Too good to be true? United Nations peacebuilding and the democratization of war-torn states

Abstract

This article examines the effectiveness of UN peacebuilding missions in democratizing war-torn states, emphasizing those missions that include democracy promotion components in their mandates. Based on a multinominal logistic regression, we reveal that democratization is significantly more likely if a UN peacebuilding mission is deployed. Furthermore, regimes categorized as more liberal at the outset have an increased risk of revealing antidemocratization trends over the post-war period. Oil wealth impedes democratization and clear victory of one conflict party makes regime transitions more likely, yet in both directions. Descriptive statistics suggest that an increase in the mission’s capacities may be conducive to democratization.

Keywords

Civil war democracy promotion peacebuilding post-conflict democratization United Nations

Introduction

The United Nations (UN) is consistently the “primary candidate” to intervene in violent conflicts and provide international peace and security (Pushkina, 2006: 133). Beginning with the UNTAG deployment in Namibia in 1989, “peacebuilding” became a leitmotif of UN operations in war-shattered countries, and the blue helmets were assigned additional functions such as organizing elections, repatriating refugees and monitoring human rights (Diehl, 2008: 52).

Democracy promotion is an integral part of the peacebuilding endeavor. The 1990s were dominated by what has been described as a “rosy pro-liberalization rhetoric” (Paris, 2010: 338). According to former UN Secretary-General Boutros Boutros-Ghali (1995: 3), liberal democracy is “one of the pillars on which a more peaceful, more equitable and more secure world can be built”. With the publication of its “Agenda for Democratization”, the UN further consolidated its role in advocating a global culture of democracy, as Boutros-Ghali (1996: 7) describes:

Democracy within states thus fosters the evolution of the social contract upon which lasting peace can be built. In this way, a culture of democracy is fundamentally a culture of peace.

As a consequence, UN peace operations started to include liberal state-building and democracy promotion among their objectives.

However, liberal triumphalism faded with the US regime-change-driven invasions of Afghanistan and Iraq, and the “magic of the market and the ballot box” (Callaghy, 1993: 244) came to be seen as a delusion. Introducing democracy in the wake of war proved to be more complicated than first assumed. Both scholars and practitioners came to challenge the simplistic assumption that all good things—such as peace, security, and democracy—necessarily go hand in hand in post-conflict democratization (Grimm and Leininger, 2012: 392).

Whether UN peacebuilding efforts have contributed to the democratization of war-torn states is still open to debate. To inform this discussion, this article revisits the liberal agenda and critically examines the viability of liberal interventionism in post-civil war countries. In taking up the empirical debate over whether UN interventions promote democratization, we intend to clarify certain apparent contradictions. A number of scholars have asserted that the UN’s post-war endeavors are too ambitious and, to put it simply, too good to be true (Jarstad and Sisk, 2008; Bueno de Mesquita and Downs, 2006; Marten, 2004; Weinstein, 2005); however, our paper adapts a more optimistic perspective. We argue that UN peacebuilding missions can help democratize war-shattered societies, even though long and bloody civil wars leave little fertile ground for democratization to blossom. By focusing exclusively on missions devoted to democratization, by applying a more suitable statistical method, and by enriching our data with more recent evidence, we demonstrate that UN forces can play an essential role in paving the way for democratic transformation. In so doing, we challenge those approaches that fail to recognize the UN’s actual potential to democratize host states because they apply unrealistic standards in the evaluation of democratization (e.g. Bueno de Mesquita and Downs, 2006; Marten, 2004; Weinstein, 2005), focus on exaggeratedly long time periods by using the end of Second World War as a starting date (e.g. Doyle and Sambanis, 2000, 2006), neglect to adequately distinguish between different mission types (Bueno de Mesquita and Downs, 2006; Pickering and Peceny, 2006), and apply statistical procedures (ordinary least squares, OLS, regression) that do not take into account the specific characteristics of the dependent variable. In contrast, we focus exclusively on UN missions with a well-formulated peacebuilding mandate and apply a multinominal logistic regression design, thus relaxing the assumption that democracy measurements operate on continuous scales.

