democracy, with a strict separation of powers, is also evident. This is probably to some extent inevitable with judgmental" data of this kind. These data, therefore, need to be supplemented with other sources.
The third continuous (since 1972) and constantly updated source of information are the Freedom House surveys on political rights and civil liberties. With the help of an elaborate checklist, Freedom House scores each country on a scale of 1 to 7, with 1 being the best value. Both indicators are also added up to provide an overall assessment of, at certain thresholds, "free," "partly free" and "not free" regimes. Strictly speaking, this is not an "index of democracy," but both the "political rights" (including competitive and fair and free elections) and the "civil liberties" indicators (concerning freedom of information, organization and religion, the absence of arbitrary repressive measures by the state, etc.) cover important dimensions of democratic systems. It must be noted, however, that these are basically subjective measures, the reliability of which cannot well be controlled by outsiders. In
the beginning, the codi ng was done by Raymond Gastil alone, with some inevitable bias and, by necessity, limited sources of information (see also Gastil 1991). In the meantime, the coding system and the data sources have become more elaborate, but a certain degree of subjectivity in the coding and weighting procedures (to avoid "numerical nonsense") and a lack of transparency of these procedures persist (see also Karatnycky 1998).
Thus, each long-term measure has certain emphases, but also certain weaknesses. Taken altogether, they correlate quite highly (with values of Pearson's r in the area of 0.80 to 0.90). Still, these correlations have become considerably weaker for the greater variety of the more recent democracies or regions with often less reliable sources of information, such as considerable parts of sub-Saharan Africa (see also Mc Henry 2000). High overall correlations may also disguise stronger discrepancies in individual cases and, indeed, quite differing assessments. For example, Yeltsin's Russia was rated with highly positive values in the Gurr and Vanhanen scales yet received only intermediate scores from Freedom House.
In recent years, two additional sources of information have become available which, having solid organizational bases, will be continuously up-dated in the future. One source are the comprehensive "good governance" indicators of the World Bank (Kaufmann et al. 2006) which among their six dimensions also assess the "voice and accountability" and the "rule of law" in 213 countries and territories on a regular basis since 1996. These also may be considered as indicators of two important dimensions of democracy. Their scales range from —2.5 to +2.5 where 0 can taken as a threshold distinguishing democratic from non-democratic regimes. The World Bank research group does not collect specific data for this purpose itself, but it draws on a large variety of sources and as, again, Freedom House, the Economist Intelligence Unit, Standard and Poor's Country Risk Review, the Gallup Millenium Survey etc. from which they compile their indicators by means of an "unobserved components model" (Kaufmann et al. 1999).
The other source is the "Bertelsmann Transformation Index" which assesses the political and economic transformation in 116 states worldwide (leaving out the more established OECD countries) since 1998. The criteria for political transformation include established statehood, political participation, rule of law, institutional stability, and political and social integration. Each of these dimensions is scored between I and 5 (5 being the best value) and then aggregated as a simple arithmetic mean into a single score. No specific thresholds are indicated. The scores are based on a panel of respective country experts (mostly from Germany), but counter-checked by informants, where possible, from these countries,
To varying degrees, all these indices share some common problems. First of all, there are some difficulties with the respective data bases and the sources of information available. The Vanhanen data are the "hardest" and
34
35
most objectifiable, being based on official election statistics, but even some of these statistics have to be taken at their face value without, in many instances, the opportunity to assess whether the elections have, in fact, been "free and fair" or whether a considerable amount of fraud and manipulation occurred. The "judgmental" data of the other two indices, with their problems of varying sources of information, selective perceptions, coder reliability, etc.. are even more problematic.
Secondly, since all these indices tap several dimensions of democracy, there is the problem of aggregating them in a meaningful way. Multiplication (as done by Vanhanen), which means that if one dimension has a value of zero the entire score is also zero, is certainly better titan merely additive indices, which may simply combine "apples and oranges,"
Thirdly, even though these indices look like interval or ratio scales with equal distances between the scoring points, they should not be (mis-) interpreted in this way. At least, they indicate some ordinal degrees of difference between the countries. But a difference between 1 and 2 on the Freedom House scale or 7 and 8 on the Jaggers/Gurr scale does not mean that this is the same difference as between 6 and 7 for Freedom House or 2 and 3 for Jaggers/Gurr (see also Elklit 1994).
Fourthly, and most importantly, the thresholds established in each scale to distinguish between democratic and non-democratic systems (5 for the Vanhanen index, 2.5 for Freedom House, 8 for Jaggers/Gurr) refer to qualitative differences between different types of political systems. As aggregated multidimensional indices, they should not be used to measure the "democratic-ness" of other distinct political system types, such as traditional monarchies, military dictatorships, communist one-party states or other types of authoritarian regimes. These are not more or less "democratic" in a comprehensive sense; they are qualitatively different. Otherwise, one would measure the "banananess" of apples and oranges! Instead, it makes more sense to keep the respective measured dimensions apart and to assess their values across different system types. Then it does, indeed, make sense to measure and compare the degree of participation, of competitiveness, of civil liberties, etc. in the respective regimes, just as one can measure the varying water or sugar content, etc., of different types of fruit.
Finally, one must remember that each index is at best a partial measure of some dimensions of democracy. Each index can, to some extent, supplement each other in a disaggregated way, but combining more dimensions may cause even more problems. This may eventually be done with the help of factor analysis in order to identify some common broader dimensions, but this requires still more comprehensive and consistent data. It is also evident that, so far, none of these indices, not even all of them combined in some way, cover all relevant dimensions of democracy in a comprehensive and nuanced sense (see Figure 3 in Chapter 1).
2.3. Qualitative assessments
The current indices of democratization provide information only about the state of affairs in a country at a given point in time; they do not assess the stability and durability of democratic systems as such. Therefore, countries' scores can abruptly change. Indeed, this has happened rather often. At least, the longer a democratic system has been in existence, the more likely it is to survive even further (see, for example, Przeworski et al. 1996), but democracies' actual effectiveness and longer-term chances of consolidation cannot be assessed in this way. Similarly, present indices neglect the output and overall "performance" side of democracies. The overall economic and "developmental" performance of political systems can be assessed with the usual indicators of GDP per capita growth rates, the "Human Development Index" and similar measures. In this regard, on the whole, during the last four decades, polyarchies have performed favorably compared with other political system types (see also Berg-Schlosser and Kersting 1996, Schmidt 1998, Przeworski et al. 2000).
From this overall "system" performance, we must distinguish a more specific "democratic" performance that refers to the overall quality of democratic systems in a functional and normative sense. The functional aspects refer to the specific "responsiveness" of political systems, their ability to provide effective mechanisms to respond to the articulated and aggregated preferences of large parts of the population in a meaningful way and to satisfy, at least to a larger extent and in the longer run, dieir demands and expectations. This idea of efficient feedback mechanisms of democracies is implied in the original system models by Easton (1965) and Almond and Powell (1978), but their actual workings need to be made more explicit and should also be amenable to some land of empirical measurement (see also Westle 1989).
Taking a more comprehensive "systems" view also makes the question of democratic consolidation more understandable and open to qualitative empirical assessments. In a widely accepted conceptualization of this term, Linz and Stepan (1996) distinguish between three aspects of consolidation at the macro-, meso-, and micro-levels of democratic systems.
At the macro-level, what they call "institutional consolidation," the proper functioning of the established political institutions and their interactions can be assessed. At the meso-level, which Linz and Stepan refer to as the "behavioral" one, one must examine whether there still are any significant antidemocratic social forces and actors, such as extremist right-wing or left-wing political organizations, a landed oligarchy, or parts of the military. At the micro- level, the "attitudinal" one in Linz and Stepan's terms, one considers the widespread acceptance of democratic values and procedures, and in this sense, the overall democratic legitimacy of a system as it is rooted in the respective political culture.
36
37
Only if at all three levels, for which specific qualitative evaluations and. io some extent, quantitative survey data with certain thresholds may be used, sufficient favorable evidence can be found, can a democracy be considered to be "consolidated." But even this state may not last forever. New crises may emerge which have to be coped with.
Taking a normatively even more demanding perspective is the "democratic audit" proposed by David Beetham and Stuart Weir. This, in part, has been put into practice (Klug, Starmer and Weir 1996, Weir and Beetham 1999). In a comprehensive and painstaking exercise, they originally examined four major aspects of democratic quality and performance: free and fair elections; open, accountable and responsive government; civil and political rights; and democratic society. For example, they found in the United Kingdom (in a way the mother of all contemporary democracies) severe constraints on the effective implementation and protection of civil and political rights and liberties. Furthermore, they consider the Westminster-type parliamentary system, which is often regarded as a model, to be, in fact, extraordinarily executive-dominated. They report that "the hallowed principle of parliamentary sovereignty amounts in practice to the supremacy of the near absolute executive over parliament" (ibid. pp. 491-496).
In the meantime, Beetham and Weir have expanded the original audit to cover 14 more detailed areas: nationhood and citizenship; the rule of law; civil and political rights; economic and social rights; free and fair elections; the democratic role of political parties; government effectiveness and accountability; civilian control of the military and police; minimization of corruption; the media and open government; political participation; government responsiveness; decentralization; and the international dimensions of democracy. At the same time, they are now attempting to develop their "democratic audit" in a comparative manner, work that is supported by the Institute for Democracy and Electoral Assistance (IDEA) in Stockholm. The results of these efforts remain to be seen, but are contributing an important development to the idea of a critical qualitative assessment, even of the "established" democracies.
