I The hybridism of democratic transformation1 i ■ I'M ■ i iii. ,il views, there is an extensive volume of literature that follows a ....... ■ ■ understanding that democracy is the best type of government, with 111 it >n mi alternatives deserving consideration. The favoring view of demo-iirni . ih.n democracy promotes "freedom as no feasible alternative can" I11 ihl i ''H1' |i 89) and thiit it "provides by definition comparatively good protec-ii 11• ■ human rights" (Diamond 1999. p. 4). Proponents of democracy argue l i desirable because it is "the best and therefore unbeatable means of polit-1111/1111011" and undeniably possible because "it is the one form of govern-lueli evolves constantly to ensure that it is possible" (Gilley 2009, it ii i ' ■) However, others argue that democracy is not desirable because it ii i'u . ,ion and inequality (Riker 1982), promotes westernization, causes llMtitlnlils and inefficiency, and is impossible to achieve due to citizen stupidity nice (Gilley 2009). Regardless of democracy's desirability and feasib-'i.....11 iincntly the most celebrated form of governance which many govern-in. hi ii \ io promote and even more try, sometimes genuinely, to achieve. Even 'a .in)- to fully adopt the championed liberal democracy, some have adopted ill/ed version of democracy, with adjectives and increasingly spreading Ii i iik ilemocracies (Tisdall 2010). r lotwiths i, Hiding the long history of democracy and its acceptance by virtu-'ii in', i imntiy, at least rhetorically, to date it remains one of the most con-i. at .1 and debated concepts (Moller and Skaaning 2013). However, even if often a vague endorsement of a popular idea" (Dahl 1989, p. 2), demo-' willi other social science concepts, requires clear definition (Sartori ■ in avoid vagueness when measuring it within different cases (Adcock and I Hi, i (ion, The literature is filled with the multitude of "democracies with live;" ((oilier and Levitsky 1997). However, while scholars debate the 111 components of democracy, the regime championed by democracy i Kim i .', -even if not always called as such—is liberal democracy, which con- ii nl.....testation and participation and is supported by the rule oflaw. These iilmtml', understandings of democracy, sometimes unwittingly, have translated ' i , i ilic democracy-promotion sectors covering the political life of target iniii-i Given that these different sectors—elections, political parties, and i 'i in n i\ may be on different levels of development in the target country, it logical to assume that democracy-promotion outcomes may he different depending on the sector and domestic conditions. These different outcomes in their turn arc likely to produce democracies with adjectives, emphasizing that mere establishment of democratic institutions does not guarantee their democratic functioning. However, is there an analytical framework that would help in assessing the potential outcome and—even more importantly— anticipating it? By closely analyzing comparative politics and international relations literature, I show in this chapter that decades of democracy promotion and its research have yet to produce a meta-theory that would wield some level of predictability. In addition, the scholarly literature has been characterized by an obsolete and artificial line between international and domestic factors of democratization. In other words, democracy promotion has often been viewed in the vacuum of promoters' strategies without paying attention to the facilitating or impeding factors provided by local milieu. While arguing that democracy promotion and local democratization should be analyzed as an interconnected process, the chapter turns inter alia to international socialization (Schimmelfen-nig el at. 2006) in its endeavor to develop an analytical framework for democracy promotion. This particular understanding of international socialization provides an opportunity to equally consider developments within the international and domestic realms, differentiating different types of strategies and democratic transformations. Such close consideration of international and domestic conditions and strategies of democracy promotion allows expectation of specific outcomes of democracy promotion depending upon varied values of international and domestic conditions. However, before proceeding to the analytical framework, the chapter provides in-depth analysis of the concepts of democracy, democratization, and democracy promotion to outline various democracy promotion projects and their objectives. Democracy and democratization Understanding of the outcome and impact of democracy promotion has often been limited due to scholarly disagreements over the conceptualizations of democracy. Thus, conceptualizing one of the most empirically and theoretically debated concepts can be a daunting task. The elusiveness of the definition of democracy mainly derives from its changing nature based on international and domestic processes, along with cultural and historical specificities of different societies. Thus, regardless of the currently prevailing definition of democracy, its conceptualization is closely interconnected with the context in which it is conceptualized and should be addressed by a "constructivist approach" (Whitehead 2002, p. 7). Definitions of democracy adopted for a study of democracy promotion—a highly practitioner-orientated concept—should be not only theoretically but also empirically grounded, carefully encompassing the existent literature but at the same time not being overly ambitious empirically. The scholarly definition adopted here is the classic one of liberal democracy, which is also the ; in- nyoriuiam <>i hvuhm i (wii " ufHjvrmwiun one type of government most praised and actually promoted by the practitioners, defined as: a meaningful and extensive competition among individuals and organised groups for all effective positions of government power: a highly inclusive level of political participation in selection of leaders and policies, at least through regular free and fair elections, and a level of civil and political liberties—freedom of expression, freedom of press, freedom to form and join organizations. (Diamond el al. 1989, p. xvi) This classic definition, which is almost entirely adopted by the practitioners, as discussed later, may not be empirically applicable in all cases of democracy promotion, consequently hindering democratic transformation. Thus, one of the arguments here is that the definition of democracy in actual policies may need revision. Whitehead (2002, p. 20) has observed that "democracy has some indispensable components without which the concept would be vacuous." However, these components are not stagnant and can be differently arranged. One such component is elections. Following Schumpeter (Schumpeter 1947), many scholars have regarded democracy as a system where the "most powerful collective decision makers are selected through fair, honest and periodic elections" (Huntington 1991, p. 7). The minimalist conception of democracy has helped to avoid conceptual overstretching by "moving up the ladder of generality" and might have been appropriate decades ago, encompassing a larger number of cases. However, this narrow approach equating democracy to elections is not compatible with the current situation on the democratic scene. Since the third wave of democratization, a record number of countries have adopted elections as government-choosing procedure. Nevertheless, in just a small portion of these countries are elections truly competitive, without massive fraud and voter intimidation, and they hardly correspond to the understanding of "good democracy" (Croissant and Merkel 2004). This is particularly true in the environment of a gradually decreasing number of democracies and an increasing number of democratic setbacks (Puddington 2013). Gradually, the scholars of democracy have started including other requirements for a regime to be democratic, thus creating the "expanded procedural minimum" of democracy. Among the new features are the requirement for the elected officials to effectively govern without being overly constrained by non-elected entities, and the requirement for the civil powers to exercise control over the military. These features have been especially compatible with Latin American reality (Karl 1990: Huntington 1991: Schmitterand Karl 1991; Mainwaring el al. 1992). Other conceptions of democracy involve features corresponding to the established industrial democracy, which entails certain political, economic, and social features. The maximalist conceptions—hardly applicable to a handful of real cases—"include equality of social and economic relations and/or broad 16 Promotion, democratization, transformation popular participation in dDCision-mnking til all levels o( politics" (Collier and l.evilsky 1996. p. 8). However, no regime should be considered democratic until the offices are contested (Chcibub et al. 1996). With the spread of democracy around the globe and the subsequent development of democratization studies, simple distinction between democracies and non-democracies has become insufficient. Many formerly authoritarian countries took the road of democratization and, while some achieved democracy and even consolidated it, others have stagnated in between. The research has shown that transfer from an authoritarian rule could be to a democracy but it also could result in a liberalized authoritarianism (dictablanca) or illiberal democracy (democradura) (O'Donnell et al. 1986), or hybrid regimes of competitive authoritarianism (Levit-sky and Way 2010). These developments have shown that 'the trend toward democracy has been accompanied by an even more dramatic trend toward pseudosemocracy" (Diamond 2002, p. 27). Transitology and consolidology assumed that new democracies can equally move either forward towards consolidation of democracy or backward towards authoritarianism (Carothers 2002). Thus, new classifications of regimes that initiated democratic reforms have emerged in the literature. Close examination of democratization literature lias discovered more than 550 examples of "democracies with adjectives" (Collier and Levitsky 1997). These new definitions of democracies evolved because scholars on the one hand sought to increase conceptual differentiation in capturing new forms of democracy, and, on the other hand, sought to avoid conceptual stretching while working on cases of democracy which did not correspond to the previously discovered conceptualizations (Collier and Levitsky 1997). The need to better understand the mixed features of post-authoritarian regimes have often been voiced (Malloy and Seligson 1987: Karl 1995; Levitsky and Way 2010; Baracani and di Quirico 2013). The differentiated conception of democracy is also important as often it serves as a causal variable in research (Collier and Levitsky 1997). However, the boundaries between these differentiated conceptions are very often "blurred and controversial" (Diamond 2002, p. 27). The "diminished subtypes" (Collier and Levitsky 1997) of democracy have been generated based on the root concept of democracy, which was the procedural minimum, expanded procedural minimum, or the established industrial democracy. The diminished subtypes of democracy are characterized as a particular type of defective democracy through a specific missing or "weakened" component of the root concept of democracy (Collier and Levitsky 1997). Thus, while an authoritarian regime can become a democracy, different types of democracies can also become more democratic (Diamond 2002. p. 34). Consequently, in research on democracy promotion, democracy should be regarded as a simultaneous process within a country, and as the ultimate goal for promoters and the target country until the set objectives are achieved. The presented divergent opinions on the notion of democracy and its value demonstrate that a comprehensive conceptualization of democracy that would satisfy both academics and practitioners is practically impossible. In addition, because conceptualization