The article is organized into four sections. The first section reveals the contradictory empirical findings with regards to the effectiveness of UN democracy promotion and sets out our theoretical argument. The following section identifies the pillars of UN peacebuilding and reveals their democracy promotion components. The third section introduces the operationalization of the variables, the dataset and the statistical methods guiding the analysis. The subsequent section presents our empirical finding, and the final section discusses the limitations of the paper and highlights areas for future research.

Democratizing or not? Discussing the effects of UN peacebuilding

The scholarly debate has not yet reached a consensus on whether peace operations have had a positive effect on democratization or not. On the one hand, Doyle and Sambanis (2000, 2006) have found that multidimensional peace operations, in contrast with other mission types such as traditional peacekeeping or peace enforcement, were highly and significantly correlated with the level of democracy in target countries, as measured 2 and 5 years after the respective war’s end. The authors compiled an extensive dataset on civil wars and UN involvement that has served as a reference for subsequent studies (such as that of Fortna, 2008).

In line with these findings, a time-series cross-sectional analysis conducted by Pickering and Peceny (2006) reveals that UN missions increase the likelihood of democratization tendencies in host-states by 6.1%, when all other variables are kept constant. Furthermore, the authors find that the blue helmets seem to be more effective in political liberalization endeavors than American, British and French interveners. Similarly, Heldt (2011) assesses the probability of democratic transition after UN peace operations, finding a positive and significant effect. Interestingly, Heldt observes a difference when comparing the impact of all UN missions with those explicitly including democratization elements in their mandates: the latter group’s effect on democracy was 30% higher. The author also examines democratic interventions by non-UN actors, finding a statistically significant negative effect on democratization.

On the other hand, however, a number of studies have reached the opposite conclusion. For example, Bueno de Mesquita and Downs (2006) compare the democratization record of states that have experienced interventions with those that have not. Based on their results, the authors argue that UN interventions have not improved democracy levels; in fact, they claim, such projects have even brought about a moderate decline in democracy, and the “targets of intervention by the UN fare no better and generally do worse than would have been expected had they not suffered an intervention” (Bueno de Mesquita and Downs, 2006: 643). Weinstein (2005: 8) also rules out the hypothesis that outsiders’ attempts can foster democratization abroad, instead favoring the concept of “autonomous recovery”; he presumes that UN interventions have a negative impact on post-war democratization, as they undermine local capacities. In line with this perspective, Marten (2004: 155) states that “the notion of imposing liberal democracy abroad is a pipedream”. Fortna (2008) takes a more moderate stance, suggesting that peace operations are neither good nor bad at promoting democratization. Accordingly, it is rather the case that positive effects (e.g. decreasing the level of mistrust) and negative effects (e.g. undermining indigenous efforts) of peacebuilding cancel each other out. Fortna analyzes democratization trends 1, 2 and 5 years after the war’s termination and further distinguishes between different mission types. However, she could not confirm the positive effect of multidimensional peacekeeping missions reported by Doyle and Sambanis (2006).

A brief sensitivity analysis was conducted to explain variations in the findings of previous studies. First, the above-cited studies applied different timeframes: while the datasets used by Doyle and Sambanis (2006), Pickering and Peceny (2006) and Bueno de Mesquita and Downs (2006) covered a period from 1946 until 1996/2001, Heldt (2011) and Fortna (2004) based their analyses on shorter timeframes, namely from 1960 to 2005 and from 1989 to 1999, respectively. As Fortna (2004) studied a relatively short period of 10 years, her sample size is reduced in comparison to the other studies—which may partly account for the fact that her results remain statistically insignificant. Apart from this, the highly contradictory findings may result from inconsistent methods of operationalizing peace operations. Fortna (2004) and Doyle and Sambanis (2006) categorize UN deployments into (a) observational, (b) enforcing, (c) traditional peacekeeping and (d) multidimensional, and Heldt (2011) specifies which UN operations were based on a democracy-support mandate. Pickering and Peceny (2006) and Bueno de Mesquita and Downs (2006), in contrast, coded UN involvement as a dichotomous variable without qualifying the underlying mandates. It was found that missions with democracy-promotion components (Heldt, 2011) and missions defined as multidimensional (Doyle and Sambanis, 2006) could yield the highest effect sizes with regards to post-war democratization. Similarly, a sensitivity analysis based on our own data revealed that the magnitude of the effect on democratization increased if characteristics of the respective mandate were taken into consideration (relative risk ratio: 6.06 vs 3.59). Lastly, variations in the results of previous studies could emerge from inconsistent coding of the dependent variable: while most studies observed changes in Polity IV rankings over a follow-up period between 1 and 5 years, Bueno de Mesquita and Downs (2006) observe Polity scores 10 years after the intervention, and their results diverge from the other quantitative analyses. In a similar vein, we observed considerably different results in our own data where magnitude as well as direction of the coefficients changed when comparing 5 and 10 year follow-ups.