Theo Schiller (1999) has listed such and similar criteria in a more systematic and comprehensive way (see Figure 4). They constitute a kind of "ideal type" of a full-fledged modern democracy, against which the "real, existing" ones can be contrasted and measured. He distinguishes five major principles of democratic systems (basic human rights, openness of power structure, political equality, transparency and rationality, and political effectiveness), which cover both the input and output sides and the respective feedback mechanisms. In addition, he lists the usual micro-, meso-, and macro-levels (see also Chapter 1), which results in this differentiated matrix with the respective emphases. Not all parts of this matrix are covered by currently available indices or have been sufficiently operationalized. But a variety of sources of information is available for such purposes, and it can be
further enhanced through the use of modern in format ion technologies and the Internet's potential to bring together expert judgments from practically all countries and backgrounds.
Figure 4: Principles of Democracy
Level: principles:	micro: individual citizens	meso: social and political groups and organizations	macro: political system, institutions
1, Basic human rights	personal rights, legal protection, freedom of opinion	freedom of organization, protection of minorities	limited state power, independence of judiciary, rule of law
2. Openness of pawer structure	free access to political communication and political power, rights of control	organizational pluralism, elite pluralism	separation of powers, limited terms of office, mutual checks and balances
3. Political equality	equality of voting rights, equality of political recruitment	equal opportunity for organizational resources	equal opportunity In the electoral system
4. Transparency and rationality	plurality of sources of information, chances for political education	independence and plurality of media, critical public	transparency of decision-making processes, rational discourses, documented bureaucratic procedures
5. Political efficiency and effectiveness	political interest, political participation, civic competence	effective aggregation of interests, mobilization of political support	effective decisionmaking rules and institutional balance, sufficient resources
Source: Adapted from Schiller (1999), p. 33.
2.4. Democratic sub-types
From such a comprehensive systems perspective and with the use of certain qualitative criteria, some sub-types of democracies can be distinguished. Most frequently, these refer to the particularities of the institutional set-up of certain sub-types. These include the usual distinctions between these types: presidential versus parliamentary (see, for example, Linz 1994), majoritarian versus consensual (Lijphart 1984, 1999), or federal versus centralized (Riker 1975). More differentiated regime types include "semi-presidential" (Duver-ger 1980, Sartori 1994), "parliamentary-presidential" and "presidential-parliamentary" (Shugart and Carey 1992). These have been extensively discussed elsewhere (see also Lijphart 1992) and need not concern us here.
In addition, a number of typologies have been developed that refer, in a broader system sense, to the social bases of politics, the particularities of the input structures (such as the party systems or the structures of interest media-
38
39
lion) and the'extent and shape of output institutions (including She military and the welfare system). Among the better-known of these typologies are those referring to particular class structures and their dynamics (Moore 1966, Rueschemeyer et al. 1992), the "parochial," "subject" or "participatory" political cultures (Almond and Verba 1963), two-or multi-party systems (for example, Sartori 1976), pluralist or (neo-)corporatist forms of interest mediation (for example, Schmitter and Lehmbruch 1979), problems of political control of the military (for example, Diamond and Plattner 1996), or particular types of welfare states (for example, Esping-Andersen 1990). Again, these sub-system specific typologies have generated their own extensive debates, which other publications have addressed.
What is of particular concern at this point are more recent notions of specific sub-types that cut across several of these sub-systems and mark specific deficiencies or defects of contemporary democracies (compare, for example, O'Donnell 1994, Merkel 1999). These can also be located in the overall systems framework and deserve some further discussion.
One of these, the "exclusionary" sub-type, refers to particular social groups or strata being excluded, de jure or de facto, from the regular institutionalized democratic processes of participation and decision-making. In the past, criteria of wealth or gender were often used to discriminate in this respect against some parts of the population (see, for example, Nohlen 2000). In many of today's new democracies, certain ethnic, religious or regional groupings are in some ways excluded from current political processes (compare, for example, Horowitz 1985, Diamond and Plattner 1994), and some socially and economically marginalized strata do not fully participate in local and national politics (Berg-Schlosser and Kersting 2000). This also raises the question of national citizenship, in particular that of recent immigrant communities and their respective legal status and possibilities of naturalization. This problem also concerns the more established democracies (Shapiro and Hacker-Cordon 1999a).
A second sub-type mat is relevant for this discussion concerns "enclave" democracy. This refers, for example, to countries where important groups and representatives of the preceding authoritarian regimes have secured a specifically protected position in the transition process that leaves them outside the mechanisms of usual democratic and judicial control. This includes particular constitutional prerogatives of the military, as in Brazil and Chile, or strong informal pressures, as in Paraguay, Thailand or Turkey (Loveman 1994). A special land of "enclave" or rather "exclave" outside the regular control of democratic authorities can also be found in certain favelas in Rio de Janeiro, for example, or, even more extremely, in some regions of Colombia where drug barons and/or rebel groups have created virtually "stateless" areas.
A third sub-type is constituted by "illiberal" democracies. In such systems, the independent normative and judicial control of the political execu-
tive is defective, and the rule of law both for the citizens in general and political opponents in particular is seriously impaired. This may even be endorsed by some populist measures, such as general referenda, to-change the constitution in order to extend the president's term of office, as in Fujimori's Peru. O'Donnell's (1994) notion of "delegative democracy" also refers to such illiberal or populist practices.
Finally, a strongly "clientelist" sub-type can be distinguished. Here, the informal ties between leaders and followers and their particular groups or regions prevail. In such asymmetric and unequal relationships, political support for the patron is exchanged for some material, often personal, benefits to the clients (Eisenstadt and Lemarchand 1981). These informal ties may be based on feudal or neo-feudal relationships in a modern context (as in a number of countries and regions where large landholdings prevail), on kinship or ethnic relations (as in a considerable number of African states), on religious affiliations and leadership (as with the marabouts in Senegal, for example), common regional or local origins, or populist personal appeals. There are also cases where hegemonic or long-time dominant political parties (as in Mexico, Japan, and Italy up to the early 1990s) or consociational systems, with their respective "political families" (as in Austria or Belgium, for example) have developed strong clientelist structures (Lauth and Liebert 1999).
As the latter examples show, the boundaries between particular defective sub-types and consolidated democracies that function adequately can be relatively hazy and fluid, and transitions (in both directions!) may occur (see, for example, Fox 1994). Similarly, some of these sub-types may be more prevalent in particular groups, regions, or localities of a certain country, where they may reinforce themselves over long periods of time and where particular "political cultural" patterns persist (Putnam et al. 1993).
In all such cases, it is, therefore, all the more important that the usual corrective mechanisms—open elective and parliamentary procedures, independent judicial control, pluralistic and independent media, and a well-informed general public and "civil society" (including attentive foreign observers)—be enhanced, and that stronger countervailing forces be created. Even some small steps in these directions, as with a more "competitive cli-entelism" in the new "electoral" democracies, the proliferation of new and more independent media, and greater international attention, may become important in the longer run. Once more and more groups and citizens become accustomed to democratic norms and procedures, they tend to become committed to maintaining and expanding those norms and procedures, and they can set in motion significant "self-cleansing" processes.
40
41
2.5. Conclusions
In this chapter I have presented a brief overview of some of the current concepts of empirical democratic research, some concrete procedures for their quantitative and qualitative assessment, and certain "defective" sub-types that are evident in some of the recent developments. I also made some criticisms and further proposals. Over and above diese conceptual considerations and problems of empirical measurements and operationalizations, the mare general theoretical concerns in which these are embedded must also be briefly addressed.
Regarding comparative empirical democratization research, we all share to some extent a "critical-rational" perspective in a Popperian sense. This need not be strictly "falsificatory" at all times with regard to our highly complex and malleable subject matter (see Popper 1972, Almond and Genco 1977), but, I would say, a certain progress can nevertheless be noted. We have a clearer perception and much broader and varied empirical evidence of what has happened during the last 15 years in particular, and even before that (see also, for example, the comprehensive analysis of the conditions of democracy and its failures in the inter-war period in Europe, Berg-Schlosser 1998, Berg-Schlosser and Mitchell 2000).
As is mentioned in Chapter 1, meta-theoretical, competing paradigms and approaches continue to confront political scientists. One concerns the "be-havioralist" tradition, which has been dominant for a long time, at least in some countries. This tradition still plays an important role, in , for example, survey research of democratization processes and political cultural changes at die micro-level (cf., for example, Rose, Mishler and Haerpfer 1998). A second school of thought continues to work in a historical-sociological and social-structuralist tradition (for example, Rueschemeyer et al. 1992). Thirdly, in-stitutionalist and "neo-institutionalist" approaches are receiving renewed attention (March and Olsen 1995). A fourth paradigm, which has gained increased popularity, in particular in the U.S. in the 1990s, is the "public" or more generally "rational choice" one (see, for example, Weingast 1996). At the same time, current national and sub-national developments are increasingly embedded in international trends of economic and political globalization, as emphasized, for example, by the "international political economy" school (see, for example, Cox 1987) and more general "cosmopolitan" concerns (for example, Archibugi and Held 1995). All these differing paradigms and emphases can and must also be seen from a "critical-normative" perspective (see, for example, Habermas 1992, Held 1995).
In my view, these paradigms, along with their respective groups of followers, need not be mutually exclusive. To some extent, they can meaningfully supplement each other and can be integrated into more comprehensive conceptual frameworks (such as structure- and actor-oriented
ones, and longitudinal historical and contemporary comparative analysis), through which we can share some of the respective advantages and overcome some of the specific limitations. Nevertheless, certain conflicts and confrontations, as in the real world, will continue. This too, in the long run, will advance our common knowledge and mutual understanding.