is an evolving activity that is closely correlated with ih< hybridism of democratic transformation \i the explanatory power of a theory (Kaplan 1964). the argument over the "correct definition" is redundant ((iutlman 1994. p. 12). Thus, the suggestion is to "avoid the extremes of including too much or too little in a definition relative to their theoretical goals" (Munck and Vcrkuilen 2002. p. 9). While maximalist definitions of democracy are of little analytical use because they are too overburdened, the minimalist definitions bear the danger of including all actually divergent cases under one subtype (Munck and Verkuilen 2002). From a practitioner/ democracy promoter point of view, adoption of a maximalist definition—which includes inter alia such attributes as freedom from war, provision of social rights, and transition to welfare state—might be unrealistic, especially when democracy promotion is implemented in a relatively poor country. On the other hand, there is hardly a rationale for the adoption of a minimalist electoral definition for a study on democracy promotion because it does not illustrate the democratic reality of the target country, as elections might be in place though still largely violated and even restrictive. While the concept of liberal democracy or polyarchy (Dahl 1989) might be well-known and ideal-typical, democratization studies show that there are various types of "democracies with adjectives" (Collier and Levitsky 1997) in the real world. However, democracy promotion activities do not aim to develop hybrid regimes (Karl 1995). electoralist democracies (Vanhanen 1997). procedural minimums of democracy (O'Donnell et al. 1986; Mainwaring et al. 1992), and certainly not competitive authoritarianism (Levitsky and Way 2010). Though efforts of democracy promoters may be inconsistent and sometimes contradictory, they aim to promote liberal democracy, which includes effective participation, voting equality, inclusion of adults (Dahl 1989). and the provision of civil and political liberties. Thus, in DahTs terms they should aim to ensure competition and contestation. While an electoralist regime can be the starting goal for a democratizing country, neither domestic nor international actors should be satisfied with short-term results. Excessive praise from a promoter for mere organization of elections, or absence of any social and especially material shaming for rigged elections, has the potential of endangering future democratic transformation as domestic actors may regard the current situation as the ultimate democratization goal of their donors. Therefore, in order to not hinder the democratization process, democracy promoters should set feasible goals to achieve in a democratizing country, without falling into the minimalist trap. It is not assumed that promotion of liberal democracy necessarily results in its inevitable establishment. Instead, it may result in the establishment of a democracy with adjectives. Nevertheless, the definition of liberal democracy addresses the requirements of contestation and participation, and avoids the minimalist exclusion of attributes or maximalist overstretch of the concept. While contestation includes the right to form political parties and participate in elections avoiding intimidation mainly from the incumbents, participation ensures fairness of the voting process, access of candidates to public financing and media, freedom of expression, freedom of the media, and equal provision of these civil and political rights (see Table 1.1). The empirical *,»c*RYKOVAUN1VE_RZVTA imi nyonatsm oj aemocranv P'atvpormaiion iv ik'hiiiiniii ol democrat\ vviilim democracy promotion, which is the main reference point for I he empirical research and is explored in relevance lo the chosen promoters later in the text, may slightly differ. As with democracy, democratization has been an issue of heated debate among scholars, who have not come to an agreement either on the nature of democratization, its preconditions, or the entire process of initiation and completion. If it is conceptually and practically appropriate to accept a contextually variable definition of democracy, then the definition of democratization cannot be rigidly fixed either, and may depend upon the contextual variations of certain processes. However, there are several methods of understanding the democratization process and, more importantly, when it is complete. The "two-turnover test" suggests that the process of democratization can be considered underway once the authoritarian regime has collapsed, and it is complete—and thus democracy is on the way to consolidation2—after there have been two successful and peaceful transfers of government between competing parties (Huntington 1991). Others argue that democratization is over when all actors consider the electoral process of democracy to be the "only game in town" (Di Palma 1990), and in this case actions of actors are not as important as their beliefs and perceptions. Apart from the debatable possibility of measuring beliefs and perceptions, this definition is rather dubious when thinking of the cases of Spain and Italy (Whitehead 2002). Thus, while taking these measurements into consideration, democratization should rather be viewed as an open-ended process, with the possibility of different outcomes, especially given recent backlashes (Ambrosio 2009). The definition of democratization should be closely related to the adopted understanding of democracy. The understanding of the beginning and completion of democratization should closely correlate with the understanding of the promoted type of democracy. Thus, the practitioner's understanding of democracy and democracy promotion should play an even greater role than the scholarly one when evaluating the process of democratization from a policy point of view. The practitioner's understanding of democracy plays an important role because "democratization is best understood as a complex, long-term, dynamic and open-ended process [and] consists of a progress towards a more rule-based, more consensual and more participatory type of politics" (Whitehead 2002, p. 27), and can provide the necessary closure to the process. Thus, democratization is not determined by its outcome. It is rather determined by a process, which may lead to a specific outcome, clearly outlined in a democracy-promotion policy and the democratization initiatives of local actors. While the process of democratization may also vary, since actors may play important roles to varying degrees, democratization processes within democracy promotion can be predetermined by the adopted democracy-promotion policy. As for the theoretical understanding of democratization, it can be regarded as underway when certain actions are taken on the way to reach the outcome mentioned in the adopted definition of democracy. Thus, while there are certain approaches to democratization, these should be adapted to specific examples of democratization to provide a comprehensive understanding of the process. I rumwifi/r«! «nrnnnm tttusn, irurw/furmwi Detnocrscj promotion I he collapse of the Soviet Union became "an historic opportunity lor a transition to a peaceful and stable international order" because the "international community has a vital interest in the success of this transition" (US Congress 1996), and donors have to provide necessary support and expertise. Democracy promotion has become one of the main foreign-policy objectives of many already-consolidated democracies and some of the "rising democracies" (Carothers and Youngs 2011; Petrova 2012). Reasons behind the rise of democracy promotion activities may range from materialistic pursuit of stretching economic and territorial power, and altruistic care for the well-being of other societies. In addition, promotion of democracy may sometimes conflict with other foreign-policy objectives (Wolff et at. 2013). One of the most influential theories of explaining rationales for democracy promotion is the democratic peace theory, arguing that democracies do not wage wars against other democracies (Kegley and Hermann 2002). Thus, in pursuit of a peaceful international system, democratic states opt for turning autocracies into democracies to avoid military conflicts and to achieve peaceful resolution of disputes. Democracy promotion is considered to be "the most effective long-term measure for strengthening international stability; reducing regional conflicts: countering terrorism and terror-supporting extremism; and extending peace and prosperity" (Bush 2006). Another rationale, though not in contradiction with the previous one, is the pursuit of a prosperous international system. Thus, democratic states promote democracy because of conviction that democracies are more effective and efficient in producing development and economic growth than are autocracies and dictatorships (Johnson 2002). However, the focus of this study is on democracy promotion itself rather than its rationales, which largely play a secondary role in the choice of implementation strategy and do not closely correlate with the chosen variables to be specified in the following chapters. Scholars of democracy promotion have repeatedly expressed concerns over the lack of an adequate theoretical framework offering predictive value for democracy-promotion studies (Burnell 2007, 2008a). The literature mainly relates to the practitioner (Carothers 1999, 2004) and ex-post (Burnell 2008a) views of democracy promotion, which is overwhelmingly a narrative of the democracy-promotion efforts of the USA (Carothers 1999) and the EU (Gillespie and Youngs 2002; Youngs 2002), and the role of democracy promotion in their foreign policies. The lack of a theoretical framework for studying democracy promotion ex-ante as a process and indicating its potential effectiveness is obvious. So far, only some practitioner tools for ex-post evaluation of democracy promotion have been available. They have been developed by different foundations and development agencies—the United States Agency for International Development (USAID), the Swedish International Development Cooperation Agency (SIDA), the Deutsche Gesellschaft für Technische Zusammenarbeit (GTZ) etc.—and criticized by academics (Crawford 2003a, 2003b). Carothers even claims that "democracy promoters treat political change in a turn iim'i miMii a/ miiim t urn trufKyurmuttun pMudOsetentiflc manner" (2004, p. 102), thus their democracy promotion not have a theoretical background. At the same time scholars criticize democracy promoters, arguing that they "rarely have much sense of history about what they Jo, cither unli regard to the countries in which they are working on or to the enterprise of using aid to promote democracy" (Carothers 1999, p. 19). Though this is potentially true, besides criticizing, academics should work on the development of an analytical framework that could help in the formulation of democracy-promotion policies. This section of the chapter presents analysis of different types and levels of democracy promotion, which are further used for developing an analytical framework. Despite large volumes of academic work on democracy promotion there are less than a handful of works suggesting definitions which can be used when researching the phenomenon. Although there may be divergence between academic and practitioner understandings of democracy promotion, initial adoption of an academic concept is essential. Thus, democracy promotion consists of: overt and voluntary activities adopted, supported, and (directly or indirectly) implemented by (public or private) foreign actors explicitly designed to contribute to the political liberalization of autocratic regimes and the subsequent democratization of autocratic regimes in specific recipient countries. (Schmitter and Brouwer 1999, p. 12) As mentioned in the definition, only activities officially labeled as democracy promotion are included in this study because other activities, regardless of their democratizing nature, may fall under covert intelligence efforts of international actors