In the light of this scholarly debate, it seems necessary to further clarify the ambivalent relationship between peacebuilding and democratization. Our analysis represents an empirical improvement on previous studies, in that it applies a fine-grained definition of peacebuilding missions in order to ensure—as advocated by Heldt (2011)—that missions lacking explicit reference to democratization in their mandates are excluded. Furthermore, our data includes evidence on the democratization records of more recent UN deployments, such as the missions in Afghanistan, Iraq, Liberia and Sudan. It should also be noted that some prior analyses focus solely on civil wars, whereas others include interstate warfare as well. Given the fact that civil war became the prevalent mode of warfare after 1945, and also because the concept of peacebuilding is more suitable to intrastate settings, our analysis focuses exclusively on civil wars. Lastly, some of the studies introduced above observe changes in democracy scores while others apply certain thresholds. We have opted to track changes in democratization indices rather than relying on rigid thresholds in order to capture not only decisive regime transformations but also slight changes in democracy rankings, thereby obtaining a more comprehensive picture. We decided to follow Collier and Collier (1991) in analysing the state of democratic change 5 years after the peacekeeping mission (what the authors refer to as a “critical juncture”) as, according to their findings, the mission’s influence on democratization dies out after this time period.

Based on the above considerations, we are especially interested in those cases in which a UN mission directly claimed to address democratization. To shed light on the causal relationship between peacebuilding and democratization in target countries, we argue that UN forces can—through preventing the recurrence of fighting, establishing a functioning political system and strengthening a country’s economy and social fabric—lay the seeds of democracy.

UN peacebuilding and the democratic transformation of post-war countries

The conditions conductive to democratization are usually absent in the aftermath of armed conflict as “the atrocities of civil war are fundamentally antithetical” to democratic norms (Fortna, 2008: 46). Fear and violence are prevalent, the economy is weak, infrastructure destroyed and society traumatized (Grimm, 2008: 535–538). Krasner (2005: 69) advocates that, considering this unfavorable outset, the seeds of democracy can only be laid from outside. Consequently, UN peacebuilding missions play a pivotal role in promoting democracy in the wake of war. Based on a comparative reading of the existing literature on UN peace-building and democracy promotion, we discern three distinct pillars of UN peace operations (a military/security, a political/constitutional and an economic/social pillar) and identify the democracy-promoting components therein (e.g. the promotion of favorable preconditions for democratization, the development of democratic institutions and the building of democratic capacity among domestic actors; Ramsbotham, 2000: 185).

The first pillar comprises security and military components of UN peace operations. The provision of a secure environment and a legitimate monopoly of force is a precondition for democratization (Lyons, 2004: 52). Hence, the initial task of each UN mission is to demobilize and disarm former combatants, thereby following the rationale of substituting military capacity with political and social benefits (Grimm, 2008: 539; Lyons, 2004: 54). The assumption is that former combatants can be reintegrated into society by being provided with financial support, land, education and employment (Russett and Oneal, 2001: 203). Based on these grounds, militias no longer have incentives to resort to the use of force but instead pursue their interests through nonviolent—political—means (Lyons, 2009: 95). A successful disarmament, demobilization and reintegration process can foster confidence-building between conflicting parties and enhance their willingness to engage in political dialogue. The UN followed this approach for example in Liberia, where 100,000 former fighters were demobilized and 28,000 guns collected within one year of the mission’s deployment (Paes, 2005: 253). Hence, the UN can promote democracy by moving conflicts “off the battlefield and into the institutions and processes of politics” (Sisk, 2008: 239).