As both political scientists and citizens of this world, we are also confronted with the dilemma of the limits to our knowledge and expertise on the one hand and our desire to bring about change where we have detected serious flaws in contemporary democracies and practical politics on the other. International organizations such as the World Bank and the IMF, in tiieir "political conditionally" and also in the "political risk" analyses of internationally operating private banks and companies, already apply certain specific criteria in their current proceedings. These are often not very transparent and may also be seriously flawed. We have learned a lot in the meantime and, though history never repeats itself, political scientists have sound advice to offer that might affect the lives of many millions of people. The dilemma between scientific rigor and political relevance, as Laurence Whitehead writes in Chapter 5, however, will always be with us, and we each have our particular role to choose and to play.
42
43
3. Democracy Studies: Agendas, Findings, Challenges'
Gerardo L. Munck
Scholarly interest in the wave of democratization that began in southern Europe in 1974 resulted in a large amount of theoretical and empirical research. These studies took the nation state as the unit of analysis and focused on democracy as the outcome or dependent variable. Beyond this common overarching interest, however, different researchers have emphasized a broad range of aspects of the politics of democratizing countries, drawn upon various theoretical traditions, and used a diverse set of methods. As the literature has grown and evolved, the need for an assessment and synthesis of this literature has become imperative. Indeed, as with any research program, such periodic assessments and syntheses play a critical role, in ascertaining whether knowledge has been generated, and in identifying the challenges that remain to be tackled and the lines of research that are most likely to be productive.
This chapter responds to this need, offering a comprehensive evaluation of the body of literature on democracy that has been produced over the past 25 years." To organize the discussion, I distinguish among three agendas, which are identified by three concepts that define their primary explanatory concern: democratic transition, democratic stability, and democratic quality. For each agenda, I discuss the ways in which the subject matter has been or can be delimited and justified, and the main research findings. I also discuss challenges in three areas—the measurement of the dependent variables, the development and integration of causal theories, and the assessment of causal theories—and offer suggestions for tackling these challenges.
The challenges facing scholars currently active in the research program on democracy studies are considerable. My emphasis on these challenges, how-
1 I would like to thank Michael Bratton, Jae Foweralcer, Venelin Gancv, Sebastian Mazzuca, Guillermo O'Donnell, Robert Pahre, Timothy Power, Richard Snyder and Kurt Weyland for their useful comments on un earlier draft of this chapter.
2 This chapter does not discuss the extensive literature in comparative politics and international relations that focuses on democracy as an independent variable. Neither does this chapter address the growing literature on notions of citizenship that reach beyond and beneath the national state.
45
ever, is not meant to suggest that this research program laces insuperable hurdles. Rather, the point of this discussion is to use this assessment of the current state of the literature to identify the most productive avenues for future research. Indeed, my assessment is positive with regard to the achievements already made in the field of democracy studies and is also optimistic concerning the likely payoffs of future efforts to advance'this research program.
3.1. Democratic Transitions 3.1.1. The Subject Matter
Research on democratic transitions is a part of the broader field of democratic theory that gains its distinctiveness from a sharply defined focus on elections or, more specifically, on the critical step in the history of democracy when a country passes a threshold marked by the introduction of competitive elections with mass suffrage for the main political offices in the land. In other words, the status of democratic transitions as a distinctive field of research is given by an undeniably Schumpeterian approach to democracy, which emphasizes some key aspects of the procedures that regulate access to political power. This delimitation of the subject matter did little to spur interest at the time university-based research was expanding dramatically in the 1960s and 1970s. Not only did the realities of world politics appear to devalue this line of research. In addition, the Schumpeterian conception of democracy was widely out of favor. Even though some landmark studies on democratic transitions were published as early as I960 (Lipset I960, Rustow 1970), interest in democratic transitions took a back seat to other, more pressing and/or more valued concerns.
The status of research on democratic transitions, however, changed quite considerably thereafter. First and most important, the wave of democratization beginning in 1974 made the subject matter immediately relevant. Second, the change in values, especially among the left in both the South and the East (Barros 1986, Heller and Feher 1987), did much to place the Schumpeterian approach to democracy in a positive light.3 Finally, the seminal work of Guillermo O'Donnell and Philippe Schmitter (1986) did much to set the initial terms of die debate and hence to crystallize a field of research on democratic transitions.4 With the boom of research in the 1980s and 1990s, by the turn of the century research on democratic transitions had attained the status of an established field, justified on political and analytic grounds.
3 Probably the most carefiil defense of a Schumpeterian approach to democracy is offered by Przeworski (1999).
4 On the landmark status of O'Donnell and Schmitter's (1986) work, see Karl (2006).
First, the real-world significance of democratic transitions is undeniable. It has affected the lives of people all over the globe since approximately 1870, a rough landmark for the beginning of mass democracy (Finer 1997: 30). It emerged as a critical issue relatively early in a number of English-speaking countries: Great Britain, the United States of America, New Zealand, and Australia. For Western Europe as a whole, however, democracy remained a key issue on the political agenda from the late 19th century through to the end of World War II. And for yet an even larger number of countries, it was a dominant issue in the last quarter of the 20th century, as a wave of democratization that started in southern Europe in 1974 swept through Latin America, East and Southeast Asia, the communist-dominated countries that were part of the Soviet bloc during the Cold War, and parts of Africa.
The continued significance of democratic transitions, moreover, should not be underestimated. To be sure, when a democratic threshold is passed, the challenge of a democratic transition fades into the past and other issues begin to dominate the agenda. Nonetheless, it is unlikely that the problem of democratic transitions will cease to be of importance to the lives of tens of millions and even billions of people. On the one hand, the challenge of a democratic transition remains one of vital importance to a large number of countries. Depending upon the precise way in which the crossing of the threshold between authoritarianism and democracy is measured, in the year 2000 a full 40 to 60 percent of the countries in the world, including cases as significant as China and practically entire regions such as the Middle East, have never achieved democracy (Huntington 1991: 26, Diamond 1999: 25-28, Diamond 2003). And the "electoral revolutions"' in post-Soviet countries (Georgia 2003, Ukraine 2004, Kyrgyzstan 2005), among other political events, showed that the push for democratic transitions continued into the new century. On the other hand, countries that have passed the democratic threshold always face the possibility of a democratic breakdown. Indeed, even in the middle of the democratic wave of the last quarter of the 20th century, numerous countries experienced breakdowns. And there are grounds to think that many of the newly minted democracies are unlikely to endure as the 21st century unfolds. In sum, a concern with democratic transitions has had, and is likely to continue to have, great relevance.
This delimitation of a field of research focused on democratic transitions is also justified on analytic grounds. The conceptualization of democratic transitions in terms of a threshold marked by the introduction of competitive elections with mass suffrage for the main political offices excludes a large number of issues that are a concern of democratic theory. For example, it is set off from such fundamental issues as the variable ways in which public policy is formulated and implemented in democratic countries; the extent to which the rule of law is respected; and the increasingly important concern
46
47
I
about the extension of democratic rule, traditionally a principle applied to the < nation-state, to a range of other units, both of different territorial scope and with different functional aims. What may appear to be an unwarranted narrowing of concerns, however, is analytically justifiable.
The decision to focus on democratic transitions is driven by two insights. First, it is based on the understanding that the introduction of competitive elections is an event that is fundamental enough to alter a country's political dynamics and that calls, therefore, for its own explanation (O'Donnell and Schmitter 1986: Ch. 6, Shain and Linz 1995: 76-78). Second, this decision is justified on the ground that a transition to democracy is a process that is distinct enough, compared with the other concerns raised in democratic theory, to suggest that it is caused by factors that probably do not affect other aspects of democracy and that it is most fruitfully theorized on its own terms (Rus-tow 1970, see also Mazzuca 2007).
A focus on democratic transitions, thus, does not deny that countries vary along other dimensions or that these other dimensions may be as important as those highlighted by a Schumpeterian approach. Indeed, as current scholarship shows, a range of issues not encompassed by Schumpeterian definitions of democracy are likely to have great relevance in countries where democracy is firmly established (O'Donnell 1999: Part TV). Therefore, the delimitation of democratic transitions as a distinct area for scholarship is not based on a judgment about the importance of a Schumpeterian approach compared with any other approach but is rather a conceptual decision, which helps to distinguish dimensions of concern within democratic theory that most likely vary independently from each other. That is, the point is not to argue that one or another issue is more important but to provide a basis for an analytic approach by breaking down democratic theory into a series of distinct and hence manageable explanatory challenges.
3.1.2. Findings
The sharp delimitation of the subject matter of democratic transitions and hence the formulation of a fairly clear question—why have some countries had democratic transitions while others have not?—had an important benefit. By providing researchers with a pointed and widely shared agenda, it allowed for the rapid generation of an impressive basis of knowledge through a succession of studies that eventually came to encompass most cases of democratic transition in world history.
Following in the wake of a key study of transitions in southern Europe and Latin America in the 1970s and early 1980s (O'Donnell, Schmitter and Whitehead 1986), major cross-regional analyses were conducted comparing Latin America with East and Southeast Asia (Haggard and Kaufman 1995), and southern Europe and Latin America with Eastern Europe and the former
I Soviet Union (Linz and Stepán 1996). Excellent region-based studies were
produced, focusing on Africa (Bratton and van de Walle 3997), Eastern Europe and the USSR/Russia (Beyme 1996. Offe 1996,-Bunce 1999), post-Soviet Eurasia (Hale 2005, Collins 2006), as well as the three major regions of the developing world (Diamond, Linz and Lipset 1989a, 1989b, 1989c; Huntington 1991). In addition, impressive efforts were made to put the transitions of the last quarter of the 20th century in historical perspective through cross-regional analyses of Europe and Latin America ranging across the 19th and 20th centuries (Rueschemeyer, Huber Stephens and Stephens 1992, Collier 1999) and, along similar lines, analyses focused on the older European cases of transitions were offered (Boix 2003, Tilly 2004).5 Finally, a number £-:':.:V'v-'.-  of statistical studies contributed to the debate.0
The richness of this literature is undeniable. It offers a wealth of ideas on the causes of transitions, a great amount of nuanced information on complex :processes, and some fruitful comparative analyses that have generated a : number of important and surprising findings. This literature has challenged the longstanding modernization argument that level of economic development is a good predictor of transitions to democracy (Przeworski and Limongi 1997).7 It has shown, again contrary to what was posited by modernization theory, that democratic transitions do not occur through a single process but rather through multiple paths defined by factors such as the power and strategies of elites and masses and the top-down or bottom-up impetus for political reform (Dahl 1971: Ch. 3. Stepan 1986, Dix 1994, Collier 1999, Tilly 2004).