and not provide full understanding of variables under consideration. The given definition does not include the implicit actions of external actors, such as diplomatic and intelligence activities, health campaigns, and so on, and also omits international factors which do not require the presence of a promoter. Thus, it also excludes cases of contagion, while militarized democracy promotion is not taken into consideration by this study since the consent of a target is not required, reducing the chances for success from the beginning. This definition provides a general understanding of what democracy promotion is and leads io further classifications of democracy promotion. While some distinguish four levels of democracy promotion targets—individual citizens, civil society, political society, and state institutions (Schmitter and Brouwer 1999)—others distinguish three sectors of democracy promotion: electoral process, state institutions, and civil society (Carothers 1999). Democracy promotion on the level of state institutions "supports institutions of public authorities not to improve their repressive capacity, but to reform those institutions" (Schmitter and Brouwer 1999, p. 21) by strengthening legislation, aiding the rule of law, and developing local government (Carothers 1999). Democracy promotion at the level of political society is understood as "assistance to the specialized organizations and movements of political society," usually involving competition for office (Schmitter and Brouwer 1999. p. 21), by political party uilding unci electoral aid (('iiroihers 1999). At the civil society level, democracy promotion supposes assistance to "organi/iilions thai arc al least partially voluntary and are relatively independent from the state" (Schmitlcr and Rrouwer 1999, pp. 19-20) by NGO building, civic education, media strengthening, and union building (Carothers 1999). At the level of individual citizens, the objective of democracy-promotion programs is "to transfer knowledge about democratic institutions and practices, socializing individuals to democratic values, and changing their behavior" (Schmitter and Brouwer 1999, p. 19). Carothers includes individual citizens in the civil society sector. Though these categories were developed based on examining US democracy promotion, they can be applied to other promoters too and can be modified if necessary. These groups largely overlap, thus in Table 1.2 they are combined into one group which entails goals and types of promotion from both groups. The template given in Table 1.2 makes clear the adherence of some democracy promoters to the idea that democracy should be promoted through bottom-up and top-down approaches. However, though these approaches should be used simultaneously, strategies usually differ from promoter to promoter. Encouragement of a multiparty system with increasing both supply (state institutions) and demand (civil society) are equally important for successful democracy promotion Table 1.2 Sectors and levels of democracy promotion Level/sector Goal Type of democracy promotion State institutions Political society Civil society Individual citizens Democratic constitution' Independent judiciary and other law-oriented institutions Representative legislature Responsive local government Pro-democratic military Free and fair elections Strong national political parties Free and fair elections Active advocacy NGOs Strong independent media Free and fair elections Politically educated citizenry Constitutional assistance Rule-of-law aid Legislative strengthening Local government development Civil-military relations Electoral aid Political party building Electoral aid NGO building Media strengthening Elections observation Voter education Professional and educational exchange Source based on Schmitter and Brouwer 1999. p. 44 and Carothers 1999, p. 88. Note I Assisting in the establishment of a democratic constitution would mean that the promoters work from scratch and the transition to democracy has just began However, in most or the cases of democracj promotion by consent, a democratic constitution has already been adopted based on the experience of other established democracies. Thus, in such cases the promoter moves to ensure that furthei democratic features enshrined in the constitution do not stay on paper only: free and fair elections, independent judiciary, representative legislature etc. ithers 1999). I litis, stale institutions should be established through a democratic process, be stable, and have the capacity to perform their functions without being pressured by the executive or the military. At the same time, vibrant and independent civil society should be able to represent the interests of the citizens and provide checks over the government. According to some promoters, the democratization process proceeds along a "relatively set path" (Carothers 1999. p. 87): a non-democratic regime faces popular demand for liberalization; opposition and civic actors consolidate their power: multiparty elections are held; an elected government achieves power; and democracy is further consolidated. Although this sequence may take place in many democratizing countries, the democratic quality of these events might be far from the imagined ideal. While civic actors may multiply and elections may be held, the quantity of civil society actors does not guarantee the fulfillment of their functions, and elections are not necessarily free and fair. Thus, though this sequence can be taken into consideration by promoters, instead of congratulating themselves and their domestic counterparts on groundbreaking performance, they also need to pay attention to whether these events carry a democratic character. Schminer and Brouwer (1999) also emphasize the importance of differentiation between democracy promotion and democracy protection. In contrast to democracy promotion, democracy protection is "overt and voluntary activities adopted, supported, and (directly or indirectly) implemented by (public or private) foreign actors explicitly designed to contribute to the consolidation of democracy in specific recipient countries" (Schmitter and Brouwer 1999, p. 14—emphasis added). Democracy protection does not intend to change the current political regime, especially if it is democratic, but acts to make it more effective and efficient. Likewise, organization of police training for enforcement of human rights and support for privatization of trade unions are activities directed at consolidation of democracy. While democracy-promotion activities are likely to be more effective on the state and political society sectors, democracy protection activities can be influential when targeting civil society and individuals. However, some of the target sectors in democracy promotion and democracy protection overlap, achieving more results in one case than in the other. In addition, the boundary between democracy promotion and protection is often blurred in the actual activities of promoters who do not strictly differentiate between the two. The introduced analytical frameworks help with grasping the concept of democracy promotion and in differentiating between its types and sectors/levels. However, they do not elaborate on the mechanisms for implementing democracy-promotion policies. Moreover, they do not specify the conditions under which a democracy-promotion policy may have a certain outcome. This gap in the literature on democracy promotion is possible to bridge only by analyzing the interaction between the international and domestic variables. Further, an analytical framework is suggested that can be helpful in democracy-promotion studies. These projects are selected as operationalization of political-society projects (party development), civil-society projects (media development), and projects implemented on all levels (elections). Democracy promotion varii-s no! only in its possible implementation strategics but also in its targets and sectors. Classification of democracy promotion according to state, political society, civil society, and individual citizens helps to understand what the most important features of democracy are according to democracy promoters. 11 also helps in establishing a link between the academic and the practitioner approaches to democracy and democratization. Selection of elections, party development, and media-development projects not only reflects the academic definition of democracy as constituting contestation and participation, but also reflects the conventional democracy-promotion package of promoters. Although democracy promotion has been launched more than two decades ago and covered more than 100 countries, the academic world has not paid sufficient attention to its constituting elements and the practitioner world is still in a search for effective policies. In addition, not all sectors are equally targeted by all promoters. This overview shows that some of the projects overlap and a project targeting one component of democracy may also indirectly affect the other two. Another conclusion is that, despite at least two decades of democracy promotion, these projects still wander in twilight and need improvement and most importantly applicability to domestic contexts. Democracy promotion through international socialization Despite two decades of democracy promotion, "there is still a dearth of theoretically informed comparative studies" (Wolff et al. 2013, p. 3). In addition, there is a lack of analytical frameworks (Burnell 2007, 2008b) or "conceptual models" (Kumar 2012, p. 3) for democracy promotion with strong theoretical and empirical foundations. Schimmelfennig et al. (2006) have previously endeavored to fill this lacuna by developing a robust and empirically testable theoretical framework, based on the notion of international socialization, that can be applied to the implementation stage of democracy promotion and can also incorporate domestic democratization. However, this framework still needs development and application to other cases of democracy promoters and the targets of their activities in order to identify and provide full explanation of the concept of democracy promotion and identify the effective strategies. Here it is updated in terms of its structural and geographical scopes, as well as its independent and dependent variables. This section presents the original framework, providing definitions of important concepts, presenting the rationalist-constructivist debate, identifying the types of socialization and strategies, and specifying necessary international and domestic conditions for successful socialization. Since here the analytical framework of international socialization is applied to democracy promotion, any mention of international socialization components, approaches, and strategies should be equally regarded as those of democracy promotion. International socialization is "a process in which states [or other targets] are induced to adopt the constitutive rules of an international community" (Schimmelfennig et al. 2006, p. 2). The state or another target for socialization is iii, nyhrtaism of democrats transformation 25 considered successfully socialized when it adopts rule-creating domestic mechanisms and powerful institutional and political processes that guarantee compliance and discourage opposition to this rule. From the point of view of democracy-promotion studies, this definition of international socialization is somewhat rigid and one-dimensional because it does not take into consideration possible varying degrees of democratic transformations. It is one-dimensional I Mise it supposes the creation of institutions and processes to guarantee compliance to democratic principles, but does not elaborate on whether those principles are guaranteed to be complied with. Thus, this definition is used as a 11, iriing point for the development of a more comprehensive understanding that would encompass various possibilities of policy outcome. Analyzing international socialization from a forward-looking perspective, "as a process directed al or potentially leading to