Apart from that, UN missions may also comprise measures to reform the military and police forces of a country. Former militias can be integrated into a new national army under civilian control (Ramsbotham, 2000: 185). The inclusion of former combatants into a country’s security forces ideally goes along with adequate training in human rights standards and democratic rules (Grimm, 2008: 539).

The second pillar involves the political and constitutional elements of UN peacebuilding missions and seeks to facilitate the formation of domestic democratic institutions and actors’ capacities (Russett and Oneal, 2001: 203). First, UN forces can help to stimulate the emergence of a competitive multiparty system, one of the basic ingredients of each democratic system (Reilly, 2004: 131). In the course of this process, insurgent groups are transformed into political actors vested with sufficient credibility and legitimacy (Manning, 2007: 253; Boutros-Ghali, 1996: 17). Since rebel movements typically lack any form of political experience and are impaired by a shortage of skilled and literate personnel, they do not have the organizational capacity to reach constituents (Carothers, 2011: 86f; Manning, 2001: 165; Zeeuw and Kumar, 2006: 286). By providing financial assistance and expert advice, the UN can help empower these groups and give them a possibility to advocate their interests through political means (Paris, 2002: 644). Emphasizing the significant role of UN post-conflict party assistance towards RENAMO in Mozambique, Manning (2007: 268) states: “democrats are made, not born”.

In addition to that, UN electoral assistance has become a core component of multidimensional peace operations. Measures comprise the planning of election logistics, advice on the electoral system, registration of voters, and the conduct, supervision and verification of elections, for example in Cambodia and East Timor, where UN forces closely monitored the first post-conflict elections (Grimm and Merkel, 2008: 466; Joyner, 1999: 342ff; Ludwig, 2004: 176). Free and fair elections help democracy to gain a first foothold (Heldt, 2007: 7; Lindberg, 2006). By casting their votes and thus choosing a new political leadership, citizens are socialized into the democratic system (Manning, 2007: 270). Elections help build citizens’ trust in the democratic process, establish legitimate institutional frameworks and offer incentives for political leaders to encapsulate their voters’ interests in order to be re-elected. Furthermore, as demonstrated in the cases of Bosnia and Herzegovina and Kosovo, elections symbolically mark the transition “from bullets to ballots” (Gromes, 2007: 123) and allow the struggle for power to continue, albeit through peaceful institutional channels (Knight, 2003: 249). Ramsbotham (2000: 172) denotes this process as “Clausewitz in reverse”, whereby conflict continues, yet with non-military means.

Another important aspect of political democracy promotion is rule of law programs comprising measures such as drafting of laws, retraining of judges, lawyers and bureaucrats, and elevation of judicial budgets. In a political system where the rule of law prevails, laws are clear in meaning and equally valid for each citizen. Established laws allow economic institutions to function in a fair and efficient way (Carothers, 2004: 122f). Further, the second pillar may comprise constitutional engineering, media assistance and anticorruption measures (Durch et al., 2003: 30; Jarstad, 2006: 12; Zeeuw and Kumar, 2006: 12). In Kosovo, for instance, the UN mission has introduced and administered a constitutional framework with the purpose of preparing the new state for self-rule (Tansey, 2009). In Sierra Leone, the UNAMSIL mission has established its own ratio program to promote civic education and mutual understanding among diverse groups (Kumar, 2006: 144).

The third pillar entails social and economic measures. In the immediate aftermath of war, UN forces can provide humanitarian relief and essential services. Displaced persons and refugees are returned, resettled and reintegrated (Russett and Oneal, 2001: 204). This is best exemplified in the case of Mozambique, where the ONUMOZ mission provided humanitarian assistance to resettle 3.7 million displaced people, who had fled in large numbers to neighboring countries South Africa, Zambia and Zimbabwe (West and Kloeck-Jenson, 1999).