The codification of these distinct paths of democratic transition has led to other important findings. First, it has allowed analysts to establish that the path toward democracy that a country follows is strongly influenced by its ; . type of prior, non-democratic regime, and that the very likelihood of a transi-
tion to democracy is affected by the type of actors that oppose authoritarian
5 See also Janoski (1998: Chs. 6 and 7), who carries out n useful test of influential theories against European history from 1200 to 1990, and Halperin (1997), who studies Europe in the period 1789-1945 and draws some comparisons between Europe's and Latin America's experience.
6 GasiDroivski (1995), Przeworski and Limongi (1997), Coppedge (1997), Przeworski, Alvarez, Cheibub, and Limongi (2000), Boix (2003), Boix and"Slokes (2003), Brinks and Coppedge (2006).
7 See also Przeworski, Alvarez, Cheibub and Limongi (2000: Ch. 2), Mainwaring and Perez-Linan (2003), and Gleditsch and Ward (2006: 925-26). Relaledly, a new literature considers the actual resources that underpin a country's level of economic development and theorizes haw these resources are politically used and that wealth does not necessarily lead to democratization (Ross 2001, Bellin 2004, Jensen and VVantchekon 2004, Snyder 2006). However, die extent to which level of economic development is or is not a good predictor of democratic transitions continues to be debated and various authors have contested Przeworski and Limongi's (1997) argument (Geddes 1999, Boix and Stokes 2003, Epstein, Bates, Gold-slone, Kristensen and O'Hnlloran 2006).
48
49
rule (Linz and Stepan 1996, Bratton and van de Walle 1997: 9-14, Munck I99S: 17-22, Ch. 7, Leff 1999). More pointedly, because pacts may be a necessary condition for a successful transition to democracy in the context of certain types of regime, the prospects of democracy are enhanced when opposition demands are amenable to negotiated resolution. This is most likely, in turn, when the supporters and opponents of authoritarianism are economically interdependent—that is, class actors—than when opposition to authoritarianism is led by a nationalist or fundamentalist religious movement (Arfi 1998, Roeder 1999, Wood 2000, Hamladji 2002).
With regard to specific classes, though much research has been conducted to ascertain whether the bourgeoisie (Moore 1966), the middle class (Lipset 1960) or labor (Rueschemeyer, Huber Stephens and Stephens 1992) is the prime agent of democratization, and whether the landed elites are an inherently undemocratic force (Moore 1966), the literature is mostly inconclusive. Indeed, probably the only clear finding about the social origins of democracy is that landed elites that depend on labor-repressive practices have a negative effect on the installation of a democratic regime (Mahoney 2003: 137-45, Bernhard 2005, see also Mainwaring and Perez-Linan 2003: 1046-50).8
Research has also shown that democratic transitions are closely linked with matters of the state, conceived in Weberian terms. As shown, processes of regime change that lead to state decay or state collapse reduce the prospects of democracy (Linz and Stepan 1996: 17-19). Thus, a key finding is the principle: "no state, no democracy."9 And a growing body of literature offers considerable evidence that international factors have an influence on democratic transitions. The scholarship that considers the broad historical sweep of
8 To anticipate partially some of the suggestions offered below, there are two significant problems with the literature an the class origins of democracy. On the one hand, efforts to theorize and test hypotheses about the role of different classes have operated with an aggregate dependent variable. The role of different classes may have a strong impact at different stages in the process of democratization, but this impact may not be discerned or may become diluted when democratization is viewed os an aggregate process. Relatedly, inasmuch as there are various paths to democracy, one would expect that different classes would play a more prominent role in certain paths (Collier 1999), a finding that again gets lost when the dependent variable is studied at an aggregate level. On the other hand, this literature has tended to Focus on the impact of each class viewed in isolation of odier classes and other explanatory factors, and to conceive of the impact of classes in linear terms. Yet it seems quite obvious that theorizing requires attention to interaction and threshold effects. For example, the strength of the musses may induce elites to extend the right to vote, as a way to faster moderation.. But if the masses are very strong relative to the elites, democracy may be forestalled by a successful revolution tram below or a retreat from a commitment to reform hy elites fearful of the redistributive consequences of mass democracy. Similarly, the disposition of middle classes to fight for democracy has oscillated in response to shifts in the relative power of other actors (Rueschemeyer, Huber Stephens and Stephens 1992: 272). Inasmuch as these issues are tackled, it is quite possible that this important line af research will yield clearer findings than it currently offers.
9 1 am indebted to Richard Snyder far his suggestion of this phrase.
democratization makes a strong case for the role of conquest and colonization (Therborn 1977, 1992, Tiljy 2004). Furthermore, recent statistical research ha's found that contiguous neighbors and regional contexts are associated with a diffusion effect and, specifically, that the likelihood that a country will undergo a democratic transition increases when neighboring countries and a country's region are more democratic.10 Likewise, belonging to international organizations with high membership of democratic states increases the probability of a transition to democracy (Pevehouse 2005: Ch. 4).
It is important not to exaggerate the confidence placed in these findings. There are many exceptions. Indeed, numerous works on the USSR and post-Soviet countries strongly question the extension of generalizations based on democratic transitions in Southern Europe and Latin America to cases where communism held sway in the post-World War II decades (Bunce 2000, 2003, McFaul 2002, Gelman 2003). Moreover, the results of much statistical research are neither very robust nor based on strong research designs (Robinson 2006: 504, 517-25). Nonetheless, with this caveat, it is important to have a clear sense of the current state of knowledge and to build on this knowledge as research continues and as new challenges are tackled.
3.1.3. Challenges
The challenges faced by students of democratic transitions concern many basic research tasks. The way in which the outcome of interest has been measured is open to improvement. The need for greater integration of causal theories is increasingly apparent. And the assessment of causal theories that combine qualitative and quantitative forms of research is emerging as yet another important challenge, indeed, the future development of the research agenda on democratic transitions is likely to hinge on the ability of scholars to tackle some broad and fundamental challenges.
Measuring the Dependent Variable
A first challenge concerns the dependent variable of research on democratic transitions. As stated, this research has focused on a sharply defined subject matter. But the common practice of using an event—the holding of free and fair elections that lead to the installation of authorities with democratic legitimacy—as an indicator that justifies changing the way an entire country is scored on the outcome of interest is problematic.
To be sure, this way of coding cases, which draws on the notion of a "founding election" (O'Donnell and Schmitter 1986: 61), has some validity when applied to transitions in die post-1974 period. The reason is that a
It) Kopstein and Reilly (2000), Brinks and Coppedge (2006), Gleditsch and Ward (2006), Mainwaring and Perez-Linan (2007).
50
51
common elite, strategy in the late-i 9th and early-20th centuries—the gradual extension of voting rights, first to propertied males, then.to all males, and subsequently to women—was probably not viable and thus not used in late-20th century transitions. Thus, to a certain extent, it is appropriate to view recent democratic transitions as unfolding in a non-incremental fashion and along the various dimensions of democracy all at once. But even when applied to recent transitions, the limitations of this approach to measuring democracy are significant.
For example, though some researchers use this approach to code Chile as a democracy from 1990 onward (Przeworski, Alvarez, Cheibub and Limongi 2000: 64), it is obvious that even though Chile became fully democratic along some dimensions of democracy, it did not do so along others. Specifically, the fact that a sizable portion of the Senate was not popularly elected meant that it suffered from an important democratic deficit concerning the range of offices filled through elections. Moreover, as this example illustrates, the use of a dichotomous measure does little to capture the incremental nature of Chile's democratic transition and hence the distinctive nature of Chile's politics in the 1990s: the gradual and incomplete nature of its transition to democracy. What is needed is a measure of democratic transitions that clearly distinguishes among multiple dimensions and can capture the possibility of gradual change.
Some efforts have been made to address this problem. A recent literature on hybrid regimes has drawn attention to a key insight: that a considerable number of countries seem to be neither fully democratic nor blatantly authoritarian and thus are best characterized with intermediate categories (Diamond 2002, Schedler 2002, 2006). And a well-developed literature on quantitative measurement offers several examples of multi-dimensional measures that do not use dichotomous scales (Epstein, Bates, Goldstone, ICristensen and O'Halloran 2006, Mainwaring, Brinks and Perez-Linan 2007). But these efforts fall short of solving the problem. The discussion of hybrid regimes is characterized by the proposal of an unwieldy number of dimensions and has still not been linked with any efforts to develop systematic data. And efforts to develop quantitative measures are still grappling with the problem of establishing thresholds that correspond to the concept of a democratic transition. Thus, they still are not able to clearly distinguish changes of regime, that is, from an authoritarian to a democratic regime, from changes within a regime. In short, much remains to be done before the study of democratic transitions can rely on good data."
11 For more an these issues, see Munek and Verkuilen (2002), and Munck (2006). 52
Integrating Causal Theories
A second challenge concerns the need for greater integration of causal theories. The evolution of the literature on democratic transitions has been characterized by the frequent introduction of new causal factors considered critical to an explanation of why democratic transitions occur. These new explanatory variables sometimes reflect the experience of new cases of transition to democracy, which have brought to light factors that had not seemed important in the cases until then considered. In other instances, the focus on new variables has been driven more by an effort to rescue insights from older bodies of literature. Over time, then, the number of explanatory variables has multiplied, pointing to an important trade-off in this literature between theoretical fertility and orderly theory building.
