COMPARATIVE
POLITICS
DANIELE CARAMANI
xl
WORLD DATA
"2 -a '
3 I :
aj a.
		o		**
		v£>	So	En
		eo	CN	ETJ
		in		in
				in
				
			in	CN
			Ü	Ln
				ses
			F*	vS
		r>-		ON
<N			in	CN
				O
-,		t--	so	
		t	00	o>
Ö				
		a\	„	
1	r--	\C-	Mí	VC
		On		O
Ö	r?	f-		
				c
				o
				Jg
			1	CL o
			o	Cl
	c3	§	"o	emal
1	J?	c	c	lg
c c	i	o	a ? Cu	
	3			1
1		q	ía -O	
Dß C O	i	popula	force f tion agi	force f i-64)
	I	ti	3 J	3 —
Popi	Life	Urb!	.O O, <tí o -I cl	2 SP
I i.
c — 'S
S9f
Introduction to comparative politics
Daniele Caramani
Chapter contents
Introduction 2
What is comparative politics? 2
The substance of comparative politics 4
The method of comparative politics 9
Conclusion 14
Reader's guide
Comparative politics is one of the three main disciplines in political science, alongside political theory and international relations. It deals with internal political structures (institutions like parliaments and executives), individual and collective actors (voters, parties, social movements, interest groups), and processes (policy-making, communication and socialization processes, and political cultures). The main goal of the discipline is empirical: describe, explain, and predict similarities and differences across political systems, be they countries, regions, or supra-national systems (such as empires or the European Union). This can be done through the intensive analysis of few cases (even one case) or large-scale extensive analyses of many cases, and can be either synchronic (based on data collected at only one time point and not accounting for change over time) or diachronic (including a temporal dimension). Comparative politics uses both quantitative and qualitative data. Increasingly, the analysis of domestic politics is challenged by the growing geographical scope and interdependence between countries through globalization. This brings comparative politics and international relations closer.
2
DANIELE CARAMANI
INTRODUCTION TO COMPARATIVE POLITICS
3
Introduction
This book is about politics. It is a book about the most important dimensions of political life, not about one specific aspect (such as elections or policies). Furthermore, it is a comparative book, meaning that we look at a variety of countries from all over the world. It is not a book about politics in one place only. Also, it is not only about politics today, but rather about how politics changed over time, beginning with the transition to mass democracy in the nineteenth century. In sum, it is a book about the long-term comparative study of politics.
But what, precisely, is politics? Politics is the human activity of making public authoritative decisions. They are public because, in principle, they may concern every aspect of a society's life. Political decisions can apply to everyone who is part of a given citizenship and/or living in a specific territory (a state) and to every area (religion, environment, economy, and so forth). They are authoritative because the government that makes such decisions is invested with the (more or less legitimate) power to make them binding, meaning that they are supported by the possibility to sanction individuals who do not comply with them. 'Authorities' have the authority—as it were—to force individuals to comply through coercive means. Politics is thus the exercise of the power of making such decisions. However, politics is also the activity of acquiring (and maintaining) this power. It is therefore both the conflict or competition for power, and its use. Who makes political decisions? How did they acquire the power of making them? Where does the authority to make such decisions come from? What decisions have been taken, why, and how do they affect the life of societies? These are the questions that comparative politics seeks to answer.
It goes without saying that these are important questions. Which decisions are made concerns our everyday life. The decision to increase taxation is a political decision. So are the decisions to cut welfare benefits such as malernity leave, introduce military conscription or carry out military intervention in a foreign country, and invest in nuclear power as a source of energy. But also how decisions are made is important. The way in which public and authoritative decisions are made varies a great deal. In democracies we, as citizens, are directly involved through elections or referendums. If we are unhappy with them we can protest through demonstrations, petitions, or letters, or vote differently at the next election. In other types of government, individuals are excluded (as in authoritarian regimes). And, finally, who makes or influences decisions also counts. Many decisions on the maintenance of generous pension systems today are supported by elderly cohorts in disagreement with younger ones who pay for diem. Or, as another example, take the decision to introduce high taxation for polluting industries. Such a decision is heavily influenced by lobbies and pressure
BOX 1.1 Definition of comparative politics'
Comparative politics is one of the three main subfields of political science (alongside political theory and international relations) focusing on internal political structures, actors, and processes, and analysing them empirically by describing, explaining, and predicting their variety (similarities and differences) across political systems (and over time)—be they national political systems, regional, municipal, or even supra-national systems.
groups and by environmental activists. Configurations of power relationships can be very different, but all point to the basic fact that political decisions are made by individuals or groups who acquired that power against others through cither peaceful/democratic or violent means.
KEY POINTS
Politics is the human activity of making public and authoritative decisions. It is the activity of acquiring the power of making such decisions and of exercising this power. It is the conflict or competition for power and its use.
Who decides what, and how, is important for the life of societies.
What is comparative politics?
A science of politics
Even though the questions addressed above are very broad, they do not cover the whole spectrum of political science. Comparative politics is one of the three main subfields in political science, together with political theory and international relations.1
Whereas political theory deals with normative and theoretical questions (about equality, democracy, justice, etc.), comparative politics deals with empirical questions. The concern of comparative politics is not primarily whether participation is good or bad, but rather investigates which forms of participation people choose to use, why young people use more unconventional forms than older age groups, and if there are differences in how much groups participate, Even though comparative political scientists are also concerned with normative questions, the discipline as such is empirical and value-neutral.
On the other hand, whereas international relations deals with interactions between political systems (balance of power, war, trade), comparative politics deals with interactions within political systems. Comparative
litics does not analyse wars between nations, but rather investigates which party is in government and why it has decided in favour of military intervention, what kind of electoral constituency has supported this party, how strong the influence of the arms industry has been, and so on. As a subject matter, it is concerned with power relationships between individuals, groups and organizations, classes, and institutions within political systems. Comparative politics does not ignore external influences on internal structures, but its ultimate concern is power configurations within systems.
As subsequent chapters clarify, the distinction between these three disciplines is not so neat. Many argue that, because of globalization and increasing interdependence and diffusion processes between countries, comparative politics and international relations converge towards one single discipline. Indeed, the brightest scholars bridge the two fields. What is important for the moment is to understand that comparative politics is a discipline that deals with the very essence of politics where sovereignty resides—i.e. in the state: questions of power between groups, the institutional organization of political systems, and authoritative decisions that affect the whole of a community. For this reason, over centuries of political thought the state has been at the very heart of political science. Scholars like Aristotle, Machiavelli, and Montesquieu—and many others—were interested in the question 'how does politics work?'
Despite being a vast and variegated discipline, comparative politics constitutes a subdiscipline of political science in its own right and, as Peter Hall has asserted, '[n]o respectable department of political science would be without scholars of comparative politics' (Hall 2004:1),
Types of comparative politics
The term 'comparative polities' originates from the way in which the empirical investigation of the question 'how does politics work?' is carried out. Comparative politics includes three traditions (van Biezen and Caramani 2006).
1 The first tradition is oriented towards the study of single countries. This reflects the understanding of comparative politics in its formative years in the US, where it mainly meant the study of political systems outside the US, often in isolation from one another and invoking little comparison. Today many courses on comparative politics still include 'German polities', 'Spanish polities', and so on, and many textbooks are structured in 'country chapters'- As discussed in Chapter 3, case studies have a useful purpose, but only when they are put in comparative perspective and generate hypotheses to be tested in analytical studies involving more than one case, such as implicit comparisons, the analysis of deviant cases, and proving grounds for new techniques.
BOX 1.2 Important works in comparative politics: Aristotle
Aristotle (350 BC)f Ta Politika (Politics)
The typologies of political systems presented in this work are based on a data compilation of the constitutions and practices in 158 Creek city-states by Aristotle's students. Unfortunately, this collection is now lost (with the exception of The Constitution of Athens). This work represents the oldest attempt on record of a comparative empirical data collection and analysis of political institutions. Aristotle distinguished three true forms of government; those ruled by one person (kingship), by few persons (aristocracy), and by all citizens (constitutional government), of which the corrupt forms are tyranny, oligarchy, and democracy.