Furthermore, peacebuilding operations can contribute to the economic reconstruction of a country by rebuilding infrastructure, reviving agriculture, assisting demining, regulating property rights and providing expert advice for the development of a functioning tax system (Grimm, 2008: 539; Ramsbotham, 2000: 185). Severe economic grievances and humanitarian crises after a prolonged civil war pose a major impediment to popular participation in a democratic process. Also, poverty has been identified as a widespread cause for human rights abuses as well as an obstacle for democratic development (Burnell, 2000: 351). Consequently, peoples’ immediate and basic needs have to be satisfied before they can turn into politically motivated citizens (Russett and Oneal, 2001: 209).

The promotion of human rights can be achieved through monitoring practices and the assistance of non-governmental human rights organizations (Zeuuw and Kumar, 2006, p. 13). Human rights standards can be further consolidated through war crime tribunals, truth commissions and the systematic investigation of past human rights abuses (Russet and Oneal, 2001: 204). These measures can contribute to reconciling former belligerents. In the long term, they may even help to heal the wounds of war and transform the war-ravaged society into a genuine political community (Ramsbotham, 2000: 185). Lastly, the UN may be engaged in community-building and civic education programs (Grimm, 2008: 542; Santiso, 2002: 582). Democracy can be fostered by educating and empowering voluntary associations, women’s groups, religious groups, public policy institutions, trade unions and regional NGOs (Hearn and Robinson, 2000: 245; Heldt, 2007: 7). Giving a political voice to these actors expands the public sphere and strengthens democratic norms and civil rights (Hearn and Robinson, 2000: 242). Moreover, a strong civil society is supposed to foster democracy by propagating citizens’ interests, animating political participation and counterbalancing the state (Carothers, 2004: 102).

In the following, we use the three pillars and the elements described in each of them as a framework for analyzing the mandates of the missions and determining whether the defining features of a peacebuilding mission are met. Prior studies have considered mandates as proxies for peacebuilding practices (Zürcher et al., 2013: 59). The mandate details what peacebuilders are supposed and allowed to do during a mission. It can therefore be assumed that democracy promotion is an integral part of peacebuilding activities in the field if specified accordingly in the respective mandate (Heldt, 2011; Rich, 2004: 65). Vice versa, given that democratization is a complex endeavor that shall trigger substantial institutional and behavioral change, an unintended side-effect of any (minor) peace-building effort on democratization cannot be expected. Instead, it requires targeted and planned substantive action by democracy promoters.

Based on the above elaborations, the following hypothesis is derived and will be tested in the empirical analysis of this paper:

H: A post-civil war country is more likely to experience democratization if a UN peacebuilding mission whose mandate includes democracy promotion is deployed.

Research design

For the empirical analysis, we adopt a large-N approach. In the following section, we describe the variables, data sources, and the statistical model. A detailed description of all variables is presented in Online Appendix¹ 1.

Dependent variable

To operationalize democratization, we use data from Freedom House (2012) (rating countries on a scale from 2 to 14, where 2 indicates the most democratic). Although we are aware of its shortcomings (Collier and Adcock, 1999; Munck and Verkuilen, 2002), the index is valuable because it provides both historical depth and large geographic scope. In addition, the choice is substantiated by findings from Casper and Tufis (2003): testing a democratization model in more than 120 countries based on Freedom House, Polity IV and Polyarchy scores, the authors revealed that most variance was accounted for when using Freedom House data. In order to proxy democratization, we compare Freedom House rankings in the first mission year with those 5 years later. In the absence of a UN peacebuilding mission, we compare rankings from the first year after the war’s end with those 5 years out. The resulting 5 year deltas record the trends in democratization, whereby negative Freedom House Deltas indicate a transition towards a more democratic system. By way of illustration: if a hypothetical country had a Freedom House ranking of 8 at the first time point and 5 at the 5 year follow-up, the resulting delta is −3, thus denoting an increase in democracy. We chose a time period of 5 years as it was the most consensual measure of post-war democratization from previous studies. Longer timeframes (e.g. 10 years) seem inappropriate as the impact of a military intervention on the target country “decays” over time (see Pickering and Peceny, 2006; see also Collier and Collier, 1991). Further, we will check the robustness of our results by estimating an equivalent model with Polity IV data (rating countries on a scale from −10 to 10, where 10 indicates the most democratic; Jaggers and Gurr, 2013) to examine whether the findings are replicated.