rule adoption by the target states" (Schimmelfennig et al. 2006. p. 2). addresses the concern that domestic political actors and processes have not received sufficient attention (Schmitter and Brouwer 1999; Schmitz 2004). The literature on international socialization defines two general approaches through which international organizations promote their rules and norms. These methods are strategic actions of incentives and coercion on the one hand, and appropriate actions of persuasion and example on the other. The first one is the logic of appropriateness advocated by constructivists. and the second one is the logic of consequentiality advocated by rationalists (Olsen and March 2004). and these represent "opposing ideal-types" (Schimmelfennig et al. 2006, p. 16). From the rationalist perspective, states act in the technical environment of the international system, and international socialization is not a relevant concept per se. Socialization is only possible as a strategic action via incentives or coercion, and aims to change the behavior of the target but not its identity or interest (Iken-berry and Kupchan 1990; Schimmelfennig 2002; Jupille et al. 2003). From the constructivist perspective, international socialization argues that states act in a social environment and international socialization is based on the concept of appropriate action. Thus, an agency socializes target states by social persuasion and benign example, acting as a role model and changing the identity and interests of the socialized (Risse et al. 1999; Checkel 2001). However, none of the ideal-types alone can provide plausible and empirically grounded explanations for the success or failure of international socialization. Though international socializers or democracy promoters publicize the image of a socially constructed role model pursuing benign purposes, it is unlikely that these purposes do not derive from their material interests. Likewise, domestic actors, nurtured in their domestic yet non-socialized environments, are unlikely to regard foreign rules and norms as appropriate because of their mere intemationality. Schimmelfennig and his collaborators design an analytical framework that regards socialization as a "strategic action in a community environment" (Schimmelfennig 2003) and views it as "a bargaining process with normative constraints" (Schimmelfennig et al. 2006, p. 25). Thus, they adopt perspectives from both rational institutionalism and social constructivism, seeing domestic actors ""■'"I'm. ii< i/n '1 I III I. I II l( III. II III! t/i >l III,lllllll as rational and risk-averse, trying to maximize their utilities, while promoters are seen as "realist actor|s| in normative clothes" (Sceberg 2009, p. 95). While the framework adopts the constructivist vision of a cultural international environment and strong international organizations as socialization agents, it does not agree with the logic of appropriateness and disregards the possibility that domestic actors will change their self-defined interests based on interaction. Consequently, a theoretical framework of international socialization, analyzing the actions of international actors and relevant reactions of domestic actors, needs to consider that not always are actions based on rational calculations and not always are they driven by social and appropriate motives. To understand whether and how strategic action is formed under the influence of the international environment, Schimmelfennig and his collaborators proceed with theorizing based on a pure rationalist bargaining approach that can capture the concept of a strategic action in the international community. From the rationalist perspective, the socializing agency and the target of socialization are motivated by their own self-interested political preferences, which, according to Schimmelfennig et al. (2006, p. 18), are "material and power-oriented." These can include security issues, welfare, or maintaining political power. Thus, a socialization agency and a target consider the socialization process as a pathway to achieve their own interests. Neither of them take the norms of the community for granted, but rather view them as undeniable facts that can either constrain or promote certain behavior. Following the same logic and echoing the critique of democracy promotion mentioned earlier, the socialization agency does not necessarily possess a genuine belief in democracy or other issues that they promote. They rather use the former to achieve other security or power-related goals. However, regardless of the motivation, the socializing agency has an interest in the socializing process, otherwise it would not venture into it. In addition, the spread of its own community rules would help create new alliances and prove the viability of its own system. Similarly, the targets when adopting the promoted norms or rules take into consideration potential benefits of adoption. However, the utility for the target in socialization, at least from its own perspective, is not taken for granted, otherwise the whole socialization process would be redundant because the norms would be adopted even without the socialization agency. Thus, in any case of socialization process there is a certain degree of domestic resistance (Schimmelfennig et al. 2006). To overcome this resistance, a socialization agency has to be willing to employ certain strategies and tools that can also be costly to the international community. The extent to which a socializing agency is willing to pay to socialize a specific target depends upon the importance of the target to the interests and preferences of the agency. From this strategic perspective, the socialization agency reinforces its community rules or democracy, and the capability of reinforcement through punishments or rewards depends upon the bargaining power of the agency. The bargaining power of the agency at the same time is constrained by the credibility and the size of its rewards, and the credibility and severity of its punishment. Thus, to induce the target to internalize the socialized in? iiyrnuiism of th'itiOi rath transformation 27 or promoted rules, the agency needs to be able to match its "offer" with domestic adaptation costs. II the adaptation costs of a target are higher than the promised rewards, (hen there is little chance for successful socialization. In a nutshell, the main proposition of the framework states that successful socialization depends Upon the agency's bargaining power, credible constraints and incentives, well-developed monitoring systems, and the size of domestic adaptation costs. I lowever, given the fact that socializers act within a social environment, the pure form of material bargaining is not possible and is constrained by structural features of the international community and the target. Schimmelfennig et al. (2006) further distinguish between strategies of reinforcement and persuasion. Reinforcement entails specification depending upon the employment of incentives: reinforcement by reward, by support, or by punishment. The framework also categorizes incentives as tangible or intangible. A socializing agency can also opt for different channels of socialization: inter-'overnmental (targeting the governments directly) and transnational (targeting non-governmental actors such as social movements, interest groups, or business actors) (Schimmelfennig et al. 2006). The last categorization correlates with the categorization of levels in democracy promotion: state level with the intergovernmental channel, and civil society level with the transnational channel. The political society level of democracy promotion can be considered as intermediary because it also serves as the channel of expression for the public will to the government and the state. The targets of socialization face the reinforcement strategies that control incentives and disincentives, and persuasion strategies that imply arguing for the justification of the promoted norm. As research shows, socializing organizations employ both strategies, though not to equal extent (Schimmelfennig et al. 2006). The reinforcement strategies in their own turn are categorized depending upon the tangibility of incentives they offer or withdraw. Social reinforcement employs "socio-psychological" (Schimmelfennig et al. 2006, p. 33) instruments of reward (international recognition or public praise), punishment (shaming or shunning), and support (additional meetings with the agency or arrival of expert groups). This strategy is generally used by the Council of Europe and the OSCE, which are socializing agencies without considerable economic or military leverage. The material reinforcement strategy is usually used by socializing agencies that have the capability to enforce the promoted norm by means of their material leverage, e.g. NATO and the EU. The most widely used strategy of material reinforcement is the reinforcement by reward, better known as political conditionality (Schimmelfennig and Sedelmeier 2005: Schimmelfennig et al. 2006). This strategy supposes tangible rewards when the target state complies with the conditions set by the socialization agency. While reinforcement by support supposes additional support in case of compliance, reinforcement by punishment supposes not only withdrawal of current support but also introduction of specific sanctions (as in the case of NATO's 1995 Operation Deliberate Force in Bosnia-Herzegovina). However, due to their costly nature, the last two strategies are used only if political conditionality fails and if due to high . *■. -; ■ unit am nterdepeiuleoce the mkinli/ation of a mrgi-i state is more important than the actual costs oI support or punishment (Schimmelfennig et al. 2006). In choosing a socialization strategy, a rational socialization agency pursuing successful socialization should lake into consideration not only its own preferences and capabilities but also the domestic conditions of the target state. As noted above, the usual strategy of socialization is reinforcement by reward, which leaves the decision of compliance strictly to the target state. Assuming that domestic political actors are rational and try to maximize their utilities, and taking into consideration "state-centric domestic structure and the electoral volatility [of post-communist states]" (Schimmelfennig et al. 2006. p. 53), it is unlikely that the target state complies with liberal norms threatening its current state of affairs if the domestic costs of adaptation are higher than the tangible rewards. Therefore, it is doubtful that social reinforcement alone would be a successful strategy of democracy promotion, especially in countries where adaptation costs are rather high. Variables and arguments or international socialization framework The dependent variable of Schimmelfennig and his collaborators' study is compliance, which analyzes how a state reacts to international conditions and under which conditions it complies or does not comply with promoted norms. Besides the set of already-presented variables, they also add a set of constructivist variables to reflect the choice of understanding international socialization as a strategic action in an international community. Being interested primarily in the "thresholds of effectiveness," Schimmelfennig et al. (2006. p. 57) prefer dicho-tomous categorization of their variables. The dependent variable is distinguished as compliance and non-compliance; whereas the independent variables are distinguished as having negative and positive values. The positive value implies the likelihood of compliance, while the negative value would most probably hinder compliance. The compliance is conceptualized as law passed or a treaty signed by a target state in accordance to the promoted rule or norm. However, Schimmelfennig and his collaborators do not proceed with a more in-depth analysis to see whether the target of the socialization actually adheres to the promoted rule after the initial compliance. Omission of the behavioral compliance as a dependent variable is explained by the difficulties in obtaining