2 The second tradition is methodological and is concerned with establishing rules and standards of comparative analysis. This tradition addresses the question of how comparative analyses should be carried out in order to enhance their potential for the descriptive cumulation of comparable information, causal explanations and associations between key variables, and prediction. This strand is concerned with rigorous conceptual, logical, and statistical techniques of analysis, also involving issues of measurement and case selection.
3 The third tradition of comparative politics is analytical in that it combines empirical substance and method. The body of literature in this tradition is primarily concerned with the identification and explanation of differences and similarities between countries and their institutions, actors, and processes through systematic comparison. Its principal goal is explanatory. It aims to go beyond merely ideographic descriptions and aspires to identify law-like explanations. Through comparison, researchers test (i.e. verify and falsify) whether or not associations and causal relationships between variables hold true empirically across a number of cases. It can be based on large-N' or 'small-NT research designs (N indicates the number of cases considered) with either mostly similar or different cases. It can use either qualitative or quantitative data, or logical or statistical techniques, for testing the empirical validity of hypotheses, lint ultimately this tradition aims at causal explanation.
This book takes the latter approach.
Thus, like all scientific disciplines, comparative politics is a combination of substance (the study of political institutions, actors, and processes) and method (identifying and explaining differences and similarities following established rules and standards of analysis). Like all sciences, comparative politics aims to say something
DANIELE CARAMANI
general about the world, i.e. formulate generalizations beyond one or few cases.
What does comparative politics do in practice?
1 To compare means that similarities and differences are described. Comparative politics describes the world and, building on these descriptions, establishes classifications and typologies. For example, we classify different types of electoral systems.
2 Similarities and differences are explained. Why did social revolutions take place in France and Russia but not in Germany and Japan? Why is there no socialist party in the US? Why is electoral turnout in the US and Switzerland so much lower than in most other democracies? As in all scientific disciplines, we formulate hypotheses to explain these differences and use empirical data to test them—to check whether or not hypotheses hold true in reality. It is through this method that causality can be inferred, generalizations produced, and theories developed and improved.
3 Comparative politics aims at formulating predictions. If we know that PR electoral systems favour the proliferation of parties in the legislature, could we have predicted that the change of electoral law in New Zealand in 1996 from first past the post to PR would lead to a more fragmented party system?
Why is comparative polities' called 'comparative polities'?
Comparative politics as a label stresses the analytical, scientific, and 'quasi-experimental' character of the discipline. It was in the 1950s-60s that the awareness of the need to carry out systematic comparisons for more robust theories increased. The comparative' label before 'politics' was added to make a methodological point in a discipline that was not yet fully aware of the importance of explicit comparison. However, single-case studies can be comparative in an implicit way, like Tocqueville's Democracy in America (1835). As John Stuart Mill noted in his review of the book in 1840, Tocqueville contrasted US specificities with France in a quasi-experimental way. Similarly, books on single countries in the 1960s and early 1970s—on Belgium, Italy, Norway, Spain, Switzerland— not only showed that 'politics works differently over here', but also included systematic, if hidden, comparison with the better-known cases of the US and Britain,
In practice the label comparative' was needed as a battle horse. In an established discipline this label could and should be dropped. Today it goes without saying that the analysis of political phenomena is comparative, i.e. entails more than one case. Therefore we should conclude that—since comparative politics covers all aspects of domestic politics—the discipline of comparative politics becomes 'synonymous with the scientific study of politics' (Schmitter 1993:171). All the dimensions of the political system can be compared, so all is potentially comparative politics. As Mair noted, '[ijn terms of its substantive
BOX 1.3 Important works in comparative politics: Machiavelli
Niccolo Machiavelli (written 1513, published posthumously 1532)» (/ Principe (The Prince, Florence: Bernardo di Ciunta)
This book was novel in its time because it told how principalities and republics are governed most successfully from a realist, or empirical, perspective and not how they should be governed in an ideal world. Machiavelli makes his argument through examples taken from real-world observations compared with one another. In The Prince he compares mainly different types of principalities (hereditary, new, mixed, and ecclesiastic), whereas in The Discourses on Livy {Discorsi Sopra la Prima Deco di Tito Law) his comparison between princely and republican government is more systematic.
concerns the fields of comparative politics seem hardly separable from those of political science tout court, in that any focus of inquiry can be approached either comparatively (using cross-national data) or not (using data from just one country)' (Mair 1996; 311). The generality of the scope of coverage of comparative politics leads us now to talk about its substance in more depth.
KEY POINTS
Comparative politics is one of the three main subfields of political science, alongside international relations and political theory.
Comparative politics is an empirical science that studies chiefly domestic politics.
The goals of comparative politics are: to describe differences and similarities between political systems and their features; to explain these differences; to predict which factors may cause specific outcomes.
The substance
of comparative politics
What is compared?
Tire classical cases of comparative politics are national political systems. These are (still) the most important political units in the contemporary world. However, national systems are not the only cases that comparative politics analyses.
1 First, non-national political systems can be compared: sub-national regional political systems (state level in the US or the German Länder) or supra-national
INTRODUCTION TO COMPARATIVE POLITICS
units such as (1) regions (Western Europe, Central-Eastern Europe, North America, Latin America, and so on), (2) empires (Ottoman, Habsburg, Russian, Chinese, Roman, etc.), and (3) supra-national organizations (European Union, NAFTA, etc.).
2 Types of political systems can be compared (e.g. a comparison between democratic and authoritarian regimes jn terms of, say, economic performance).
3 Comparative politics compares single elements of the political system rather than the whole system. Researchers compare the structure of parliaments of different countries or regional governments (or other institutions), the policies (e.g. welfare state or environmental policies), the finances of parties or trade unions, and the presence or absence of direct democracy institutions and electoral laws.
The various chapters of this book compare the most important features of national political systems. As can be seen in the Contents at the beginning of the volume, the varietv of topics is large, and comparative politics covers—in principle—all aspects of the political system. It has been argued that precisely because comparative politics encompasses 'everything' from a substantial point of view, it has no substantial specificity, but rather only a methodological one resting on comparison (Verba 1985; Keman 1993«). Yet, there is a substantial specificity which resides in the empirical analysis of internal structures, actors, and processes. It is also true that comparative politics is a broad discipline and, over the decades, it has been through phases in which it focused on particular aspects. This evolution is described in the next two sections.
From institutions to functions ...
Comparative politics before the Second World War was mainly concerned with the analysis of the state and its institutions. Institutions were defined in a narrow sense, overlapping with state powers (legislative, executive, judiciary), civil administration, and military bureaucracy. The type of analysis was formal, using as main 'data' constitutional texts and legal documents. The emphasis on the study of formal political institutions focused, naturally, on the geographical areas where they first developed, namely Western Europe and North America.
While the study of state institutions remains important, the reaction against what was perceived as the legalistic study of politics led to one of the major turns in the discipline between the 1930s and the 1960s—a period considered by some the 'Golden Age' of comparative politics (Dalton 1991). The behavioural revolution—imported from anthropology, biology, and sociology—shifted the substance of comparative politics away from institutions. Pioneers of comparative politics such as Gabriel A. Almond, founder of the Committee
©
BOX 1.4 Important works in comparative politics: Montesquieu
: Charles de Secondat, baron de Montesquieu :  (1748) De I'Esprit des Loix (On the Spirit of the I  Laws, Geneva: Barritfot et fils)
I   In this influential book, in which the idea of the separation of powers is presented systematically for the first time, Montesquieu distinguishes between republics, monar-:   chies, and despotic regimes. He describes comparatively ■   the working of each type of regime through historical examples. Furthermore, Montesquieu was really a pioneer of'political sociology' as, first, he analysed the influence :   of factors such as geography, location, and climate on a :   nation's culture and, indirectly, its social and political in-:   stitutions; and, second, did so by applying an innovative naturalistic method.
on Comparative Politics in 1954 (an organization of the American Social Science Research Council), started analysing other aspects of politics than formal institutions, and observing politics in practice rather than as defined in official texts.