As challenging as the task of theoretical organization and integration is likely to be, it is facilitated somewhat because theoretical debates have evolved around a number of central axes. One main axis contrasts short-term factors and the choices made by actors (O'Donnell and Schmitter 1986, Przeworski 1991) with medium-term factors, such as the characteristics of the old regime (Linz and Stepan 1996, Chehabi and Linz 1998) and long-term, more structural factors, such as the mode of production or the model of development (Rueschemeyer, Huber Stephens and Stephens 1992, Castells 1998). Another axis of debate contrasts elite-centered explanations (Dogan and Higley 1998) with mass-centered explanations, which focus either on class actors (Rueschemeyer, Huber Stephens, and Stephens 1992, Collier 1999, Bellin 2000), social movements (Foweraker 1995: Ch. 5, Tilly 2004), or ethnic groups (Offe 1997: Ch. 4). Yet another axis contrasts political with economic determinants of transitions (Haggard and Kaufman 1995, Przeworski and Limongi 1997). And one more critical axis of debate opposes domestic factors to international factors (Whitehead 1996, Levitsky and Way 2006), an axis along which one might also locate explanations centered on stateness and "intermestic" nationality issues (Linz and Stepan 1996).
This way of organizing the literature has merit and helps to introduce some order into the debate. Moreover, it is noteworthy that, as the literature on democratic transitions grew and introduced new explanatory variables, scholars sought to impose some organization on theorizing, either by pulling together the range of explanatory variables (Diamond, Linz and Lipset 1995) or by attempting to syndiesize a range of these explanatory factors (Kitschelt 1995: 452-55, Mahoney and Snyder 1999). However, the challenge of integrating and synthesizing the diverse set of explanatory factors proposed in this literature and the generation of a more parsimonious theory remains to be adequately tackled.
In this regard, the potential gains associated with efforts to build rational choice-theoretic and game-theoretic models of democratic transitions should
53
be noted. This literature is distinctive in that it employs a common theory, which facilitates theoretical cumulation. Moreover, inasmuch as it employs a [ formal methodology, it also brings to bear the power of deductive logic, which has the advantage of demonstrating what implications follow from a given set of assumptions. These advantages notwithstanding, it is worth highlighting that to a large extent, the rational choice literature on democratic transitions has reproduced the problems of the broader literature.
On the one hand, much as with any approach to theory generation, game-theoretic models are driven by insights about specific cases or regions. As a result of this inductive aspect of the modeling process, game-theoretic models propose explanatory factors that diverge widely in terms of their empirical scope. On the other hand, the explanatory variables themselves differ from model to model. Thus, some rational choice theorists seek to explain democratic transitions with tipping models, which focus on proximate factors and draw attention to the contingent nature of processes of democratic transition, specifically by highlighting the critical role of triggers or tippers (typically students, intellectuals or dissidents), and cognitive aspects, such as belief cascades (Kuran 1995, Petersen 2001). Others offer models that emphasize the explanatory role of the prior, non-democratic regime, seeking to show how actors within certain institutional settings engage in patterned forms of action (Geddes 1999). And yet others develop what might be labeled political economic models that focus on the long term and see action as driven by the interests of actors, which are conceived either in class terms or more broadly as elites and masses (Acemoglu and Robinson 2001, 2006, Boix 2003). In sum, much as the rest of the literature, rational choice theories of democratic transitions diverge in terms of their empirical scope and explanatory variables.
Even as the search for theoretical principles that would provide a basis for theoretical integration and synthesis thus continues, another approach to the task of theoretical integration that deserves emphasis is closely connected to the previous challenge of defining and measuring the dependent variable more carefully. This strategy has gone, for the most part, unrecognized. But it has the potential to yield important payoffs. Indeed, inasmuch as the concept of democratic transition is carefully disaggregated and measured in a nuanced way, such work can be used to break down the big question at the heart of research on democratic transitions—why have some countries had democratic transitions while others have not?—into smaller, more analytically tractable questions.
The disaggregation of the broad problem of democratic transitions into its constituent parts, and the use of measures that distinguish a variety of meaningful thresholds, is likely to assist in the identification of explanations by helping analysts distinguish and avoid the conflation of aspects of democracy that are likely to be driven by different processes. For example, because there
are good reasons to believe that the extension of the right of suffrage for men is driven by a different process than the extension of the right to vote for women, the disaggregation of the explanatory challenge .in such a way as to explicitly capture this distinction is likely to help analysts uncover stronger associations. And, in turn, such an approach may help to show how arguments that are presented as competing may actually be complementary. Indeed, once a disaggregated approach to democracy is employed, there would be little reason to consider the theses advanced by Barrington Moore (1966) and Dietrich Rueschemeyer, Evelyne Huber Stephens and John Stephens (1992) as rival explanations. Rather, in that democracy is defined by Moore (1966: 414) in terms of the dimension of contestation, and by Rueschemeyer, Huber Stephens and Stephens (1992: 303-04) in terms of participation, it seems clear how their theories might be considered as partial contributions to a general theory of democratic transitions.
In short, it is important to focus on the challenge of theoretical integration. And to tackle this challenge it is worth considering the way in which theoretical debates have already been organized to a large extent along certain axes, to continue the search for theories that serve as unifying principles, and to recognize how analysts might integrate research findings regarding conceptually connected parts of a broad question.
Combining Qualitative and Quantitative Forms of Causal Assessment
Finally, a third challenge, concerning causal assessment, touches upon the as-yet barely addressed problem of how to combine qualitative and quantitative forms of research. Research on democratic transitions has been pioneered by researchers who have given primacy to small-N and medfum-N comparisonsi The reason for this strategy is obvious, in that the comparison of a small number of cases has been particularly well suited to the crafting of fertile concepts and has also provided a sufficient basis for introducing new ideas into the debate—and for doing so rapidly. Moreover, the use of qualitative forms of analysis has had the added benefit of being useful for the task of causal assessment, in particular because its intensive nature and its emphasis on process tracing makes it suited to address theories that highlight the role of actors and changing situations.
This strategy, however, has also had its problems. Qualitative researchers are limited in their ability to test the generalizability of their dieories and to offer precise estimates of causal effect that take into consideration a variety of sources of bias. Moreover, they have not always been as systematic as they could be, For example, though this literature has generated a great amount of nuanced data, researchers have not always gathered data on all the explanatory variables for all the cases they analyze, nor have they always coded cases ex-
54
55
plicitly according to a set of clear criteria. Finally, small-N researchers have not given enough attention lo issues of research design and have rarely conducted stringent tests of their theories. As a consequence, researchers' ability to test their theories and draw strong conclusions has been constrained.
Though the weaknesses of qualitative research on democratic transitions are not all inherent to the methods used and thus much progress can be made by improving qualitative research, they certainly point to the need to combine qualitative and quantitative forms of analysis. But, unfortunately, combining these two types of research is far from easy. Indeed, though quantitative research on the question of democratic transitions has been produced, the links between qualitative and quantitative research on democracy have been very weak. First, the measures of democracy used by quantitative scholars tend to differ significantly from those used by qualitative scholars. What these scholars think of as democratic transitions, thus, may be quite different things. Second, the causal theories quantitative scholars actually test are often caricatures of the theories discussed in the qualitative literature. In this regard, existing statistical tests have been very limited. Practically without exception, they have focused on a narrow range of independent variables, related primarily to social and economic questions, ignoring a variety of theories cast in terms of the role of actors and choices. Moreover, tests have tended to use additive models and also, for the most part, linear models that severely misrepresent the causal argument in the literature. Finally, large-N data sets have typically consisted of one observation per case per year. This restricts their sensitivity to issues of time and process, which rarely obey the cycle of calendar years. Indeed, it is important to recognize that there is a very steep trade-off in the level of richness of information and explanatory arguments discussed as one moves from the literature based on intensive but relatively narrow comparisons of a small set of cases to the statistical literature based on a large number of cases.1"
The difficulties of using a genuine multi-method approach that combines qualitative and quantitative methods suggest that future research should probably be based on a continuation of the multi-track approach used so far. The qualitative track is likely to yield significant dividends by extending the intensive analysis of a small to medium number of cases to some relatively unexplored questions. Some significant works offer a historical perspective on the democratic transitions that have been at the heart of the debate, those occurring in the last quarter of the 20th century (Ruesche-meyer, Huber Stephens and Stephens 1992, Collier 1999). But much
12 Indeed, from this perspective, the most fruitful comparative studies, in that they use hnrd-to-callect data to test complex and dynamic theories, while reluming □ broad enough basts to moke claims about generalizability, have focused on a medium number of cases (that is, roughly 8 to 20 cases) (see, far example, Huntington 1991, Haggard and Kaufman 1995' Linz and Stepan 1996, Collier 1999).
remains to be learned by cross-time comparisons and a re-analysis, in light of new theories, of the older cases of transitions discussed by Barrington Moore (1966), Reinhard Bendix (1978), and Michael Mann (1987, 1993),
In addition, qualitative research can make contributions by broadening the variation on the dependent variable it seeks to explain. The existing literature has tended to focus on positive cases and has introduced variation longitudinally by studying the process whereby countries that were authoritarian become democratic, and through the concept of modes of transition (Main-waring 1992: 317-26). Beyond this, some insightful work has been done comparing cases of transitions that led to democracy but also to other outcomes (Collier and Collier 1991, Yashar 1997, Snyder 1998, Mahoney 2001). But, overall, when it comes to events in recent decades, little attention has been given to the need to explain failed democratic transitions, that is, cases where transitions from authoritarian regimes did not occur or lead to new authoritarian regimes.13 Indeed, important questions that remain to be fully answered are: why did many countries that saw the collapse of authoritarian regimes during the last quarter of the 20th century experience transitions that did not lead to democracy? And, why have some countries not had transitions at all? Especially inasmuch as this research is explicitly connected to the existing literature and both draws upon its strengths and hones in on its lingering problems, the continued use of qualitative methods focused on these and other questions is likely to be highly rewarding.