Explanatory variable

In “An Agenda for Peace” (United Nations, 1992: 5), peacebuilding is described as “action to identify and support structures which will tend to strengthen and solidify peace in order to avoid relapse into conflict”. However, the concept of peacebuilding still lacks an authoritative definition and interpretations of the term are “tremendously varied” (Call and Cook, 2003: 240). For the purposes of this paper, we apply a narrow operationalization of peacebuilding operations in order to avoid conflation with other mission types, thereby filtering out the genuine influence of UN peacebuilding on democratization.² We exclude all those missions lacking the stated aim of democratizing the host-state, namely observer missions, traditional peacekeeping and peace enforcement missions. Missions’ mandates provide the basis for our case selection (coding scheme presented in Online Appendix 2).

Previous studies (such as those of Bueno de Mesquita and Downs, 2006; Pickering and Peceny, 2006; Fortna, 2008) can be criticized on the basis that the link established between peacebuilding and democratization did not consider the individual characteristics of the UN missions. Indeed, poor mission capacity may thwart democratization, and UN operations are “habitually under-funded, under-equipped and understaffed” (Fortna, 2008: 39). In addition, the timeframes of mission deployments are often criticized as being too short (Ramsbotham, 2000: 179). Consequently, we take into account the number of troops deployed, the budget of the mission, and its duration. We will look at the association between mission capacity and democratization for all identified peacebuilding missions using descriptive statistics, given that the sample size for this subgroup does not allow for inferential statistics. Data on mission endowment was retrieved either from the Department for Peacekeeping Operations (n.d.) homepage or from annual reports provided by the Advisory Committee on Administrative and Budgetary Questions (n.d.).

Control variables

UN peacekeepers are not deployed to post-war countries “at random”. Rather, previous research by Beardsley and Schmidt (2012: 43), Fortna (2008: 61) and Gilligan and Stedman (2003: 51) suggests that UN missions tend to be sent to post-war countries that have suffered the greatest human catastrophes—precisely those countries that lack favorable prerequisites for democratic transformation. Assuming that the costs of war are positively associated with the likelihood of a peacebuilding employment, but negatively associated with the odds for democratization, we can conclude that UN targets are mainly those countries that are more difficult to democratize. Consequently, the effect of a peacebuilding mission would be underestimated if this selection bias was not accounted for. This assumption was tested in a logistic regression where we regressed—in line with Gilligan and Stedman (2003)—the number of battle deaths, the existence of a peace agreement, victory of one conflict side, the democratic tradition of the target country and oil wealth on UN peacebuilding deployment. We found that the odds of a peacebuilding deployment increased significantly with the number of battle deaths (see Table 1). Therefore, we control for the costs of war—proxied by the number of fatalities of the respective civil war (see Fortna, 2004: 494; Gilligan and Stedman, 2003: 44)—hypothesizing that the more people a civil war kills, the less likely a transition to democracy becomes (Fortna, 2008: 61). In so doing, it is more likely to detect a positive effect—if present.

Table 1.

Likelihood of UN peacebuilding deployment

	β	Odds ratio
Battle Deaths	0.25** (0.12)	1.26** (0.15)
Peace Agreement	0.65 (0.55)	1.92 (1.05)
Victory	−0.70 (0.60)	0.50 (0.30)
Oil	−0.75 (0.73)	0.47 (0.35)
Democratic Tradition	−1.53*** (0.60)	0.22*** (0.13)
IMR	0.01 (0.01)	1.01 (0.01)
N	102
McFadden’s pseudo R²	0.18
LR χ²	20.92***

Notes: *p < 0.10, **p < 0.05, ***p < 0.01.

Source: Authors’ calculation.