cross-national data and customary satisfaction by the socializing or promoting agent with the formal compliance of passing a law. However, in the case of democracy-promotion policies the behavioral compliance is the actual goal of promoters, and omitting it in research would lead to practically inapplicable results. This book intends to fill this lacuna and introduces its own dependent variable in the next chapter. To test the hypotheses, the framework of international socialization uses a set of rationalist and constructivist variables that can also be distinguished on the basis of international-domestic divide (Schimmelfennig et al. 2006, pp. 57-60). /we tiynruitsm a) iuiiik, i.un irarajormanon e* I he rationalist independent variables arc incentives (kind and size of tangible rewards), credibility (of promise to pay the reward), and costs (which are low in cases where rule conformity docs not threaten the current distribution of power). The constructivist independent variables are legitimacy (measuring whether the socializing agency itself complies with the promoted norm and promotes it on a constant basis), identification (measuring the extent to which a target state identifies with the international community and promoted norms), and resonance (measuring the extent to which domestic institutional design matches with the promoted norms). Closer consideration of these variables is given in Chapter 3, which reveals the research design and main theoretical and methodological contributions of the book. The hypotheses containing the test and control variables are tested on nine European country-cases—Belarus, Yugoslavia (Serbia). Turkey, Slovakia, Romania, Estonia, Latvia, Northern Cyprus, and Montenegro—as targets of socialization by the European community organizations—EU, NATO. OSCE, and CoE—using the Qualitative Comparative Analysis (Ragin 1989). Schimmelfennig et al. (2006, p. 55) argue that "credible membership perspective and low domestic political costs of rule adoption are both individually necessary and jointly sufficient conditions of successful socialization." Thus, their research findings support the rationalist perspective of understanding the mechanisms of international socialization. Schimmelfennig et al. also argue that, though social constructivist factors of identity, legitimacy, and resonance can to some extent constrain the decisions of socializes in choosing goals and targets, they do not account for the mechanisms of successful socialization. The rationalist variables tend to explain better the successful mechanisms because the main targets of the socialization process studies by Schimmelfennig and his collaborators are states that are not members of the community yet. Being outside of the community and not facing conditional incentives makes them exempt from social obligations to the community norms. This comparative-static analysis also demonstrates that "the gap between liberal countries with or without major minority rights problems first narrowed and then disappeared" (Schimmelfennig et al. 2006, p. 253), proving socialization to be a continuous rather than a short-term process. The international socialization framework focuses on the promoters' strategies and domestic conditions of the target countries, thus enabling the researcher to identify effective democracy promotion strategies in the presence of specific domestic conditions. The Schimmelfennig et al. study (2006) was completed with cases of international socialization in Europe by Western international organizations. They conclude that successful socialization requires a specific set of domestic conditions supported by a membership incentive. Thus far, the framework has been applied exclusively to international organizations and has examined exclusively European cases, limiting the scope of its applicability. However, individual states ignored by the framework, such as the USA, also have an important role in democracy promotion, even though they cannot offer membership per se. In addition, the EU, the primary promoter in the i sc.iuIi ul Schimmcllennig <7 «// (.'Odd). also promotes demoi i,u s v\ iiluuil oliering membership. I hus, the geographic and structural dimensions ot the framework should be widened by applying it to other types of cases: those outside of Europe and those not offering membership. Only by doing so would it be possible to provide plausible arguments and reasonable recommendations on democracy promotion, avoiding the limitations of generalizing based on exclusively European and "membership-ofTer"-based research. This study has adopted the international socialization framework and adapted it to democracy promotion by widening its scope of applicability, changing its dependent variable to democratic transformation, and introducing new independent variables. Notes 1 mSSSSf1chapler have appeared previously in Babayan 2009 **jn Bai»>™ 2 For more in-depth discussion of consolidation see Schneider 2009. 2 Conditions of democratic transformation, cases, and methods This chapter brings together the theoretical and empirical parts of the book. It further specifies the research questions and hypotheses, and operationalizes variables used in this comparative research on the outcome of democracy promotion and democratization. Afterwards, it elaborates on case selection and describes how the cases are analyzed. This book has derived from an observation which has led to empirical and theoretical puzzles. Since the collapse of the Soviet Union, numerous democracy-promotion activities have been launched with various targets and strategies. However, while some post-communist or authoritarian countries quickly democratized and even launched their own democracy-promotion programs (e.g. the Baltic states. Poland, and the Czech Republic), others are still democratic laggards despite almost two decades of democracy promotion activities (e.g. Armenia, Georgia, and other post-Soviet states). Though on a lawmaking level some countries have at least embraced democratic principles, others have not even bothered with democratic pretence, creating the paradox of long-lasting and widespread democracy-promotion activities and domestic championing of democracy, but scarce spread of liberal democracy on the global level. This discrepancy has triggered the main puzzle of this study: why, despite numerous and extensive democracy promotion initiatives, has democracy not become "the only game in town " even in countries where these initiatives are welcome? Effective democracy promotion entails not only intermediate success or completion of a democracy-promotion project but also an overall improvement in the state of democracy in the target country. Depending upon the country-specific situation, overall improvement in democracy can involve a move to free and fair elections, reduction of political violence during campaigns, increased independence of the judiciary, or a reduction of political corruption. Thus, for example, elections-related projects should not only raise awareness by the voters about their rights or increase the understanding by political parties about the value of free and fair elections, but also ultimately contribute to the conduct of such elections and the peaceful transition of power. On the level of specific sectors targeted by democracy promoters, it can increase their own compliance with their primary functions based on democratic principles. In addition, domestic political actors should not only spcnk but also behave democratically (Linden 2002). There is a need to understand the combination of international and domestic conditions used in promoting democracy in countries without a membership incentive. Based on these considerations, and in an endeavor to develop an analytical framework while understanding the outcome of democracy promotion policies, this study posits the following question: under what conditions is democracy promotion more likely to result in democratic transformation? There is a need to understand the combination of international and domestic conditions used in promoting democracy in countries without EU-membership incentive. To assess the effectiveness of a democracy-promotion policy, this research analyzes different outcomes the policy may have and whether outcomes of the same type of democracy promotion are different or the same depending upon the target sector of democracy promotion. Echoing the lament of the scholarly community that democracy promoters simply transfer their policies and success indicators from country to county, it is important to observe possible changes in policies in the target country in the course of democracy promotion and depending upon specific socio-political and economic developments in the target country. Democratic transformation The dependent variable of this study is democratic transformation as the combined outcome of democracy promotion and democratization. Taking "compliance" as their dependent variable. Schimmelfennig et al. (2006) are only interested in whether or not the country complies with the rule of the international organization, having as an indicator "legal rule adoption" (Schimmelfennig et al. 2006, p. 58). They do not investigate further to see whether the new rule or law is actually enforced because "the Western organizations were generally content with legal adoption and greeted the passing of norm-conforming laws as indications that their demands were fulfilled" (Schimmelfennig et al. 2006, p. 58). However, the examples of many post-Soviet countries show that mere legal adoption of a rule does not guarantee rule-based behavior within any of the sectors of democracy promotion or democratization. In addition, the assessments of democracy promotion projects acknowledge that, even if the rule has been legally adopted, "the challenge, however, lies in converting new formal rules into working rules" (ARD 2002, p. 24). Moreover, the formal adoption of the rule on civil-society level does not guarantee its prevalence also on political or state levels, often due to the weakness of the civil society in democratizing countries. Thus, while "compliance" is useful to understand short-term effects of rule promotion, democratic transformation within domestic political and societal systems as a dependent variable is more useful for understanding the potential overall outcome of democracy promotion, providing a better understanding of its impact. Exclusively, democratic transformation as an outcome is taken as a dependent variable, and not "success" or "effectiveness," as the latter two concepts, though widely used in literature. ;ire too sub|ective and normative. What may be a measurable and an achieved success for the promoter can be a failure from the standpoint of the improvement oi democracy in the target country. In its own turn, democratic transformation cannot be always the same on all levels of promotion, and will often depend upon the implemented policy and domestic conditions. Meanwhile, long-term outcomes and unintended outcomes of democracy promotion are harder to detect since these often remain "unforeseen" and "unforeseeable"' (Morell 2005) and are similar to impact (Kumar 2012). Within literature on evaluating democracy promotion, an outcome is operationalized as "behavioral changes that result from outputs" (Kumar 2012, p. 64), which in their own turn arc the immediate results of democracy-promotion activities. Since democracy promotion does not happen in vacuum, the analyzed outcome is the result of a complex interaction between specific domestic and international factors. The outcome of democracy promotion cannot be measured looking only at democracy-promotion strategies but should also involve analysis of domestic factors. Thus, while still identifying a number of independent variables, my goal here is not to measure direct causal effects but rather to identify the complex interaction by case studies and comparative analysis (Ragin 1997; Wolff et al. 2013). The variation in the outcome of democracy promotion—formal, behavioral, and unintended (Schimmelfennig 2002, p. 9)—is