What triggered this revolution? Primarily, more attention was devoted to 'new' cases, i.e. a rejection of the focus on the West and the developed world. Early comparativests like James Bryce, Charles Merriam, A. Lawrence Lowell, and Woodrow Wilson—as Philippe Schmitter calls them, 'Dead, White, European Men, but not Boring' (Schmitter 1993: 173)—assumed that the world would converge towards Western models of 'political order' (Fukuyama 2011, 2014). With this state of mind, it made sense to focus on major Western countries. However, the rise of communist regimes in Eastern Europe (and, later, in China and Central America) and the breakdown of democracy where fascist dictatorships came to power—and in some cases lasted until the 1970s, as in Portugal, Spain, and Latin America, and to some extent also in Greece (Stepan 1971; Linz 1978; O'Donncll and Schmitter 1986)—made it clear that other types of political order could exist and needed to be understood. After the Second World War, patterns of decolonization spurred analyses beyond Anglo-Saxon-style liberal democratic institutions. New patrimoni-alist regimes emerged in Africa and the Middle East, and populist ones in South America (Huntington 1968; O'Donncll 1973).
These divergent patterns could not be understood within the narrow categories of Western institutions. New categories and concepts were required, as was greater attention to other actors, such as revolutionary parties and clans under patrimonialistic leadership. The mobilization of the masses that took place in communist
DANIELE CARAMANI
INTRODUCTION TO COMPARATIVE POLITICS
and fascist regimes in Europe, as we]] as under populism in South America, turned attention away from institutions and directed it towards ideologies, belief systems, and communication. '1 his motivated comparativists to ask which were the favourable conditions for democratic stability, and thus to look into political cultures, social capital, and traditions of authority.2
Finally, the closer analysis of Europe also contributed to a shift away from the formal analysis of institutions. From the 1960s on, European, comparative political scientists started to question the supposed 'supremacy'—in terms of stability and efficiency—of Anglo-Saxon democracies based on majoritarian institutions and homogeneous cultures. Other types of democracies were not necessarily the unstable democracies of France, Germany, or Italy. The analyses of Norway by Stein Rokkan (1966), Austria by Gerhard Lehmbmch (1967), Switzerland by Jiirg Steiner (1974), Belgium by Val Lorwin (1966a, b), and the Netherlands by Hans Daalder (1966) and Arend Lijphart (196&*)—most published in Robert Dahl's influential volume Political Oppositions in Western Democracies (1966)—as well as Canada, South Africa, Lebanon, and India, all showed that politics worked differently than the Anglo-Saxon model.
Although ethnically, linguistically, and religiously divided, these societies were not only stable and peaceful, but also wealthy and 'socially just' (most remarkably in the case of the Scandinavian welfare states). On the one hand, these new cases showed that other types of democracies were viable. Besides the 'Westminster' type of majoritarian democracy, these authors stressed the 'consociational' type with patterns of compromise between elites (rather than competition), 'amicable agreements and accommodation'—in short, alternative
BOX 1.5 Important works in comparative politics: Tocqueville
Alexis Charles Henri Clerel de Tocqueville (1835) De'° Democratic en Amerique (On Democracy in America, Paris: C. Cosselin)
Although this book represents a 'case study'—an analysis of democracy in the United States—it is an example of comparison with an 'absent' case, i.e. France and, more generally, Europe. In his implicit comparison, Tocqueville analyses the uniqueness of conditions in American society and geography that were favourable to the development of modern democracy. Tocqueville follows Montesquieu in going beyond public institutions to include social and cultural aspects. He speaks of aristocratic and democratic societies when comparing France with the US. Tocqueville was also strongly influenced by Montesquieu's use of naturalistic methods.
practices of politics beyond format institutions. On the other hand, these new cases stimulated the investigation of the role of cleavages (overlapping vs cross-cutting), as in the case of welfare economies, as well as the role of elite collaboration in the political economy of small countries, which later led to important publications (see e.g. Esping-Andersen 1990; Katzenstein 1985).
What have been the consequences of the broadening of the geographical and historical scope?
First, it increased the variety of political systems. Second, it pointed to the role of agencies other than institutions, in particular parties and interest groups, civil society organizations, social movements, and media (Almond 1978: 14). Third, it introduced a new methodology based on empirical observation, large-scale comparisons, statistical techniques, and an extraordinary effort of quantitative data collection (see next section).3 Fourth, a new 'language^ namely systemic functional-ism, was imported in comparative politics. I he challenge presented by the extension of the scope of comparison was to elaborate a conceptual body able to encompass the diversity of cases. Concepts, indicators, and measurements that had been developed for a set of Western cases did not fit the new cases. It also soon became clear that 'Western concepts' had a different meaning in other parts of the world. What Sartori has called the 'travelling problem' (Sartori 1970: 1033) is closely related to the expansion of politics and appears when concepts and categories are applied to cases different from those around which they had originally been developed (see Table LI).
The emphasis on institutions and the state was dropped because of the need for more general and universal concepts. Since the behavioural revolution, we speak of political systems rather than states (Easton 1953,1965«, b). Concepts were redefined to cover non-Western settings, pre-modern societies, and non-state polities. Most of these categories were taken from the very abstract depiction of the social system by Talcott Parsons (1968). These more general categories could not be institutions that did not exist elsewhere, but their functional equivalents.
Functions dealing with the survival of systems were perceived as particularly important. From biology and cybernetics David Easton and Karl Deutsch (Deutsch 1966a, h) imported the idea of the system—ecological systems, body systems, and so on—and identified 'survival' as its most important function. Similarly, in the 1950s—still in the shadow of the dark memory of the breakdown of democratic systems between the two world wars through fascism and communism—the most important topic was to understand why some democracies survived while others collapsed. Almond and Verba's The Civic Culture (1963) is considered as a milestone precisely because it identified specific cultural conditions favourable or unfavourable to democratic stability.
and back to institutions
It soon also became clear, however, that the price to be paid for encompassing transcultural concepts was that of an excessive level of abstraction. This framework was not informative enough and too remote with regard to the concrete historical context of specific systems. In the 1970s* European comparative political scientists like Rokkan, Lehmbruch, and others (and even more so area specialists from Eastern Europe, Latin America, Africa, and Asia) had already noted that the ahistorical categories of systemic functionalism did not allow the understanding of concrete cases. The chapter by Lipset and Rokkan on 'Cleavage structures, party systems, and voter alignments' (Lipset and Rokkan 1967) is the emblematic piece of research that puts history and context back in the equation.
The counter-reaction to systemic functionalism sLarts precisely in 1967 and involves (1) a shift of substantial focus, (2) a narrowing of geographical scope, (3) a change of methodology, and (4) a theoretical turn devoting greater attention to the rationality and strategies of actors.
Bringing the state back in
The shift of substantial focus consists of a return to the state and its institutions (Skocpol 1985). In recent decades there has been a re-establishment of the centrality of institutions more broadly defined as sets of rules, procedures, and social norms. In the new-institutionalism theory (March and Olsen 1989; Hall and Taylor 1996; Thelcn 1999; Ostrom 2007; Pierson and Skocpol 2002; Przeworski 2004a) institutions are seen as the most important actors, with autonomy and being part of real politics. Institutions, furthermore, are seen as determining the opportunity structures and the limits within which individuals formulate preferences.4
Mid-range theories
The excessive abstraction of concepts in systemic functionalism was also countered by a return of attention to varying historical structures, cultural elements, and geographic location, in which the specific context plays a central role (Thelen and Steinmo 1992). Rather than general universalistic theories, mid-range theories stress the advantages of case studies or in-depth analyses of a few countries.