The quantitative research track, in turn, is likely to contribute to the debate inasmuch as it addresses two tasks. One is the need for statistical research that is more keenly aware of problems of omitted variables and en-dogeneity, and hence that is more concerned with matters of research design and is more closely connected to theory. A second, related task concerns the collection of data. Not only should data collection focus on factors other than the standard economic and institutional ones, which are the staple of statistical analyses. Data collection should also be driven by the need for data that reflect the unfolding of events more closely than the standard practice of gathering one observation per case per year. Indeed, the full benefits of statistical tools are likely to be felt in the debate on democratic transitions only once data sets are generated with information on the actors involved in the process of democratic transitions, the choices these actors make, the sequence of events whereby democratic transitions unfold, and the institutional setting in which actors operate.
The tasks facing quantitative researchers are formidable but they promise important payoffs; thus, they are well worth pursuing. It is this sort of research that will finally bring the strengths of distinct research traditions to
13 See, however, the discussion of post-Soviet cases in McFaul (2002) and Collins (2006), of the Middle East in Bcllin (2004) and Anderson (20D6), and of China in Pei (1994,2006).
56
57
bear on the same research question, rather than remain as two somewhat disconnected approaches that never quite talk to each other. Indeed, a multi-track approach, if properly implemented, could offer an important stepping-stone and gradually give way to a truly multi-method approach that would show how qualitative and quantitative research can be conducted in a complementary fashion.
3.2. Beyond Democratic Transitions 3.2.1. The Subject Matter
Research on the politics that follows the completion of democratic transitions is harder to assess than research on democratic transitions for the simple reason that there, is a lack of consensus concerning the subject matter. Moreover, some ways of defining the subject matter do not offer a clearly delimited focus for research. Overall, the agenda put forth by what might be labeled regime analysts does share certain common elements. Thus, it can be contrasted as a whole with the voluminous research on narrower, institutional issues, which are standard in the study of advanced democracies and are increasingly a concern of students of new democracies.14 This commonality notwithstanding, regime analysts have conceptualized post-transitional politics in such diverse ways that the organization of the field of research around clearly defined questions has been hampered.
The core of the problem is as follows. Initially, one concept—democratic consolidation—was widely used as a way to identify the subject matter of interest. This concept was useful. It helped to identify issues that went beyond those discussed in the literature on democratic transitions. And it provided an overarching frame for theorizing (Schmitter 1995, Schedler 1998, Merkel 1998, Hartlyn 2002). However, over time this concept was used in such different ways that it ended up creating severe confusion. Then, as a way to clarify the agenda of research, scholars gradually introduced a new concept, the quality of democracy.'5 But this new concept was itself defined in such a variety of different ways that it did little to solve the problems
14 Institutional issues are of obvious relevance to fundamental questions in regime analysis. This much is evident, for example, from the debate over the relative impact of consocia-tional versus majoritarian arrangements, and presidentialism versus parliamentarism, on the durability of democracies. But institutionalists more frequently take the democratic nature of the regime for granted, while regime analysts are explicitly concerned with the ongoing salience of the democracy question.
15 Przeworski et ol. (1995; 64), Linz and Stepan {1996: 137-38, 200), Linz (1997: 406, 417-23), Diamond (1999: 28, 132), Huber and Stephens (1999: 774), Kitsehelt, Mansfeldova, Markowski and Toka (1999: 4-9), O'Donnell, Vargus Cullell and lazzetia (2004), Diamond and MorlinD (2005).
associated with the concept of democratic consolidation. Indeed, the concepts of democratic consolidation and the quality of democracy used by scholars have varied so much that research could simply not build around a clear set of shared questions.
To move forward, hence, some basic terminological and conceptual choices must be made. First, with regard to democratic consolidation, it is probably best, as some researchers have suggested, to simply jettison the term "democratic consolidation" (O'Donnell 1996) and focus instead on democratic stability, understood as involving nothing more than the sustainability or durability of the democracy defined in Schumpeterian terms, which result from successful democratic transitions. Second, with regard to the quality of democracy, it is necessary to make some theoretically based decisions concerning the concept that would serve to specify precisely how this agenda is distinct from those of democratic transitions and democratic stability. Indeed, these choices are essential to future progress in the field of democracy studies.
3.2.2. Democratic Stability
A focus on democratic stability, as opposed to democratic consolidation, helps to articulate a delimited yet still relevant subject matter. Indeed, research on democratic stability focuses on a clear question, why have some democracies been more stable than others? And the relevance of this question is hard to dispute. Very few countries have followed the path of Great Britain, which moved toward democracy without ever suffering any major reversal of its democratic gains. Thus, the potential breakdown of democracy has been an important concern of students of democracy.
In the context of Western Europe, the history of France offers dramatic evidence of the potential for democratic reversals. In turn, the interwar period not only gave us the paradigmatic case of breakdown, Weimar Germany, but also showed how the breakdown of democracy could become a widespread phenomenon (Berg-Schlosser and Mitchell 2000, 2002). And the collapse of democracy in Greece in 1967 showed that even post-World War II Europe was not immune to the forces that could lead to an authoritarian bacldash.
Beyond Western Europe, the history of post-World War II Latin America is punctuated by frequent democratic breakdowns, including the dramatic replacement of democracies by harsh authoritarian regimes in the 1960s and 1970s. Similarly, the African continent witnessed the breakdown of numerous democracies in the early post-colonial period, and the history of important cases such as Nigeria is essentially one of die oscillation between democracy and authoritarianism. Even in Asia, where India provides a nota-
58
59
ble exception,'" cases such as Pakistan are a reminder of the lack of guarantees that the establishment of democracy does not always lead to democratic stability.
Finally, even the most recent wave of democratization did not end the continued relevance of concerns about democratic stability. In Russia, the closing and bombing of the parliament in 1993, the serious doubts about whether elections were going to be held in 1996, and the lack of basic freedoms during the Putin years, helped drive this point home. Developments in Belarus under Lukashenka, added further weight to worries about the erosion of democracy in post-communist countries. Even more unambiguously, democratic breakdowns in several cases in Latin America (Haiti 1991 and 2004, Peru 1992, Ecuador 2000), Africa (Nigeria 1983, Sudan 1989, Niger 1996, Sierra Leone 1997, Ivory Coast 1999, Central African Republic 2003, Guinea-Bissau 2003) and Asia (Thailand 1991 and 2006, Pakistan 1999) raised concerns about the potential of significant democratic losses. In short, democratic stability offers a more tractable object of analysis and one that has been and continues to be of great relevance.
Findings
A focus on the more delimited subject matter of democratic stability also greatly aids the task of sorting through the large relevant literature and uncovering findings.17 Much of this research, and especially the qualitative research, relies on the concept of democratic consolidation. But it is usually possible to separate out a narrower understanding of consolidation as stability. And it is also possible to identify some surprising and some not so surprising findings on this delimited subject matter.
It is worth highlighting at the outset that a set of factors that were considered as potential determinants of the durability of democracy have been
16 But even this exception is somewhat tainted by the restrictions placed on Indian democracy during the 1975-77 years.
17 The research on democratic stability, much as thai on democratic transitions, draws on various approaches. It includes important regional studies, on Latin America (Karl 1990, Mainwaring, O'Donnell and Vulenzuela 1992), Southern Europe (Gunther, Diamandouros and Puhle 1995, Mariino 1998), Eastern Europe (Elster, Offe and Preuss 1998, Tismaneanu 1999, Janos 2000), and Africa (Joseph 1997, Wiseman 1999, Bratton, Mattes and Gyimah-Boadi 2005, Lindberg 2006). Noteworthy works also offer cross-regional analyses, comparing southern Europe and Latin America (Higley and Gunther 1992*), Latin America with East and Southeast Asia (Haggard and Kaufman !995: Part III), southern Europe and Eastern Europe (Maravall 1997), southern Europe and Latin America with Eastern Europe and the Former Soviet Union (Linz and Stepnn 1996, Przeworski et al. 1995, Diamond 1999), and Asia with Eastern Europe and the former Soviet Union (Shin and Lee 2003). Moreover, thinking about democratic stability has been influenced by statistical studies to a greater extent than has research on democratic transitions (Remmer 1990, 1991, 1996, Diamond 1992, Przeworski and Limongi 1997, Power and Gasiorowski 1997, Gasiorowski and Power 1998, Mainwaring 2000,1-ladenius and Teorell 2005).