based on the discussion on conceptions of norms (Finnemore and Sikkink 1998) and is often used in the norm diffusion literature (Raymond 1997; Cortell and Davis Jr. 2000; Risse 2013). Though this research partially adopts the terms used in the literature, it operationalizes the outcome of democracy promotion differently in order to show the transition from legal rule adoption to rule-based behavior and to account for unintended consequences of democracy promotion. While the international socialization framework uses a dichotomous categorization of its dependent variable (compliance versus non-compliance), to fully account for possible outcomes this research opts for a trichotomous categorization. At the same time, taking into consideration the implementation process of a policy and customary monitoring and evaluation strategies of promoters, the measurement of democratic transformation involves indicators from two dimensions: promoter and target. This is justified by the fact that often promoters measure the effectiveness of their projects not by domestic political dynamics but according to their checklists of completed activities. In addition, it is important because the outcome of democracy promotion occurs as a result of interaction between promoters and targets. The formal democratic transformation of this study most resembles the compliance variable of Schimmelfennig et al. (2006). Formal democratic transformation from a promoter's perspective occurs upon project completion and the accomplishment of specific project goals, such as training of a specific number of officials or reporters, organization of a number of conferences etc. Within the domestic target realm, formal democratic transformation entails formal adoption of a law (in the case of the state level), of a code of conduct or a specialized law (in ihe ease of political society) or a code of conduct/ethics and media (in the case of the civil society/media). It is less likely to occur on the individual target level of democracy promotion. Knowing that democracy promotion projects do not always proceed smoothly, this book also introduces the possibility of an unintended outcome. Unintended outcome/transformation is understood as an early project shutdown (regardless of reasons), democracy setback, neglect of democratic functions of the target level, or worsened evaluation in comparison with previous assessments. Essentially, an unintended outcome is a negative trend in democratization, which can also entail regress to autocratic tendencies. Behavioral democratic transformation entails the overall improvement of democracy, which is inter alia understood as the conduct of free and fair elections and a free media, and is measured by various democracy indices such as Freedom House, Polity IV, and the Bertelsmann Transformation Index. This category of democratic transformation fully resembles the components of liberal democracy and happens when those are not only theoretically but also practically adhered to. Behavioral democratic transformation is also the most difficult one to achieve because it would mean that democracy has become "the only game in town." In addition, behavioral democratic transformation is measured within each analyzed sector of democracy promotion (elections, party politics, and media freedom) and entails fulfillment of specific functions ascribed by democratic rules. Thus, if elections are not free and fair, it is concluded that behavioral democratic transformation has not happened. Each sector under consideration is assessed based on its specific functions and indicators (see Table 2.1). Following Levitsky and Way (2010, p. 366), elections are considered as unfair in the evidence of any one of the indicators given below. Consequently, if all the indicators are absent, behavioral democratic transformation has happened. 1 At least one major candidate is barred for political reasons; 2 Centrally coordinated or tolerated electoral abuse is asserted by credible sources. Indicators include: serious partisan manipulation of voter rolls; • large-scale voter intimidation or disruption of voting • ballot-box stuffing, multiple voting, or other forms of ballot tampering; falsification of results; 3 Significant formal or informal impediments—coordinated or tolerated by the national government—prevent the opposition from campaigning nationally on reasonably equal footing. Indicators include: • violence against opposition party activists, candidates, or infrastructure; • use or abuse of law regulating public meetings limits the oppositions ability to campaign; ( umlilians, cases. an Promotion, democratization, transformation democratic transformation happens when lite media performs Ks watchdog runt-lion, which is within the powers of the media to choose whether or not to perform. Partially based on definitions of watchdog media given by Waisbord (2000) and Hughes (2006), behavioral democratic transformation within the media sector is registered if media outlets do all of the following: 1 report on the topic of their choosing; 2 critically cover powerful political actors, including the incumbent; 3 do not impose self-censorship; 4 provide equal and unbiased coverage to all election contestants; 5 do not show bias towards the incumbent. The choice of being a watchdog may depend upon other conditions, including economic or physical pressures, state regulations, or imposed censorship, which boil down to freedom of the media. Although freedom of the media does not indicate behavioral democratic transformation within the media sector, it indicates overall democratic improvement in the country, and its absence is a possible reason for not performing the watchdog function. Thus, freedom of the media is also observed and measured by indices of Freedom House, Reporters without Borders, IREX. and the Committee to Protect Journalists, chosen based on their "prominence and longevity" (Annenberg School of Communication 2007, p. 26) in evaluating media freedom. Factors influencing democratic transformation While the initial independent variables of the international socialization framework are investigated, additional international and domestic (independent) variables are added to account for the absence of a membership incentive and enhance the explanatory leverage of the framework. Yet again taking into consideration the importance of the promoter-target interaction, this book looks into the combination of international and domestic independent variables. Independent variables borrowed from the international socialization framework are incentives, credibility, and legitimacy on the international dimension, and costs, identification, and resonance on the domestic dimension. Initially assigned positive and negative values, these variables were also originally categorized as rational (incentives, credibility, and costs) and constructivist (legitimacy, identification, and resonance) (Schimmelfennig et al. 2006, p. 59). However, a dicho-tomous classification of the variables and their sometimes limited operationalization does not allow capturing of the full picture of promoter-target interaction and possible variations in international and domestic factors. Thus, substantial modifications are made, and each of the original and new variables is assigned a low (-). moderate (+), or positive value (++). In addition, more detail is given to the elaboration of variables and one of the variables is assigned a different term merely to fall under the logical reasoning of the methodological approach. The operationalization of the variables is empirically motivated. taking into consideration possible outcomes of democracy-promotion projects rather than the original constructivist rationalist divide, which, though it seems a solid theoretical motivation, does not add to policy-driven research. Relying only on the attractiveness of democracy's virtues would be shortsighted for promoters, inducing them to offer certain incentives for democratic transformation of their targets. Incentives are understood as the type and size of rewards offered to the target country for compliance. While a combination of political and economic incentives (++) are more likely to produce behavioral democratic transformation, economic incentives without the political ones would have only a moderate effect on behavioral change (+). The absence of any incentives would make the chances of a behavioral democratic transformation highly unlikely. Credibility is understood as "the credibility of the threat of withholding the rewards or inflicting punishment in case of non-compliance, and the credibility of the promise to pay the reward or abstain from punishment in case of compliance" (Schimmelfennig et al. 2006, p. 59). The credibility of threats and promises can be of major importance for the outcome of democracy promotion not to create an image of Westerners who "will make a fuss [about violations of democracy] for a few days, and then they will calm down and life will go on as usual" (Shevarnadze in Karumidze et al. 2005, p. 24). Lack of credibility—not fulfilling the threat or promise (-)—is likely to damage the target's reliance on the promoter, making further similar actions useless. Such threats or promises are as a rule made with an expiration date, thus a fulfillment that passed the deadline set by the promoter or a partial fulfillment (+) will not be as negative as in the case of the lack thereof, but would signal the target that it may in future either receive a small reprimand instead of the promised sanction or get a marginal reward despite good performance. In contrast, timely fulfillment of a threat or a promise has the potential of underlining the commitment of the promoter and encouraging similar commitment from the target. Legitimacy refers to the application and observation of promoted rules by the promoter, in order to avoid resistance from the target that may claim the "incompetence" of the promoter. Thus, "the minimal condition is that demands on the target governments must be based on organizational rules rather than ad hoc interests" (Schimmelfennig et al. 2006, p. 59). The legitimacy of the promoter would be low if it shows neither rhetorical nor behavioral commitment within its own territory or structure (-), in other words if it neither talks nor walks the walk of democracy. Merely rhetorical commitment without corresponding behavior would make the promoter moderately legitimate (+) for democracy promotion, while demonstrating both rhetorical and behavioral commitment would also have a positive (++) influence on achieving behavioral democratic transformation. As mentioned, the variable of costs (Schimmelfennig et al. 2006, p. 59) is changed to the variable of utility oj adaptation to harmonize its operationalization with the ones of other variables and reflect the utility for the domestic stakeholders to adapt to democratic rules. This variable is defined as any material or ..