Some authors argue that the reawakening of attention to the state and its institutions is in fact a consequence of this narrowing of geographical scope (Mair 1996). The general language introduced by systemic functionalism—and which nearly discarded the state and its institutions—was needed to encompass a greater variety of political systems. Institutions have recently been re-appreciated because of a closer focus. Systemic functionalism did not forget institutions; simply, they were 'absorbed upward into the more abstract notions of role, structure and function' (Mair 1996: 317). A regionally more restricted perspective giving up global comparisons does not require the same level of abstraction of concepts. Therefore, the shift of substantial focus is a consequence of less ambitious theoretical constructions. The change of substantial focus has been favoured by the narrowing of the geographical focus.
Case-oriented analysis
This narrowing of scope also entailed a methodological change. The counter-reaction to large-scale comparisons came from the development of methods based on few cases ('small-N') (see Ragin 1987). They revitalize today a type of comparative investigation that had long been criticized because few cases did not allow the testing of the impact of large numbers of factors—the problem
Table 1.1 Comparative politics before and after the 'behavioural revolution'
Dimensions of analysis	Before	After
Unit	State	Political svstem
Subject matter	Regimes and their formal institutions	AH actors in the process of decision-making
Cases	Major democracies: US, Britain, France; analysis of democratic breakdown in Germany and Italy; authoritarianism in Spain and Latin America	Objective extension of cases (decolonization) and subjective extension with spread of discipline in various countries
Indicators/ variables	West-centric, qualitative categories, typologies	Abstract concepts (functional equivalents); empirical universals, quantitative variables
Method	Narrative accounts and juxtapositions between cases	Machine-readable data sets and statistics; quasi-experimental comparative method
Data	Constitutional and legal texts	Survev and aggregate data
Theory	Normative: institutional elitism and pluralism; no elaborate conceptualization	Empirical: structural tunctionalism, systems theory, neo-institutionalism, rational choice, cultural theories
INTRODUCTION TO COMPARATIVE POLITICS
that Lijphart (1971, 1975) named 'few cases, many variables! This difficulty made the analysis of rare social phenomena, such as revolutions, impossible with statistical techniques. Hence, the great importance of this 'new' comparative method. It provides the tool for analysing rigorously phenomena of which only few instances occur historically (see next section and Chapter 3 for more details).
Rational choice theory
At the end of the 1980s another turn took place in comparative politics, strengthening further the place of institutions. It was the turn given by the increasing influence of rational choice theory in comparative politics.
Whereas the behavioural revolution primarily imported models from sociology, the turn at the end of the 1980s was inspired by developments in economics. In addition, the rational choice turn does not revolve around a redefinition of the political, for it applies a more general theory of action that applies equally well to all types of human behaviour, be it in the economic market, the political system, the media sphere, or elsewhere {Munck2001;Tsebelis 1990).
This theory of action is based on the idea that actors (individuals, but also organizations such as political parties) are rational. They are able to order alternative options from most to least preferred and then, through their choice, seek the maximization of their preferences (utility). For example, voters are considered able to identify what their interest is and to distinguish the different alternatives that political parties offer in their programmes with regard to specific policies. Voters then maximize their utility by voting for the political party whose policy promises are closest to their interests. It is rational for political parties to offer programmes that appeal to a large segment of the electorate as this leads to the maximization of votes.
It is clear from these premises that the place for 'sociological' factors on which the behavioural revolution insisted—such as socio-economic status and cultural traits—assume a lower key in rational choice models. These models have been crucial to understanding the behaviour of a number of actors. In the field of party politics, examples include work by Downs (1957), Przeworski (Przeworksi and Sprague 1986), and Cox (1997). Other examples include the work of Popkin (1979) on peasants in Vietnam, Bates (1981) on markets in Africa, Fearon and Laitin (1996) on ethnicity, Przeworski (1991) on democratization, Gambetta (1993, 2005) on the Mafia and suicide missions, and Acemoglou and Robinson (2006) on the origins of political regimes.
Rational choice theory in political science owes a lot to the work of William Riker. He is the founder of the 'Rochester School' (Riker 1990; see also Amadae and Bueno de Mesquita 1999). Today, rational choice
theory comes in various forms and degrees of formalization. They range from 'hard' game-theoretical versions, in which the degree of mathematical formalization is very high, to 'softer' versions in which the basic assumptions are maintained but in which there is no formal theorizing. What is important to note is that the rational choice turn did not lead to a redefinition of comparative politics as a subject matter precisely because it does not offer a meta-theory that is specific to politics. The subject matter did not change under the impulse of rational choice theory. On the contrary, it has reinforced the pre-eminence of institutions in comparative politics. Rational choice institutionalism, in particular, sees institutions as constraints of actors' behaviour (Weingast 2002). An example of this approach is the concept of 'veto player' developed by Tsebelis (2002).
What is left?
As we have seen, there has been an almost cyclical process.5 However, comparative politics did not simply return to its starting point.
1. Despite the recent narrowing of scope and the tendency to concentrate on 'mid-range theories', the expansion that took place in the 1950s and 1960s left behind an extraordinary variety of topics. A glance at the Contents shows how many features of the political system are dealt with in comparative politics,
2. Also, the great contribution made by the systemic paradigm has not been lost. We continue to speak of a political system and use this descriptive tool to organize the various dimensions of domestic politics. In fact, the structure and coverage of the book mirrors the political system as described by David Easton (see Figure 1.1 and Box 1.6). Easton's work is a monumental theoretical construction of the structural-systemic paradigm, still unrivalled and probably the last and most important attempt to build a general empirical theory including all actors and processes of political systems.
I
POLITICAL SYSTEM
w£M BOX 1.6 Important works in comparative politics: Easton
David Easton (1953) The Political System: An Inquiry into the State of Political Science (New York: Alfred A. Knopf)
This volume is the first of a series of books by Easton on the political system. His work represents the most systematic and er.compassing effort on the 'theoretical side' of the behavioural revolution. Scholars like David Easton and Karl W. Deutsch imported the notion of system from other scientific disciplines (biology and cybernetics). This notion soon replaced the formal concept of state and enlarged the field of comparative politics to non-institutional actors. The framework developed by Easton and his colleagues, and its conceptual components (input, output, feedback Iood, black box, etc.), are common language today. Easton's work remains the last major attempt to develop a general empirical theory of politics.
Feedback Environment
Figure l.i The political system Source: Easton (1965).
Easton's concepts have marked the minds of political scientists, as well as those of the wider public. His attempt has been an extremely systematic one, with subsequent and cumulative contributors drafted towards one single goal. His concept of political system—as a set of structures (institutions and agencies) whose decision-making function is to reach the collective and authoritative allocation of values (output, i.e. public policies) receiving support as well as demands {inputs) from the domestic as well as the international environment which it shapes through outputs in the feedback loop—includes all aspects of what is described in this book, from communication to culture, socialization and behaviour, interest articulation through parties, movements or pressure groups, institutions in democratic and authoritarian regimes, decision-making and policies, as well as the interaction with other systems—addressed in the last section of this book. ■ The substantive scope has not ceased to grow and this trend has continued over the most recent decades. As discussed in Chapter 1, there has been a change in focus from 'input' processes to 'output' processes, namely public policies, policy-making, as well as the outcome and impact of policies. This is the reason why a specific section of this book is devoted to these topics. In particular, recent trends of 'what' is compared include industrial relations, trade, and economic policies (aspects stressed in Chapters 22 and 24); the reawakening of ethnic, religious, and nationalist movements, trends towards regionalization (aspects stressed in Chapters 11, 15, and 17); and the growing role of pressure groups (see Chapter 14).
New trends also include awareness of the interdependence between national systems (discussed below more extensively). Chapter 23 analyses the integration between member-states of the European Union, Chapter 24 addresses the blurring of national boundaries, and Chapter 25 shows how states influence others through democratic promotion and peacekeeping.