shown not to have much explanatory power. This applies to various propositions about the impact of the old regime and the modality of transition to democracy (Karl 1990, Karl and Schmitter 1991),'" sequencing of economic and political reforms (Haggard and Kaufman 1992, Przeworski 1991: 180-87). economic performance and crises (Przeworski 1991: 32, 188), the strength of civil society and political parties (Ruescliemeyer, Huber Stephens and Stephens 1992: 6, 49-50, 156, Mainwaring and Scully 1995: 1-2, 21-28), and the presidential or parliamentary form of democracy (Linz 1994). 9 In brief, countries that became democratic since 1974 display a tremendous amount of variation with regard to these explanatory factors, yet they have had a fairly common outcome: a durable democracy. Moreover, even departures from this trend toward democratic stability do not appear to be strongly correlated to these factors,
This research has also produced some positive findings about the conditions of democratic stability. To a considerable extent, the. evidence confirms Danlcwart Rustow's (1970) broad proposition that die causes of the origins of democracy are likely to be different from those that account for the stability of democracy (see also O'Donnell and Schmitter 1986: 65-66). Most notably, this proposition is supported by the finding that even if economic development is not a determinant of democratic transitions, the part of Seymour Lipset's (1959, 1960) old hypothesis that states that there is an association between the level of economic development and the stability of democracy does hold (Przeworski and Limongi 1997).20
But Rustow's proposition should not be pushed too far. Indeed, another old hypothesis that has received empirical support concerns the argument that
18 See also Valenzuela (1992: 73-7B), Linz and Stepan (1996: Ch. 4), and Munck and Luff . (1997).
19 Concerning Linz's (1994) hypotheses thai parliamentary democracies are more stable than presidential democracies, some tests indicate strong support for the argument that parliamentary forms of government better promote democratic stability (Linz and Valenzuela 1994a, 1994b, Stepun and Skach 1993, Przeworski, Alvarez, Cheibub and Limongi 1996), but others purport to show equally strong support for llie argument that presidential forms of government also promote democratic stability (Shugart and Carey 1992, Mainwaring 1993, Mainwaring and Shugart 1997a, 1997b, Power and Gasiorowski 1997). As various authors have stated, more plausible hypotheses would have to focus on variations within the broad choice between parliamentary and presidential forms of government, as well as consider the link between the power of presidents and the other institutional features such as the fragmentation of the party system and parly discipline (Shugart and Carey 1992, Mainwaring 1993, Mainwaring and Shugart 1997a, 1997b). It is unclear, however, whether such hypotheses would refer to the likelihood of die survival of democracy as opposed to the variable workings of stable democracies.
20 See also Przeworski, Alvarez, Cheibub and Limongi (2000: Ch. 2), Diamond (1992), Ged-des (1999), and Mainwaring (2000). For a theory and empirical test that shows why economic development can have a different impact Dn democratic transitions and democratic breakdowns, see Gould and Mangio (2007). However, for a skeptical view of the argument that economic development accounts for democratic stability, see Robinson (2006: 519-24).
60
61
democratic stability is less likely in plural societies or multinational states (Dahl 1971: 108-11, Powell 1982: 40-53, Diamond, Linz .and Lipset 1995: 42-43), even if, as Arend Lijphart (1977, 1984) stresses, this negative factor is mediated and potentially ameliorated by elite choices and power sharing arrangements (see also Dahl 1989: 254-60, Linz 1997: 414-14).21 Thus, what might be labeled as the national question seems to affect, in broadly the same manner, the prospects of democratic transition and democratic stability.
Another finding is that, much as there are multiple paths to democracy, so too are there multiple equilibria that can sustain democracy. This basic thesis is best established in research on the orientation of class actors in more equal and less equal countries. In more equal countries, as research on post-World War II Western Europe shows, a class compromise underpins the stability of democracy (Przevvorski 1985, Boix 2003). In this scenario, democratic stability was premised on a political exchange, whereby the moderation of the demands of labor and the left—a key goal of elites—is exchanged for redis-tributive policies, which is a core demand of mass actors. Both elites and masses, thus, have an incentive to accept democracy. In less equal countries, in contrast, a class compromise does not represent an equilibrium. As evidence from Latin America during the 1950s to 1970s shows, the redistributive consequences of democracy threatened elite interests and thus weakened the commitment of elites to democracy (O'Donnell 1973, 1999: Ch. 1). Thus, democratic stability in less equal countries rests on a different basis: the breaking, rather than the establishment, of any link between democracy and redistribution.
The stability of the democracies that emerged in less equal countries in the post-1974 period can then be related to two sets of factors. The potential destabilization of democracy due to the polarization of politics has been reduced due to the weakening of popular sector actors and labor as a result of recent experiences with authoritarian rule (Drake 1996, Munck 1998: Ch. 7) and the conscious lowering of expectations and self-restraint, especially among the left, which is a result of a learning process begun in the context of repressive, authoritarian regimes (McCoy 1999, Mainwaring 2000). Yet, more broadly, democratic stability is also the result of the widespread adoption of neoliberal policies in the 1980s and 1990s. Put in different words, because democracy in these countries is currently not associated with redistribution, business elites, who previously felt threatened by democracy and
21 One institutional proposal that has been the focus of much discussion is federalism. Some authors argue dial federalism is a particularly apt institutional choice Tor multinational societies (Stepan 2004). However, others show that at least under certain circumstonces edino-federalism can be destabilizing and lead to the breakup of the state (Hale 2004).
frequently sought to undermine democracy, have come to accept democracy (Payne and Bartell 1995, Huber and Stephens 1999: 775-80)."
Finally, research has also shown that democratic stability is influenced by international factors. Much as transitions to democracy, the stability of democracy is aided when neighboring countries, a country's region and the global context are more democratic (Gleditsch and Ward 2006, Mainwaring and Perez-Linan 2007). In addition, joining international organizations with high membership of democratic stales increases the probability that democracies will endure (Pevehouse 2005: Ch. 6).
Challenges
The accomplishments of this literature notwithstanding, scholars of democratic stability face a series of challenges that are quite similar to those discussed in the context of research on democratic transitions. With regard to the manner in which democratic stability, the dependent variable in this research, is measured, the problem and the solution are largely the same as in the literature on democratic transitions. Indeed, though discussions of the erosion, in contrast to the breakdown, of democracy introduce nuance in the discussion of democratic stability, there is still a need for explicit criteria for distinguishing changes of regime, that is, from a democratic to an authoritarian regime, from changes within democracy. This challenge, it bears noting, is simply the flip side of the challenge of measuring democratic transitions. Thus, it does not constitute a new challenge in the context of the broader study of democracy.
A second challenge concerns the pressing need for integration of causal theories. Some scholars have proposed causal factors that are structural in nature and focus primarily on economic aspects (Lipset 1959, I960,' O'Donnell 1973, Przeworski and Limongi 1997), Others have emphasized the explanatory significance of a range of institutional arrangements (Shugart and Carey 1992). And yet others advance theories that stress the importance of choice (Linz 1978) and strategic issues (Przeworski 1991: Ch. 1, O'Donnell 1992). As scholars have argued, each of these types of factors
22 This basic point can be fleshed out further. As Boix (2003: 3844) argues, democratic stability is affected by the types of assets owned by elites and, specifically, is positively correlated with factor mobility (see also Rogowski 1998 and Wood 2000). In this argument, then, die turn to ncDliheralism bos a positive effect on democratic stability in that it has increased the credibility of the threat of capital flight, which induces moderation among the poor and reduces the likelihood that electoral majorities would even propose redistributive policies, and hence makes democracy acceptable to elites. This argument has some distinct implications. First, it suggests that the moderation of labor and the left eon be induced by globalization as much as by harsh authoritarian rule. Second, and relatedly, it points to the possibility that the rare and happy coincidence of democracy and prosperity that was the trademark of post-World War II Europe may become rarer as neoliberalism takes root.
62
63
seems to have some explanatory' power; hence, a theory that ignored any of these types of factors would be incomplete. The problem, however, is that with a few exceptions (Lijphart 1977, Collier and Collier 1991), this literature has treated these variables in isolation even though processes affecting the stability of democracy unfold simultaneously at the various levels of analysis tapped by these variables. Thus, further progress on research on democratic stability is likely to require efforts to connect different types of explanatory factors and generate a more parsimonious and powerful theory that integrates the long list of explanatory factors highlighted by existing causal theory.23
A third challenge that scholars of democratic stability face concerns causal assessment. Statistical analysis has been more common in the study of democratic stability than that of democratic transition, Thus, the need to find ways to combine literatures using different methods is a prime concern. As with the literature on democratic transition, however, future research on democratic stability would still benefit from a multi-track approach. Specifically, qualitative researchers are likely to derive important payoffs from efforts to extend their comparative analyses beyond the current successes and failures to secure stable democracies. This might include comparisons with older positive experiences, especially the successful post-World War II record of Western Europe (Przeworski 1985, Maier 1987: Ch. 4).24 Moreover, it might address older cases of democratic breakdown, either by revisiting the well-researched cases of interwar Europe (Linz and Stepan 1978, Luebbert 1991, Rueschemeyer, Huber Stephens and Stephens 1992: Ch. 423) and Latin America in the 1960s and 1970s (O'Donnell 1973, Collier 1979, Collier and Collier 1991, Rueschemeyer, Huber Stephens and Stephens 1992: Ch. 5), or analyzing the failure to establish stable democracies in Africa and Asia in the early post-colonial period (Collier 1982, Young 19SS).26
The tasks faced by researchers who use statistical methods are much the same as those they face in the context of the study of democratic transition. Indeed, a key problem is that quantitative research on democratic stability has assessed only a limited number of independent variables. Virtually all studies
23 Weingast (1997) offers a noteworthy effort from u gume-theoretie perspective to the task of integration and synthesis. He interestingly frames the issue as a problem of credible commitment and stresses how democratic stability may be threatened by those who are in power. But his model fails to acknowledge thai democracy can also be threatened from above or below. Yet another line of research is implied by Schmitter's (1995) suggestion that scholars might focus on "partial regimes," a proposal which would focus on the challenge of integration by considering the interaction among various sites of politics rather than levels of analysis. See also Mahoney and Snyder (1999).
24 For a study that begins to address this comparison, see Alexander (2001).
25 See also Kurth (1979), Zimmermann (1988), Zimmermann and Saalfeld (1988), Linz (1992), Berg-Schlosser and De Meur (1994), and Ertmnn 1998).