Further, although coefficients in the logistic regression were not significant, it is conceivable that mission deployment as well as a country’s prospects for democratization are influenced by the outcome of war. Burton and Higley (2010), Hartzell and Hoddie (2007, 2003) and Walter (2002) hypothesize that a country is more likely to experience a liberal transition when its civil war has ended in a negotiated settlement, as such agreements may involve power-sharing components and encourage credible commitments on both sides. In contrast, a clear victory by one party is presumably the least beneficial outcome for democratization (Downes, 2004; Gurses and Mason, 2008: 329). We therefore add two independent variables, one indicating whether a peace agreement has been negotiated and the other capturing whether the war has ended in a clear victory for one side.

In the following, we also consider variables that have been identified as relevant in the basic democratization literature, either as impediments to or facilitators of democratization. First, we take into account a country’s previous democratic experience and the regime type of neighboring states. Regarding the former variable, the history of democratic rule makes a liberal transformation more likely (Przeworski et al., 1996). At the same time, we found in the above logistic regression that the UN is significantly less likely to intervene in countries with a democratic tradition. That is, the overall effect of a UN mission deployment on democratization would again be underestimated if this potential cofounder was not factored in. Secondly, several scholars hypothesize that a democratic neighborhood can foster transitions to democracy by transmitting norms and diffusing pro-democratic ideas (Levitsky and Way, 2006, 2010; Teorell, 2010; and with reference to Europe: Barbé et al., 2009; Börzel, 2011; Lavenex and Schimmelfennig, 2011).

In addition, one of the most prominent and robust findings in democratization research is the assertion that oil impedes democracy (Aslaksen, 2010; Morrison, 2009; Ross, 2001; Tsui, 2011). Accordingly, countries exporting a considerable amount of oil are less likely to implement democratic structures; this is also referred to as the “resource curse” (Humphreys et al., 2013; Koubi et al., 2013; Sachs and Warner, 1995, 2001). We examine data on fuel exports, applying a threshold of at least a one-third share of total export revenues.³ In addition, democracy has often been associated with prosperity: poor socio-economic development is presumed to hinder democratization (Geddes, 1999: 117; Grimm, 2010: 81f; Lipset, 1959: 75; Przeworski et al., 1996: 41).⁴ We use the infant mortality rate (IMR) as a proxy that has proven to be highly correlated with gross domestic product (GDP) per capita, but also better reflects diverse aspects of socio-economic development (Gleditsch et al., 2007; Ulfelder and Lustik, 2007).

Moreover, it is important to control for the base level of democracy in the first mission year or the first post-war year (Bueno de Mesquita and Downs, 2006: 642; Ross, 2001: 393f). Mature and consolidated democracies have no potential to democratize; deltas in such cases are thus zero. Improvements in the indices’ rankings may thus be more easily achieved in countries at the very low end of the democracy scale. We therefore take into account the regime type at the outset of the mission. Lastly, we control for the presence of democratic interveners other than the UN who might account in part for democratization trends (see Heldt, 2011: 48).

Case selection

To select the case, we use the civil war data presented by Doyle and Sambanis (2006: 76ff), which catalogs armed conflicts that caused more than 1000 battle deaths, took place within the recognized boundaries of a sovereign state and involved the government as a principal combatant. The authors apply a threshold of 1000 battle deaths in total rather than per year, as suggested by Correlates of War (COW). Since the dataset only covers civil wars that ended before 2002, we add more recent data compiled from COW (Sarkees and Wayman, 2010) and from the Center for Systemic Peace (Marshall, 2012) while applying the same threshold of 1000 total battle deaths. Thereby, we only include cases of severe violent conflict that were considered by the UN for intervention.⁵ Furthermore, we confine the analysis to civil war data from 1989 onwards, as peacebuilding truly came into its own in the early 1990s. In addition, the international context fundamentally changed with the dissolution of the Soviet Union, which makes data before and after 1989 less comparable. The compiled dataset covers 103 cases (see Online Appendix 3).