KEY POINTS
Comparative politics is not limited to the comparison of national political systems, but also includes other units such as sub-national and supra-national organization, single political actors, processes, and policies. With the widening of the number of 'cases' (new states or other regions), the need for more general concepts that could 'travel' beyond Western countries led to a focus on functions rather than institutions. In the last three decades, however, a reaction against overly abstract analysis led back to 'mid-range theories' limited in space and time. * As for the behavioural revolution, rational choice also aims at a general and unified theory of politics applicable in all times and places. This paradigm was imported into political science from economics and stresses the role of institutions in comparative politics. Comparative politics includes as a subject matter all features of political systems and, recently, has turned its attention towards the interaction between them, approaching international relations.
The method of comparative politics
Having discussed the 'what' of comparison, we turn now to the 'how' of comparison.
A variety of methods
Comparative politics does not rely on one specific method only, for four main reasons,
1. Depending on the number of cases included in the analysis (say, 150 or two countries only), the type of data the analysis deals with (quantitative electoral results or qualitative typologies of administrative systems), and the time period covered (the most recent census or longitudinal trends since the mid-nineteenth century), the methods employed are different. The research method depends on the research question. We formulate the research question; then we look for the most appropriate data
DANIELE CARAMANI
and methods to address it. The choice of cases very often depends on the research question. As explained in Chapter 3, comparative politics may analyse one single case (a case study). Research designs can be more or less intensive or extensive (depending on the balance between the number of cases and the number of features analysed); they can be synchronic or diachronic.
2. The dimensions of comparison can be diverse. It is wrong to suppose that comparative politics is always cross-sectional, i.e. that it involves a spatial comparison between countries, areas, or groups of political systems. In fact, spatial (cross-sectional) comparison is only one of the possible dimensions of comparison. A second dimension of comparison is the functional {cross-organizational or cross-process) comparison. Take, as an example, the comparison of the liberal and the nationalist ideologies in Europe. Or the comparison of policy-making of environmental and military policies in, say, the US. Or the comparison of leadership in social movements such as the civic rights movement, the feminist movement, the green movement, and the pacifist movement. The dimension of comparison here is not territorial. A third dimension is the longitudinal {cross-temporal) comparison. We can compare institutions, actors, and processes over time as, for example, in the comparison of party organizations in the nineteenth century (cadre parties), after the First World War (mass parties), after the Second World War (catch-all parties), and since the 1980s (cartel parties).
3. Units of analysis can be diverse. As we have seen earlier, 'what' is compared can be either whole political systems or single actors, institutions, processes, or trends.
4. Comparative research designs can focus on either similarities or differences. Sometimes we ask questions about similar outcomes, such as 'why did social revolutions take place in France, Russia and China?' (Skocpol 1979) or 'why did democracy resist attacks from anti-system forces in some countries and not in others?' (Capoccia 2005). To explain similar outcomes we look for common factors (something that is present in all the cases in which the outcome occurred—either a revolution or a democratic breakdown) in cases which are otherwise very different from each other. As we will see in Chapter 3, John Stuart Mill called this research design the Method of Agreement (Przeworksi and Teune 1970 called it the 'Most Different Systems Design'). However, sometimes we use the Method of Difference (or 'Most Similar Systems Design'), in which we ask questions about different outcomes, such as 'why did Britain democratize early and Prussia/Germany late?' (Moore 1966). To explain different outcomes
BOX 1.7 Important works in comparative politics: Lazarsfeld et al.
Paul F. Lazarsfeld, Bernard R. Berelson, and Hazel Caudet (1944) The People's Choice: How the Voter Makes up His Mind in a Presidential Campaign (New York: Columbia University Press)
This book is a marvellous example of the use of statistical methods, and multivariate analysis of elections, public opinion, socialization processes, and communication through large data sets and the employment of rudimentary computing techniques. It is an application of the positivist approach to politics and has paved the way for countless studies of the determinants of people's vote: the crucial questions of which groups of people {classes, professions, age cohorts, gender, and so on) tend to turn out more often and for which parties they tend to vote. A follow-up volume entitled Voting {1954) pursued this line of research. This book is an example of the 'empirical side' of the behavioural revolution.
r
we look at factors that vary (something that is either present or absent in the case in which the outcome either occurred or did not—democracy) among otherwise similar cases. We also often combine these two methods.
From cases to variables ...
Comparative politics prior to the behavioural revolution was typically a discipline that compared few cases. Today, we speak of 'small-N' research designs. As explained earlier, it was thought that the world would converge towards the Anglo-Saxon model of democracy and that, consequently, these were the cases that comparative political scientists should concentrate upon. Therefore the number of cases ('N') was limited to the US, Britain, France, and a few other cases such as Canada, sometimes Australia and New Zealand, and the 'failed' democracies of Germany or Italy.
The behavioural revolution involved the widening of cases. On the one hand, this involved a much larger effort of data collection. Large data sets were created with the help of the development of computer technology. On the other hand, this involved the need for comparability of indicators and, as it turned out, the most general 'language' was that of quantities. It is very difficult to establish whether or not civic culture, honour, patriotism, justice, etc., have the same meaning in different continents. However, it is possible to measure the number of televisions, internet connections, or mobile telephones in all countries of the world. Both factors—the
BOX 1.8 Important works in comparative politics: Downs
Anthony Downs (1957) An Economic Theory of Democracy (New York: Harper & Collins)
This is a small book (Downs's PhD dissertation) which had an enormous impact, showing the great potential of rational choice theory for the study of politics. It introduced economic models for the analysis of actors' behaviour as well as the deductive analytical rigour in comparative politics. Today, rational choice models are one of the dominant approaches in comparative politics. Although this approach had an impact in all fields of comparative politics, in the field of electoral studies it still remains one of the most important works, alongside that of Maurice Duverger and Giovanni Sartori, inspiring pioneering research such as that of Gary Cox on voting behaviour and the impact of electoral systems on politics.
increasing number of cases and the quantification of indicators—led to the development of sophisticated statistical techniques, which became the dominant method. Therefore, research designs based on a 'large-NT typically employ techniques such as multiple regression and factor analysis (or other statistical techniques) based on numerical coefficients which allow researchers to quantify the strength of the causal association between political phenomena.
This trend turned attention away from cases and shifted it towards variables. Intensive research designs became extensive: many cases and few variables. Ragin (1987) defines the large-N research design as 'variable-oriented', implying that, with many cases, we ultimately know very little about the context of the countries. Not only did concepts become increasingly abstract in the search for the most general, most comparable, and most equivalent concepts, but the analysis itself referred increasingly to abstract relationships between variables. We would know that higher literacy levels are associated with higher turnout rates, but wc would be ignorant about patterns in single countries.
... and back to cases
More recently there has been a return to small-NT and case-oriented research designs and, today, the comparative method is in fact equated with the qualitative techniques based on John Stuart Mill's Methods of Agreement and Difference and on the search for sufficient and necessary conditions. Theda Skocpol (1984), David Collier (1991), and, most prominently, Charles
INTRODUCTION TO COMPARATIVE POLITICS
Ragin with his ground-breaking The Comparative Method (Ragin 1987), showed that rigorous empirical tests could also be carried out when the number of cases is small (for an overview see Caramani 2009; Schneider and Wagemann 2012).
This methodological shift stresses the intrinsic advantages of the study of few cases. Case-oriented scholars stress that small-N comparisons allow in-depth analyses in which configurations or combinations of factors are privileged in explanations. Cases are seen as 'wholes' rather than being divided into isolated variables. Constellations of factors represent the explanation rather than the impact of each factor individually.