-lm i i ft I, iniri.i;iM ifi m i< n physical cosls or Ihe imminence of losing siaius quo by (lie target thai can occur in case of compliance with promoted democracy rules. An imminent threat to the status quo or its loss makes the utility of adaptation to democracy low, as it may inter alia result in a loss of the occupied position (-) if elections are free and fair, in exclusion from governing coalition by the non-complying counterparts, or loss of professional occupation or even physical damages to civil-society members. An uncertain possibility of losing the status quo would make the utility of adaptation moderate (+), while preservation of the status quo would have a positive (++) influence on behavioral democratic transformation. Identification implies the willingness of domestic targets to be associated with the promoter. Identification expressed through rhetoric would be low if domestic stakeholders show commitment to non-democratic principles (-). Uncertainty between non-democratic and democratic principles and the promoter would make the identification moderate (+), while firm commitment to democratic principles and the approval of the promoter would have a positive influence on behavioral democratic transformation (++). However, there is no pretence made by this research of measuring genuine attitudes of local stakeholders towards democracy and democracy promoters. Rather than claiming the vague possibility of measuring ideas and beliefs, this research analyzes rhetorical identification of local stakeholders through their discourse. Thus, this research allows for the possibility that, in the case of strong identification but no behavioral democratic transformation, local stakeholders may simply pay lip-service to democracy. Nevertheless, even such lip-service identification provides better opportunities for democratic transformation than a pronounced opposition to democracy. The final variable of Schimmelfennig et al. (2006) is resonance, which supposes cultural or institutional reflection of the promoted rule with domestic laws, rules, and customs. To avoid arbitrariness, this variable is measured by documented rules and declarations and not by individual attitudes. Presence of opposing concepts makes the resonance of the promoted rule low (-), lack of corresponding concepts and rules makes resonance moderate (+), while presence of such rules and concepts makes resonance positive (++). All these variables are investigated in this book. However, I argue that they are not enough to account for possible changes in the domestic situation of a target and for possible variations of democratic transformation. Thus, this research introduces an additional six variables, which, as with the previous ones, can be distinguished on international-domestic dimensions. However, there is no goal of choosing variables according to the rationalist-constructivist divide. Rather than taking solely theoretical assumptions as the basis for the choice of variables, this research bases its choice of variables, as well as on empirical evidence and actual practice in the field. The newly proposed variables of the international dimension are cooperation, consistency, and involvement. The variables of the domestic dimension are party constellation, a democracy blocker, and ownership. It should be clarified that international variables do not reflect the current international situation but are attributed to the democracy promoter itself, thus are under the full control of the latter. Similarly, domestic variables »............ ..... are associated with the target of democracy promotion, hollowing the setting of the international socialization framework, these variables are also assigned values low, moderate, and positive. However, because the dependent variable is not categorized dichotomously, there is no absolute threshold value for the independent variables. Nevertheless, the highest positive value is assigned to the opcrationalization of the variable that has the strongest potential to result in a behavioral democratic transformation, as this is the ultimate goal of democracy-promotion policies and domestic democratization. Cooperation primarily refers to the joint programs between promoters in a target country and the involvement of at least two sectors of democracy promotion within the framework of one project, e.g. involving the state while the primary target of the project is civil society. As the same area of democracy promotion is often a simultaneous target for democracy promoters, democracy promotion starts being a competition for promoters. This makes the promoters hastily design democracy-promotion programs without close consideration of domestic conditions, thus leading to one-size-fits-all approach. This not only negatively affects the quality of a democracy promotion project itself but also decreases its potential influence on the overall level of democracy in a given country. Cooperation as described above is believed to alleviate the pressure from the state and to tackle reluctance from civil society and individuals, at the same time combining the resources of different promoters for augmented results. Cooperation is low (-) when it is either absent or is present only in the form of annual promoter meetings or dinners, where the yearly activities are presented. Cooperation is moderate (+) in the case of any programmatic joint activities, though targeting only one sector because in this case the promoters assure that their strategies do not conflict (which is sometimes the case in the field). Finally, cooperation is positive (++) and more conducive to behavioral democratic transformation in the case of joint activities targeting more than one sector. Democracy promotion and the practice of individual promoters is characterized by "variance and inconsistency" (Wolff et al. 2013, p. 8). Democracy promoters tend to work on the short-term, without much-needed follow-up to their projects (Zeeuw and Kumar 2006) which is necessary to avoid the usual lip-service of domestic actors. To avoid the stagnation of democratic progress on a lip-service level, democracy promoters need consistency, which entails a follow-up to the completed project with further deepening of the initiative, targeting a particular level and issue. Consistency in democracy promotion is important because "the danger is that donors and financing agencies will declare victory when the formal rules have been modified ... and will not follow through to ensure that they are enforced" (ARD 2002, p. 24). In the case of low consistency (-), the promoter is satisfied with the legal adoption of the rule (formal democratic transformation), especially if that is the final objective of a specific project, and shuts down the project without overseeing whether the rule is actually adhered to (the behavioral democratic transformation). In the case of a moderate consistency (+), the promoter launches a similar project, sometimes changing the implementer of the project and not following-up on the previous activities. In . i r in t if i fi tin ft i i rff contrast, a promoter with positive consistency ( ■ > I lollows-up with the issue and initiates projects or activities that are directed at the actual enforcement of the newly adopted rule, building on previously completed projects Involvement measures the level of presence of the promoter in the issues of national or regional importance, since positive and effective participation in the other issues of a target country (or region) may enhance domestic identification and the promoter's credibility. In addition, such involvement may eliminate outstanding issues that hinder or stagnate democratization. The involvement of the democracy promoter in areas other than democracy promotion is especially important in countries with outstanding conflicts, a low level of economic or human development, or other issues that are internally considered as national priorities. The degree of importance of a specific issue to the domestic politics can be determined through available surveys and political discourse. Involvement is considered low (-) in the case where the promoter limits itself to democracy promotion and related issues, moderate (+) in the case of mostly rhetorical involvement in other issues, and positive (++) in the case where the promoter tries to regularly and substantively engage in other local issues. However, such encompassing involvement may be possible only for large democracy promoters whose agenda can also accommodate additional issues. Smaller promoters can, however, seek cooperation with larger promoters to boost their local presence and give more weight to their democracy-promotion activities. The first domestic variable of party constellation is borrowed from further analysis of Schimmelfennig et al. (2006. p. 245), who argue that -'effective international socialization [democratization] will depend upon the party constellations in the target countries and their respective domestic power costs of compliance," as the course of reforms would depend upon the coalitional government formed by these parties. Party constellation can be distinguished as liberal/positive (++), mixed/moderate (+), and anti-liberal/low (-). Liberal party constellation is understood as liberal-, West- and democracy-oriented programs and declarations of parties with parliamentary seats. By contrast, in an anti-liberal party constellation the party programs and declarations are based on nationalistic, authoritarian, communist, and/or populist grounds. In the case of a mixed constellation, parties do not enjoy consensus on liberal reforms and. despite liberal coalitions, some of them are still in the process of reshaping their authoritarian or communist pasts. Thus, a liberal party constellation is the most conducive for behavioral democratic transformation, especially in cases of party-development projects. Although this variable is argued (Schimmelfennig et al. 2006) to have substantial influence on smooth norm transfer, here it is assumed to play a decreased role in countries with semi-presidential or presidential regimes, where the executive dominates over the parliament. Thus, it is expected that party constellation is rather an intervening variable, which encourages or further discourages democratization depending upon the values of other variables. Democracy promoters are often blamed for unilateral programmatic decisions and for not involving local stakeholders in the development or implementation (. onamons. cases, ana metnoas of democracy-promotion prqjwts. This issue was first addressed in l')W when World Bank introduced its Comprehensive Development framework, emphasizing the importance of country ownership in developmental programs. Thus, another domestic variable to be considered is ownership. It is acknowledged that local ownership of a democracy-promotion program by an authoritarian government may block the advancement of the project. However, exclusion of local stakeholders is unlikely to have positive results either, because reforms would be blocked in any case. On the other hand, the exclusion of civil-society actors from projects developed for them is likely to result in indifference and in accusations of local inapplicability. Often local stakeholders are given limited ownership in project development but not in the implementation and moreover not in the monitoring and evaluation. Accordingly, the absence of any local involvement in the project is considered as a low (-) level of ownership. Limited involvement only in the developmental stage is considered as moderate ownership (+), and involvement or ownership in the course of the development and implementation of a project as positive (++). Following the argument by Whitehead (2005) that successful democratization is possible if there is no major power in the region opposing democracy, another domestic variable of democracy blocker is introduced.' Democracy blocker is understood as a kind of a regional bully that is a powerful authoritarian regional actor and has the potential and is willing to influence—though not always directly—the domestic policy choices of a democracy-promotion target. The potential to influence stems from military, economic, and to some extent cultural interconnectedness with the target country, A mere proximity of a non-democratic neighbor is not considered as an indicator of a democracy blocker. Russia's influence over the post-communist territories may be considered as a sign of democracy blocker, while for example Iran's shared border with Armenia does not influence its political choices. It is important to note that the democracy blocker is regarded as a domestic variable even if it may seem as external to the target country, because it is not in the powers of the promoter to control its presence. Nonetheless, it is in the powers of a promoter to contain a democracy blocker's negative influence on democratization by alleviating the reasons of its influence. However, in the case of a democracy blocker the opposition to democracy is not necessarily outspoken and can be covert, aiming at distancing the target country from the democracy promoter. The absence of such an actor is a positive (++) condition