26 For examples that revisit European and Latin American cases, see Berg-Schlosser and Mitchell {2000,2002), Bermeo (2003), and Capoccia (2005).
concerned with democratic stability still consider the favorite factor of modernization theorists: level of socio-economic modernization. To this factor, others have been added. These include other facets of economic and social life, such as economic performance (Gasiorowski 1995, Gasiorowski and Power 1998), inequality (Midlarsky 1997) and political culture (Inglehart 1997). And, in what are probably the most significant departures, the quantitative literature has begun to consider political institutions (Stepan and Skach 1993, Mainwaring 1993, Przeworski, Alvarez, Cheibub and Limongi 1996, Powerand Gasiorowski 1997) and the international environment (Pevehouse 2005, Gleditsch and Ward 2006, Mainwaring and Perez-Lifian 2007). In short, this literature is richer than the quantitative literature of the 1960s and 1970s.
However, statistical research on democratic stability has remained focused on easily measurable variables and has tended teMgnore the role of actors and choices stressed by process-oriented theorists.2 And, as a result, this research is unable to address the actor-centered theories that have been increasingly appreciated and theorized by qualitative researchers. Hence, the collection of data needed to assess the range of explanatory factors in the broader literature is an important task for quantitative scholars and one that would do much to foster a fruitful dialogue among quantitative and qualitative researchers about the causes of democratic stability.
3.2.3. The Quality of Democracy
The stability of democracy does not exhaust the post-transitions agenda of research. Rather, as numerous countries that democratized in the 1970s and 1980s faced no immediate threat of breakdown, scholars of democracy ■■ gradually began to suggest that other issues deserve attention. Essentially, even though more and more countries had made transitions to democracy and even though more and more countries remained democratic, these scholars sensed that they differed in quite fundamental ways and that these differences were not being captured by research on democratic transition and democratic stability. Thus, the need for a new agenda, on the quality of democracy, was recognized."'8
This new agenda is at an early stage in its development compared with the well-established agendas of democratic transition and democratic stability. Thus, there is not much in terms of research and findings to report. Indeed,
27 A few attempts aside (see, for example, Berg-Schlosser and De Meur 1994: 270-74), most quantitative researchers have proceeded as though it were unfeasible to collect data on process-oriented factors (Gasiorowski and Power 1998: 742, 745).
28 A more extensive discussion of the points that follow is provided in Munck (2004: 450-56, 2007).
64
65
one of the most pressing challenges faced by scholars concerned with this line of thinking concerns the delimitation of the subject matter of research. This step in the research process is critical, in diat the initial definition of the agenda charts out the boundaries of subsequent research, and demanding, in that it requires making complex conceptual choices in light of both theory and empirical information. Yet, some important clues regarding how to define this agenda are emerging and the steps needed to advance this agenda are also becoming clearer.
A key point of departure in the definition of a research agenda on the quality of democracy is that it addresses matters that go beyond a Schumpete-rian conception of democracy. In other words, it differs from research on democratic transitions and stability, which focuses on the concept of electoral democracy, in that it seeks to address aspects of democracy that extend beyond the constitution of government and the question whether rulers gain access to office tiirough free and fair elections. On tiiis point, there is broad agreement: few dispute the need to broaden research so as to encompass more than electoral democracy. Moreover, even though differences remain concerning how far beyond electoral democracy this new agenda should go, current theory offers a basis for making such choices.
The work of Robert Dahl (1989) in particular offers a theoretical foundation for understanding the concept of democracy as involving governments constituted through free and fair elections, but also a process of public decision-making and the implementation of binding decisions that reflects the principle that voter preferences are weighed equally. Indeed, Dahl's concept of democracy, though procedural, is much broader uian usually assumed and provides a dieoretical basis for a research agenda that goes well beyond Schumpeterian-rooted agendas. And the question it gives rise to—why do the actions of states reflect the preferences of voters more or less equally?—is of utmost current political relevance. Thus, even as various conceptual questions are addressed, it is probably advisable that scholars start focusing on steps needed to empirically address this new question.
To this end, one key challenge concerns the production of data. Indeed, though a learning process regarding the measurement of electoral democracy is beginning to bear fruits, students of democracy have still to develop the measurement metiiodologies and measures needed to advance this agenda. Efforts to measure different aspects of democratic governance, including corruption, do offer some leads (Munck 2005). Thus, this work does not have to start from scratch. But the work to develop adequate measures to address this question has barely begun.
The other key challenge is the development of causa! theory. Here again there are several useful leads in various literatures, and particularly those on corruption (Rose-Ackerman 1999, Heidenheimer and Johnston 2002, Johnston 2005) and clientelisrn (Eisenstadt and Lemarchand 1981, Piattoni,
2001. ICitschell and Wilkinson 2007). Moreover the application of the principal-agent framework to the problem of democratic representation and accountability has yielded some valuable insights (Przeworski 2003, Shugart, Moreno and Crisp 2003). Yet some key issues remain to be tackled and hence further theorizing is called for. Most urgently, the interplay between politicians' preferences and their capacity to make and implement policies is a fundamental and complex issue that remains to be adequately treated.
But it is also important to note that this agenda is not only broader than the previous agendas on democratic transitions and stability but also subsumes these agendas. Thus, theorizing should build on and incorporate the explanatory factors that research on democratic transitions and stability has shown to be relevant. And theorizing should distinguish between questions of transition and stability, or change and order. Indeed, the goal of tiiis new agenda should be to build a theory of democratization and democracy, much as the literature on democratic transitions and stability has done, but to anchor this theory in a broader concept of democracy, one that goes beyond democracy's electoral dimension.
3.3. Conclusion
The literature discussed in this chapter makes many valuable contributions to the study of politics around the world. It has addressed many normatively pressing problems and has produced many important findings. Even though it has not always focused on clearly articulated questions, as shown, it is possible to articulate its three core agendas in analytically tractable ways. Yet future progress in the field of democracy studies hinges on analysts' ability to face up to diree closely interrelated challenges.
One challenge is the need to better measure the outcomes of interest. More pointedly, more disaggregated and more nuanced measures of dependent variables are needed. A second challenge is the development and especially the integration of causal theories. With regard to the study of the quality of democracy, developing theory is a key concern. But the study of democratic transitions and stability faces a different problem: the unwieldy proliferation of explanatory factors. Hence, the need for greater theoretical integration was stressed and Uiree suggestions were offered: to rely on the lines along which theoretical debates have already been organized, to continue the search for theories that serve as unifying principles, and to recognize how analysts might integrate research findings regarding conceptually connected parts of a broad question.
Turning to the third challenge, concerning causal assessment, the ideal to be pursued in this field of studies is a multi-method approach that considers the trade-offs associated with small-N and large-N methods and taps into the
66
67
strengths of small-N methods (the generation of rich data, the sensitivity to the unfolding of processes over time, the focus on causal mechanisms) and of large-N methods (the emphasis on systematic cross-case and over-time comparison, the concern with generalizability, the formulation of precise estimates of causal effect and statistical significance). There are good reasons, however, why such a multi-method approach is hard to use in practice. Thus, the need for large-N data sets on key, processual variables, and for statistical analysis based on stronger research designs, was discussed. And some suggestions concerning small-N research projects that are most likely to yield important benefits were presented. Finally, the hope that a multi-track approach would give way to a genuine multi-method approach was expressed:
In sum, the field of democracy has made significant strides but still faces important challenges. In this sense, it constitutes an exciting research agenda. Researchers on democracy have opened up and continue to open up new substantive agendas, and have generated some important findings. Moreover, the issues addressed by this field of study put it in dialogue with some of the main debates about theory and methods in comparative politics. Students of democracy focus consistently on core issues of modern politics, the conflict over how the power of the state is accessed and used. In turn, the study of democracy is a site of important methodological innovations and a substantive field where a range of methodological issues have come into sharp focus. In short, the study of democracy is a vibrant research program.
4. Determinants of Democratization: Taking Stock of the Large-A/ Evidence
Jan Teorell and Axel Hadenius
4.1. Introduction
Since the third wave of democracy peaked some 10 years ago, large-/; studies of the determinants of democratization have expanded across space and time, covering more countries and longer time periods. In tenns of the theories tested and variables employed, however, most analyses have been highly specialized, focussing on the effects of but one or a few major explanatory factors. A large number of studies have assessed the effect on democracy of economic development and socioeconomic modernization (Burkhart & Lewis-Beck 1994; Helliwell 1994; Londregan and Poole 1996; Barro 1999; Przeworski et al. 2000; Boix & Stokes 2003; Epstein et al. 2006), whereas others have largely focused on the impact of economic crises (Gasiorowski 1995; Bernard et al. 2001, 2003), resource wealth (Ross 2001), colonial heritage (Bernard et al. 2004), or international factors such as globalization (Li and Reuveny 2003; Rudra 2005) and diffusion effects (O'Loughlin et al. 1998; Starr and Lindborg 2003; Brinks & Coppedge 2006).
Without denying the merits of specialization, we shall argue that this large-;; literature has serious limitations. To begin with, the results pertaining, to each determinant of democratization may be incorrect if the assessment is not performed in the context of all relevant controls. In other words, specialization may lead to erroneous conclusions even with respect to the one or few explanatory variables under study.
Second, and equally important, these previous studies do not address the question of how far all hypothesized determinants together can take us in explaining movements to and from democracy. The latest most comprehensive large-;; study even reached the conclusion diat, whereas democratic survival "is quite easily predictable", transitions to democracy appear to be explained by chance factors (Przeworski et al. 2000, 137). If that really proves to be the case, it would lend support to the anti-structural, actor-oriented, "no preconditions"-approach to democratization proffered in particular by Rustow (1970) and O'Donnell and Schmitter (1986), an approach which played a key part in the "transition paradigm" recently proclaimed dead (Carothers 2002). Apart from the distinction between transitions toward
68
69