This is a reaction against the first writings on the comparative method, which advanced strong arguments precisely against configurative or combinatorial analyses in which a large number of potential explicative variables were listed (e.g. Przeworski and Teune 1970). On the contrary, the focus was on parsimonious explanatory designs, i.e. a few key variables whose impact should be tested on as many cases as possible. In two famous articles, Arend Lijphart (1971, 1975) suggested increasing the number of cases (e.g. by selecting several time points) and decreasing the number of variables by focusing on similar cases (thus reducing the number of factors that vary across them).
As we have seen, comparative methods developed in a period when social sciences were looking for a general language, i.e. theoretical and operative concepts that could be used without temporal or spatial limitations. Such a move implied 'replacing proper names with variables' (Przeworski and Teune 1970), defining concepts able to 'travel' (Sartori 1970), and using 'sets of universals' applicable to all political systems (Almond and Powell 1966; Lasswell 1968). In addition, because of the small number of cases for many research questions, a parsimonious use of variables was also invoked. This has led to 'a strong argument against... "configurative" or "contextual" analysis' (Lijphart 1971: 690) unable to give rise to generalizing statements. 'Ihirty years later, a large part of the recent debates around methods in the social sciences has focused on the opposite reaction, namely a swing away from the variable-oriented approach towards 'thick' research designs.
Critiques of case-oriented approaches denounce a return to the past. As John Goldthorpe notes, this represents a revival of holism against which Przeworski and Teune (1970) had directed their work. In addition, even if one concentrates on 'whole' cases, one still refers to selected features or attributes. Comparison can take place only when one compares cases' values of shared properties or attributes, i.e. variables (Goldthorpe 2000; see also Bartolini 1993). The accusation is that we are going back to holism. And, again, we see a cyclical pattern in the method of comparative politics just as we did for its subject matter.
12
DANIELE CARAMANI
BOX 1.9 Important works in comparative politics: Almond and Verba
Gabriel A. Almond and Sidney Verba (1963) The Civic Culture: Political Attitudes and Democracy in Five Nations (Princeton, NJ: Princeton University Press)
This book was the seminal attempt to make systematic use of individual-level data collected comparatively through survey techniques. It is a phenomenal effort of individual data collection and analysis, in the US, the UK, Germany, Italy, and Mexico, at the dawn of the computer age. Within the behavioural paradigm, it analyses the function of political culture in political systems and, in particular, the central role that the 'civic culture' plays in the survival of democratic political systems. This book opened the way to studies on values, trust, and social capital pursued most prominently by Ronald Inglehart and Robert Putnam.
From aggregate to individual data ...
For a long time, the only available data were those collected as official statistics. The term 'statistics' itself goes back to the seventeenth century and the German School of Statistics. Etymologically, the term means 'science of the state' and its purpose is, as it were, to analyse state matters. Statistics started developing during the formation of the modern mercantilist nation-states and flourished in the course of the nineteenth century when the great economic transformations (industrialization) and population movements (urbanization) strengthened the need for states to monitor increasingly complex societies.
The same period saw the development of the liberal nation-state, which, as discussed in Chapters 4 and 8, increased its intervention in the society and economy, which was accentuated with the welfare state. To act, states needed knowledge of the society and economy they were supposed to steer. Democratization also gave a big push towards the development of statistics as governments became accountable; they had to perform, which involved a systematic collection of information. To meet this need, i.e. to increase their 'cybernetic capacity' (Flora 1977:114), techniques for gathering information greatly improved.
Primarily, statistics were collected for practical reasons linked to the economic and military action of governments, The contents of national statistics relate directly to the activity of the state: security and finance (military and criminal statistics, and statistics relating to income and expense items, taxation, and natural resources). With the growth of welfare states, the transformation of the population and health issues are monitored very closely: birth rates, mortality, health, and mental
illness. As far as political statistics are concerned, they were usually included under juridical statistics. However, the presence of political statistics is less common than that of other categories, in particular electoral statistics which are linked to democratization and attempts to legitimize regimes (see Caramani 2000: 1005-15).
The landmarks of this development have been the organization of censuses—every five or ten years, depending on the country—and the establishment of the annual publication of statistical yearbooks. These often include statistics of neighbouring countries requiring a certain degree of standardization of information to allow for comparisons.
These data are called aggregate data because they are available at some territorial level: provinces, regions, countries. Typical aggregate data are election results. We never know how individuals vote because voting is secret. However, we have aggregates: the number of voters and the number of votes for parties and candidates in a constituency. Similarly, we often have data for unemployment rates, population density, and activity in a given sector (e.g. agriculture) for territorial units.
With the behavioural revolution the approach to data collection changed radically.
1. Scientific researchers became sceptical about official statistics which, especially in non-democratic states, may be subject to manipulation. Tins concerns data on elections and all aspects of civil rights, but also data on economic performance (unemployment or GDP). Therefore, the creation of large data sets by university researchers independent from politics is an important aspect of the behavioural revolution.
2. Official statistics do not include all variables of interest to researchers. On the one hand, official statistics do not include information on political actors. An example is data on political parties, their members, and their finances (very difficult to obtain and rarely reliable). On the other hand, official statistics do not include information on individuals' values, opinions, attitudes and beliefs, competence and trust in political institutions, and differences between elites and masses in political preferences. Through official statistics we would not know whether an individual has authoritarian attitudes or post-material values, and whether he or she is strongly religious. Census data are the closest type of data on this information but do not include political data. The behavioural revolution introduced surveys as a systematic instrument to collect individual data. As Chapter 17 shows, political culture cannot be analysed without this type of data, which can be found throughout the world in surveys such as the World Value Survey, Eurobarometers, European Social Survey, Latinobarometers, etc.
3. The collection of individual data involved much larger data sets as thousands of individuals are included in a survey. This amount of data can be dealt with only
through the computerization of the social sciences which began in the 1950s. The behavioural revolution involved the analysis of individual data through a new electronic computing technology. Certainly, in the past there had been examples of extraordinary data analysis without computers. Durkheim's Le Suicide (1897) is a breathtaking example of comparative multivariate analysis of a huge amount of data presented in tables and figures without the help of computers. Computerization put this type of analysis within the reach of all researchers, first through mainframe systems (usually in a university) and, in the late 1980s, through personal computers and statistical software designed for them. Today every undergraduate student has Excel, SPSS, R, Stata, or other packages on his or her laptop, 4. The year 1950 proved to be devastating for analysis with aggregate data. This was the year when William S. Robinson published his famous article about 'ecological fallacy' (Robinson 1950). This article undermined the assumption that correlations observed at the level of aggregated units could be inferred at the individual level. Problems of ecological inference arise in the attempt to infer conclusions reached at the level of territorial units down to the individual level. Put simply, what is true on an aggregated level is not necessarily true at the individual level. Robinson showed that through aggregate data (based on counties and states in the US) there was a strong correlation between race and literacy, but this correlation disappeared when individual data were used, The effect of this article was disruptive, the term 'ecological fallacy' became popular, and for a long time analyses based on ecological data were discredited.
BOX 1.10 Important works in comparative politics: Rokkan
Stein Rokkan (1970) Citizens, Elections, Parties (Oslo: Universitetsforlaget)
This book is a collection of previously published articles and chapters, complemented by unpublished bits and pieces, and conference papers by Stein Rokkan (who never v*rote an authored monograph but preferred to work his writings over and over again). Nonetheless, Rokkan's work provides the most systematic comparative picture of a huge amount of empirical material on similarities and differences between countries in their patterns of state formation, nation-building, democratization, and the structuring of party systems and electoral alignments. In the tradition of 'comparative historical sociology' (with Reinhart Bendix, Otto Hintze, and Barrington Moore, among others), his work encompasses centuries of political development and has inspired generations of scholars such as Theda Skocpol and Charles Tilly.
INTRODUCTION TO COMPARATIVE POLITICS
... and back to aggregate data
The reaction to this 'shock' began almost immediately, attempting to find solutions to 'ecological fallacy! Conferences and meetings led to collective publications (see Merritt and Rokkan 1966; Dogan and Rokkan 1969; Berglund and Thomsen 1990; King 1997; King et al. 2004).