for behavioral democratic transformation, whereas the presence of it with local resistance may moderately (+) hinder democracy promotion. The presence of democracy blocker with local support is likely to lower (-) the chances of the behavioral democratic transformation. In addition, understanding the reasons for support is important to give recommendations on the democracy promoter's involvement in specific areas and possible minimization of the democracy blocker's influence. The dependent variable of democratic transformation is measured separately for each of the target sectors and is expected to have different results. Among Table 2.2 Factors influencing democratic transformation Low (-) Social incentives Incentives (international) Credibility (international) Legitimacy (international) Utility of adaptation (domestic) Identification (domestic) Resonance (domestic) Cooperation (international) Consistency (international) Involvement (international) Party constellation (domestic) Ownership (domestic) Democracy blocker (domestic) Source: author's own compilation. No fulfillment of the threat or promise No rhetorical and no behavioral commitment to democracy Imminent loss of status quo/threat to status quo's survival With non-democratic principles/ against democracy promoter Absent or annual meetings, dinners etc. Moderate (+) Economic but no/almost no political incentives Past-deadline or partial fulfillment of the threat or promise Only rhetorical commitment to democracy High possibility of losing status quo With democratic principles/ democracy promoter and partly with non-democratic countries Lack of/confusion over specific concepts Joint promoter activities targeting only one level Similar project with a different implementer Rhetorical involvement in other issues of national importance Mixed Local involvement either in the development or implementation of the specific project Present with local resistance Positive (++) Economic and political incentives Timely fulfillment of the threat or promise Rhetorical and behavioral commitment to democracy Status quo preservation Only with democratic principles/ democracy promoter Corresponding concepts Joint promoter activities targeting more than one level within the same activity Follow-up to the completed project Other issues of national importance Liberal Local involvement both in development and implementation Absent Closure of the project without follow-up Only democracy promotion Anti-liberal Absent Present with local support Opposing concepts Iromotton, atmocrau anon, wansjormauon live) Mils icscarcli employs iiual ilal i vc research methods of ease \ by choosing a target country where two chosen promoters engage in the same type of democracy promotion within different turgct-sectors. Case studies "provide the intensive empirical analysis that can find previously unnoticed causal factors'" (Achen and Snidal 1989. pp. 167-168). The study analyzes democracy-promotion projects according to sectors and types, from the year of launching democracy promotion (in the early 1990s) until shortly after the latest general elections. Democracy-promotion policies (further categorized as projects) are analyzed by tracking their initial development and further evolution. The selected countries represent states that have legally embraced the fundamental democratic principles and have been in interaction with selected promoters since the initial transition to democracy. The domestic conditions of the target country (from the first "democratic" general elections until shortly after the latest general elections) are also classified within the analytical framework to observe the interaction with the international conditions and to provide a full account of possible effects. Domestic and international changes after the latest general elections will be discussed, albeit without covering their outcome which, according to the adopted framework, is to be revealed after the next elections. To avoid static results and account for the dynamics of democracy promotion, the country-case is viewed beyond the limits of a single observation (Kubicek 2003; Flockhart 2005). Thus, because the observations of this research are based on the interaction of various international and domestic conditions, any change in those leads to a new observation. Following the example of Schimmelfennig et al. (2006), this research uses the methods of within-case analysis—the congruence method, before-after comparison, and process tracing. Cross-case comparison is used in regard to different projects of the promoters (mostly ones targeting the same sector) to account for similarities of their strategies given the same domestic conditions or to pinpoint their differences. Comparisons between promoters and targets of democracy promotion help to explain possible differences between the strategies and effects in various domestic milieus and identify the variables that account for the specific outcome of a certain policy. These methods also provide detailed examinations of cases, which is necessary in identifying missing variables (George and Bennett 2005). Thus, comparison will be made between promoters' policies and strategies, possible outcomes of democracy promotion on different levels of promotion within the country-case by looking .it the state of democracy before promotion programs and after (before-after comparison), and possible outcomes on the levels of democracy promotion (cross-case comparison). According to George and Bennett, when using the congruence method: tlie analyst first ascertains the value of the independent variable in the case ii hand and then asks what prediction or expectation about the outcome of i he dependent variable should follow from the theory. If the outcome of the ease is consistent with the theory s prediction, the analyst can entertain impossibility that a causal relationship may exist. (George and Bennett 2005. p. 181) I lius, the values of a theoretically defined variable are checked against the expected outcomes within the period of observation. The ultimate goal of this method is to find one set of conditions that result in one type of outcome, even if ilu- establishment of a causal link between democracy promotion and "observed i hinges in democracy" (Kumar 2012, p. 4) are difficult to establish. One of the methods for verifying the results of the congruence method is the before after comparison. BecauM the country -case is not treated as a single observation, it is possible to perform the before-after comparison each time one ot the conditions (variables) or the value of the variable changes. This method allows controlling for the possible changes in the outcome and comparing different sets of conditions (independent variables). To further refine the findings and limit the possibility that simultaneous—though not causal—change has occurred in both dependent and independent variables, process-tracing analysis is used. Even if not presented here in its entirety due to space constraints, process-iiaeing helps to identify the links between the independent and dependent variables and contributes to further development of an analytical framework. It also helps in identifying the hypothesized conditions. Since the outcome of democracy promotion is the result of interaction between different variables, process-tracing will help to identify "interacting causal variables" (George and Bennett 2005, p. 212) and understand the link between the international and domestic conditions of democracy promotion and its effect. In addition, it will help to "overcome the dilemmas of small-« research by providing more observation to the implications of a theory" (King et al. 1994, p. 227). The first step is the identification of the current state of the dependent variable in the case studies. With the help of process-tracing, as discussed by George and Bennett (2005), the next step is tracking the design and implementation strategies of democracy-promotion projects and their influence on domestic change by observing the reaction of domestic actors to the conditions of promoters. Process-tracing also helps to understand whether "similarity or variance of the independent variable [democracy-promotion strategy and domestic conditions] leads to different outcomes [in the dependent variable]" (George and Bennett 2005, p. 219). Democracy promoters While a vast amount of literature is available on individual democracy-promotion efforts of specific countries or organizations, limited comparative knowledge is available. Among the reasons for the lack of comparative research has been the alleged incomparability of democracy promoters, given their structural differences as an individual state or an international governmental or non-governmental organization, and the absence of an adequate and generally applicable theoretical nework. The choice of democracy promoters is primarily based on their comparative influence as international actors, as considerable leverage is needed to persuade domestic actors to change their behavior. In addition, the general assumption characterizes them as pursuing different strategies which, however, given the low level of democratic progress, result in similar outcomes. Despite an array of international actors, the EU and the USA are the most prominent democracy promoters (Bumell 2008b). Initially these actors seem to represent two absolutely different and incomparable structures. However, closer consideration shows that "they are two different species of the same political genus" of compound democracy (Fabbrini 2007, p. 3), which is "based on territorial or state cleavages and necessarily function without a government" (Fabbrini 2007, p. 203), and where "decision-making capacity is constrained and limited through the sharing of its resources by distinct institutions" (Fabbrini 2005, p. !90). Similarities and differences between EU and US democracy promotion-policies in comparative perspective are not yet established using a clear analytical framework. The usual perception of the EU is as "one of the most important, if not the most important, normative powers in the world" (Barroso 2007), while the US is seen as a power that for the sake of democracy promotion can "send [its] soldiers, when [they] are needed" (Bush 2002). However, here the actions of the EU and the US are not viewed in isolation or in confrontation, but as complementing each other and in their own way influencing the potential outcome of democracy promotion and democratization processes. The EU promotes democracy to its candidate countries by its powerful instrument of membership incentive. However, it also promotes democracy to countries that do not have membership incentives. The US is one of the largest and most visible democracy promoters that cannot offer membership incentive per se. The case of the US broadens the geographic and structural applicability of the analytical framework, giving an opportunity for its further development. This book analyzes the democracy promotion of USAID—the largest democracy-promotion instrument of the US. Though the US also uses military means in democracy promotion, this project does not take this into consideration as it does not require the consent of domestic actors, thus increasing the probability of non-compliance with promoted principles and rules of democracy. According to the mainstream literature on democracy promotion and foreign policy, these two promoters represent different images of normative (EU) and hard (US) powers; they are supposed to have different policies of democracy promotion though still not always resulting in democratic change in the behavior of domestic actors. However, this study shows that that, despite seeming differences, the strategies of these two actors have started to gradually converge. Targets of democracy promotion The case selection reflects the independent variables of this study and activities by both selected democracy promoters. Thus the selected cases need to have all variables present, with special focus on important variables that can be assigned onautnm. .