Furthermore, the reaction to Robinson's blowtoecolog-ical data involved the creation of international networks for comparable 'hard data' worldwide. International data archives were set up. 1he most important ones today are the Inter^University Consortium for Political and Social Research (ICPSR) (at the University of Michigan), the Data Archive (at the University of Essex), the Mannheim Centre for European Social Research (MZES), and the Norwegian Data Archive (at the University of Bergen). Data archives developed in all countries are linked together in a global network (see the Online Resource Centre). Such efforts led to major publications of aggregate data collections with documentation, most notably the three editions of the World Handbook of Political and Social Indicators (Russett et al. 1964; Taylor and Hudson 1972; Taylor and Jodice 1983), but also other projects (see the 'Yale Political Data Program'; Deutsch et al. 1966). These publications arc updated today through the internet resources of the ICPSR.
International organizations such as the United Nations (UN), the World Trade Organization (WTO), the World Bank, the International Monetary Fund (IMF), the World Health Organization (WHO), the Organization for Security and Co-operation in Europe (OSCE), and so forth also contributed to the creation of large comparative data sets with aggregaLe data in their sectors of competence. The Online Resource Centre provides all the links to these data sets.
But perhaps the main reason for a 'recovery' of ecological data analysis resides in the intrinsic weaknesses of individual-level data. It is more difficult to build long time series with individual data. Only aggregate data that we can collect from the beginning of the nineteenth century allow us to understand topics that need a long-term perspective. This was particularly true during the 1960s and 1970s when modernization approaches were used to understand newly decolonized countries. Panels-surveys carried out with the same group of respondents over protracted periods of time—arc extremely costly {and, anyway, do not allow going 'back' in time). And the use of existing surveys for comparative purposes is not straightforward. Intelligence services, especially US ones, carried out a number of surveys in Europe after the Marshall Plan to investigate the public's attitudes, its favour of democratic values, and the potential of a communist menace or fascist return. However, these early studies are fragmented, with different questions asked and different groups or respondents.
13
DANIELE CARAMANI
INTRODUCTION TO COMPARATIVE POLITICS
A
^fa^   Box 1.11 Important works in
comparative politics: Esping-Andersen
i  Casta Esping-Andersen (1990) The Three Worlds of Welfare Capitalism (Cambridge: :  Polity Press)
:   This is the book that best illustrates the shift in comparative politics from input to output and public policies. It presents a typology and an explanation of what can be considered the ■   most encompassing of all public policies after the Second j   World War—the development of the welfare state as the :   latest stage in the construction of the modern nation-state :   and citizenship, where social rights complement political :   and civic rights {as distinguished by T H. Marshall}. This work is a prominent example among other large research programmes, namely on varieties of capitalism (e.g. Susan Strange's work), comparative political economy (e.g. Peter Hall), and welfare states (e.g. Peter Flora).
Therefore, aggregate data have not disappeared and sometimes provide more solid bases than individual-level data for international long-term comparisons.
KEY POINTS
Comparative politics employs statistical techniques when research designs include many cases and quantitative indicators (variable-oriented large-N studies), or 'comparative methods' when research designs include few cases and qualitative indicators (case-oriented small-N studies). Case studies can also be earned out in a comparative perspective. ■ The dimensions of comparison are multiple: spatial, temporal, and functional.
Research designs aim either at selecting similar cases and explaining their different outcomes (Most Similar Systems Designs, the 'Method of Difference'), or at selecting different cases and explain similar outcomes (Most Different Systems Designs, the "Method of Agreement').
-  Comparative politics relies on different types of aggregate and individual data.
Conclusion
The variety of comparative politics
The great variety of approaches, methods, and data of comparative politics matches the great variety of the world's societies, economies, cultures, and political
systems. At the end of this book we have inserted a number of 'Comparative tables' on various indicators. We have also inserted a number of 'World trends' figures which show how societies and political systems have changed over the decades. Readers will also find 'Country profiles', small files on political systems around the world.
The book rests on the principle that everything is comparable. Large-scale comparisons through space and time in this book are based on the idea that there are no limits to comparison. Everything—i.e. any case in the world at any point in time—is, in principle, comparable. Analytical comparison never compares cases as such (say, countries) but rather properties (e.g. turnout levels) and their values for each case—whether turnout levels are high or low according to countries. Obviously, turnout applies only where there are democratic elections, so the level of generality and the spatial and temporal scope of the comparison of turnout is limited.
The nineteenth century witnessed what is probably the greatest change in the political organization of human societies with the rise of modern nation-states and democracies. There was no previous experience of mass democracy based on principles of fundamental equality between individuals, civic liberties, political rights, and open participation to the political process and to social welfare. The scope of this change was matched only by the Industrial Revolution during the same period. This is a unique period in our history and we should be aware of its exceptional character but also of its shortness. Therefore, it is crucial to cover the development of the nation-state and mass democracy over nearly 200 years.
This Introduction has stressed the great variety of what is a huge field of study covering all aspects of domestic politics, with many areas of specialization and subdisci-plines which are reflected in the chapters of Comparative Politics. The great variety—and the consequent specialization of the field—is the main reason why it is difficult to single out the most important books (sec the various boxes scattered through this Introduction). Each subdis-cipline has its 'classic' work: in the field of coalition formation, in that of the study of electoral systems, in that of the formation of modern nations, and so forth.
It is not only the broadness of its substantial focus of the topics that gives comparative politics a character of great variety. This variety also appears in the research design and in the theoretical frameworks we apply (see the five Ts distinguished in Chapter 2). Today, this variety becomes even larger as comparative politics increasingly 'invades' the discipline of international relations (and vice versa).
From divergence to convergence ...
Comparative politics was born out of diversity. There would be no comparative politics without the diversity of political systems and their features. The literature up
to the 1950s assumed that there would be a convergence towards the model of the major Western liberal democracies. On the contrary, there has been divergence (in the form of alternative models of political order) and this has led to the actual development of comparative politics.
Is it still like that? Recently, trends towards convergence have been strong. The end of the Cold War in 1989 and the disappearance of the leading superpower that embodied one of the major alternative political models, the 'Third Wave' of democratization (the Arab Spring being its latest manifestation), the pressures towards market economy coming from world trade and globalization in a country like China, the numerous initiatives to 'export' and 'promote' democracy in Africa and Asia, democratic consolidation in Latin America—these are all patterns of worldwide convergence.
Thus, the role of comparative politics is called into question in a world that is less and less diverse. What is the future of comparative politics in a globalized world? Comparative politics—like all quasi-experimental' methods—bases its explanations on the covariation between phenomena which, quite naturally, leads to a focus on differences between analytical cases. Yet, how does such a discipline deal with the existence of commonalities, patterns of homogenization, and diffusion effects? Furthermore, comparative politics was built on the methodological assumption that cases—i.e. national political systems—are independent of each other. It has been less concerned with common aspects and interactions.6
As Sorensen notes in Chapter 24, '[t]he standard image of the sovereign nation-state is that of an entity within well-defined territorial borders: a national polity, a national economy, and a national community of citi-zensi and on this premise researchers thought that they could safely ignore what takes place outside the borders of the countries they were studying! For a long time the main concern of comparative politics has remained the study of the Westphalian territorial state.
However, it is increasingly difficult to maintain such a position and, indeed, the literature has addressed these issues. In recent years there has been a resurgence of interest in the so-called 'Galton problem^ i.e. the methodological issue raised at the end of the nineteenth century by the polymath Francis Galton concerning associations between phenomena that are, in fact, the result of diffusion and contagion between cases.
Today, most countries are open systems subject to external influences, borrowing and learning from the practices of others, and are part of multi-level governance arrangements (Marks and Hooghc 2004). For example, it is plausible to suppose that the development of welfare states in various countries (see Chapters 20 and 22) is affected by diffusion processes through policy transfers and policy learning. There is coordination when countries belong to overarching integrating organizations (the European Union, for example, as shown in Chapter
23) as well as cases of imposition by conquest, colonialism, and economic dependency (as discussed in Chapter 4, many current states were part of other states before secession). Finally, our current world, more than ever, experiences migrations (see the 'Comparative Tables' at the end of the volume).
The risk for comparative politics is—methodologically speaking—that of ending up with 'N = l! Already, Przeworski and Teune in their classic book on the comparative method have asked: 'How many independent events can we observe? If the similarity within a group of systems is a result of diffusion, there is only one independent observation' {Przeworski and Teune 1970: 52). Is our methodology fit to analyse common developments, changes without variation between cases, and situations of dependence between them? In an increasingly interdependent world, comparative political scientists realize that social phenomena are not isolated and self-contained, but rather are affected by events occurring within other, sometimes remote, societies, Within a 'shrinking world1 the problem is larger today than in the past.
... and back to divergence?
The last section of the book addresses precisely these questions, with chapters on integration, globalization, and promotion of democracy in non-Western parts of the world. This is where comparative politics and international relations become contiguous and their efforts, in the future, will increasingly be common efforts.
Nonetheless, one should not forget that while there are signals of convergence and homogenization, there are also signals pointing in divergent directions. Examples include the renewed role that religion plays in some parts of the world, such as the Arab world and Turkey, but also in Russia and the US; the emergence of right-wing populism in several European countries and the US in the last three years, and of left-wing populism in Latin America and Southern Europe; differentiation at the sub-national level that points to the resurgence of regional-ist phenomena, as with the Scottish referendum; also, supra-national integration, where it takes place, occurs to different degrees and at different paces as the Brexit referendum in Britain in 2016 witnessed.
All this is to say that it is difficult to detect linear, or cyclical, developments in world politics over short periods of time. This is one of the reasons why this book adopts a long-term perspective from the beginning of modern politics—the formation of national states, mass democracies, and industrialization in the nineteenth century. The French expression reculer pour rnieux sauler (to step backwards in order to jump further) was a favourite of Stein Rokkan, one of the pioneers of comparative politics, To have a firm ground for looking at the future fits very well with the philosophy of this book.
16
DANIELE CARAMANI
Further reading
'Classics' of comparative politics are shown in the boxes in this Introduction. These books should be on every comparative political scientist's shelves.
Overviews of the discipline
Blondel,). ('999) 'Then and Now: Comparative Polities',
Political Studies, 47(1): 152-60. Daalder, H. (1993) 'The Development of the Study of
Comparative Polities', in H. Keman (ed.), Comparative Politics
(Amsterdam: Free University Press), 11-30. Dalton, R. J. (1991) 'Comparative Politics ofthe Industrial
Democracies: From the Golden Age to Island Hopping', in
W. Crotty (ed.), Political Science {Evanston, IL Northwestern
University Press), 15-43. Eckstein, H. (1963) 'A Perspective on Comparative Politics,
Past and Present' in H. Eckstein and D. E. Apter (eds),
Comparative Politics: A Reader (New York: Free Press), 3-32. Mair, P. (1996) 'Comparative Politics: An Overview', in R. E.
Coodin and H.-D. Klingemann (eds), A New Handbook of
Political Science (Oxford: Oxford University Press), 309-35. Rogowski, R. (1993) 'Comparative Polities', in A. W. Finifter
(ed.), Political Science: The State of the Discipline (Washington,
DC: American Political Science Association), 431-50. Schmitter, P. (1993) 'Comparative Polities', in J. Krieger (ed.),
The Oxford Companion to Politics of the World (Oxford: Oxford
University Press), 171-7. Verba, S. {1985) 'Comparative Politics: Where Have We Been,
Where Are We Going?', in H. J. Wiarda (ed.), New Directions in
Comparative Politics (Boulder, CO: Westview Press), 26-38.
Recent treatments of comparative politics as a discipline
Almond, G. A. (1990) A Discipline Divided: Schools and Sects in
Political Science (Newbury Park, CA: Sage). Braun, D. and Maggetti, M. (eds.) (2015) Comparative Politics:
Theoretical and Methodological Challenges (Cheltenham:
Elgar).
Chilcote, R. H. (2000) Comparative Inquiry in Politics and Political Economy (Boulder, CO: Westview Press).
r
0E
INTRODUCTION TO COMPARATIVE POLITICS
17
Landman, T (2007} Issues and Methods in Comparative Politics:
An Introduction {2nd edn) (London: Routledge). Lichbach, M. I., and Zuckerman, A. S. (1997) Comparative
Politics: Rationality, Culture, and Structure (Cambridge:
Cambridge University Press). Peters, B. G. (199S) Comparative Politics: Theory and Methods
(Basingstoke: Macmillan). Wiarda, H. J. (ed.) (2002) New Directions in Comparative Politics
(3rd edn) (Boulder, CO: Westview Press).
Reference work
Boix, C, and Stokes, S. C. (2007} Oxford Handbookof
Comparative Politics (Oxford: Oxford University Press). See also the other titles in the Oxford Handbooks of Political Science series. For specific topics see the 'Further reading' section at the end of each chapter. Scientific comparative politics research publishes results in a number of specialized journals. The most important scientific journals with a focus on comparative politics are the following: Comparative Politics, Comparative Political Studies, Comparative European Politics, European Journal of Political Research, European Political Science Review, and West European Politics, among others.
In addition, most countries have political science journals that publish research in comparative politics. Examples include American Political Science Review, British Journal of Political Science, Revue Francaise de Science Politique, Scandinavian Political Studies, Pohtische Vierteljahresschrift, Irish Political Studies, Australian Journal of Politics and History, and Swiss Political Science Review.
Finally, for each subject (elections, parties, communication, etc.) there are specialized journals which include comparative work. Examples are: Party Politics, Electoral Studies, European Journal of Public Policy, Local Government Studies, Publius: The Journal of Federalism, journal of Common Market Studies, Journal of Democracy, Democratization, journal oj European Social Policy, Media, Culture and Society, and Political Communication.
1 [sj0t aLl authors would agree with such a division of disciplines, stressing that fields like public administration, policy analysis, political behaviour, and political economy are not part of comparative politics (see, for example, the titles of the volumes in the Oxford Handbooks of Political Science listed in the 'Further reading' section of this Introduction). More importantly this division into three main disciplines disregards methodology as a separate field. However, opinions diverge as to whether or not methods should be considered within the fields of political science, as they largely overlap with methods in other sciences, such as economics and sociology.
2 In these years the first studies on political culture were published (see, e.g., Banfleld 1958), followed by uthers stressing the differences in political cultures other than the Anglo-Saxon culture—namely based on clicntclism and patronage. For an example of cultural analysis see Putnam (1993).
3 This involved the creation of data archives, combined with the introduction of computerization and machine-readable data sets. Numbers are a universal language and thus, from a com-
parative point of view, the least problematic level of measurement of phenomena in diverse contexts.
4 Within the new-institutionalist theory different positions have emerged and have been summarized by Hall and Taylor (1996): (1) historical new-institutionalism devotes attention to the time dimension and the constraints set by past developments (path dependence) with a strong impact on policy analysis; (2) sociological new-institutionalism stresses how institutions model politics, and influence preferences by narrowing expectations and orientations; (3) rational choice new-institutionalism focuses on how institutions result from the aggregation of individual preferences and on institutions' contribution to solving collective action problems.
5 These cycles correspond to what Chilcote (1994) calls traditional, behavioural, and post-behavioural comparative politics.
6 Charles Tilly's critique of Stein Rokkan's model points precisely to Rokkan's failure to genuinely analyse the interactions between countries (Tilly 1984: 129)
For additional material and resources, please visit the Online Resource Centre at; http://www.oxfordtextbooks.co.uk/orc/caramani4e/
MASARYKOVA UNTVERZITA
Fakultu sociálních studií Jtétcvi 10 602 0U B R N O