Statistical method

Democracy scales cannot be considered a perfect continuum (Gleditsch and Ward, 1997), that is, the intervals between the ordered categories might not be equal.⁶ To illustrate this, a decrease in a Freedom House ranking from 10 to 8 might not necessarily be equal to a decrease on the Freedom House scale from 4 to 2. Therefore, the assumptions for employing mathematical operations on an ordered, categorical variable are violated. The use of OLS regression is consequently inappropriate. Previous studies (e.g. Bueno de Mesquita and Downs, 2006; Fortna, 2008; Pickering and Peceny, 2006) employing this method could be flawed and their results distorted by the use of a model that does not adequately fit the data. Based on these considerations, we have opted to use a statistical model that does not operate on the same rigid assumptions as OLS estimations. We instead apply a multinominal logistic regression,⁷ converting delta scores for each post-war country into the three categories of “decrease” (for a delta score of >0), “no change” (for a delta score of 0) and “increase” (for a delta score of <0) in democracy. Multinominal logistic regressions create probability scores for each category of the dependent variable, compared with one baseline category (here: no change) (Long, 2007: 183). The data analysis was conducted using STATA version 12.1.

Results

Descriptive statistics

To report our findings, we begin with descriptive statistics. Thirty-one of the 103 civil wars in our dataset experienced a UN peacebuilding deployment according to the classification criteria defined above (see Online Appendix 3). Table 2 illustrates the Freedom House rankings at the outset and the Freedom House deltas broken down by UN peacebuilding involvement and no involvement. It is apparent from this table that the majority of post-war countries are categorized as undemocratic in the immediate aftermath of war, meaning that the base level of democracy is in most cases very low. As expressed by the median score for UN target countries, movements towards democracy outweigh movements in the other direction.⁸ The median score for countries left to their own devices indicates no change in democracy. According to Freedom House rankings, the UNMIL operation in Liberia was the most successful regime transformation.⁹ In addition, successful democratization was achieved in Croatia with the UNTAES mission and in Mozambique with the ONUMOZ deployment. These findings are in line with previous research denoting these two operations as particularly successful: ONUMOZ facilitated Mozambique’s transition to multiparty democracy, the country having regularly held presidential and parliamentary elections since the first polling day in 1994 (Bertelsmann Transformation Index, 2012); likewise, UNTAES was able to successfully implement its well-defined mandate, contributing to war crime investigations, economic reconstruction and the establishment of human rights standards, in Croatia (Šimunović, 1999). According to Freedom House, democracy promotion decisively failed in Central Africa with the MINURCA mission and in Haiti with UNSMIH. Deltas captured by Polity IV indicate the same tendencies in the evaluation of democratization in post-war countries. The two indices correlate with ρ = −0.59.

Table 2.

Descriptive statistics: democracy at the outset and deltas

	Median	Range
Freedom House at the outset
Peacebuilding mission (n = 31)	10	6–14
No peacebuilding mission (n = 72)	10	5–14
Δ Freedom House
Peacebuilding mission (n = 31)	−1	−5–5
No peacebuilding mission (n = 71)	0	−7–4

Source: Authors’ calculation.

Table 3 classifies post-war countries by the three categories of “decrease”, “no change” and “increase” in democracy. According to the Freedom House rankings, 58% of UN host-states adopted more democratic forms of government within the 5 years after the mission start, compared with only 36% where the UN was not deployed. This difference is significant (χ²-test p < 0.05). On a cautionary note, as becomes apparent when looking at Table 3, UN peacebuilding does not always go hand in hand with democratization, that is, there are some incidents of antidemocratization trends in UN host states, although outweighed by those in countries left to their own devices.

Table 3.

Contingency table: peacebuilding and democratization

	Decrease	No change	Increase	Mission total
Peacebuilding	6	7	18	31
	19.4%	22.6%	58.0%	100%
No peacebuilding	16	30	26	72
	22.2%	41.7%	36.1%	100%

Source: Authors’ calculation.

Following the above-mentioned theoretical considerations, Table 4 breaks down democratization trends for the security, political and economic pillars, coded according to the respective mandate of the mission. It has to be noted that most mandates list more than one of the above pillars. The political/constitutional pillar is most prevalent in peacebuilding missions. In addition, mandates including a military/security pillar seem to be associated with the largest number of “democratized” cases. However, generalizations should be cautioned against in consideration of the small sample size of 31 peacebuilding missions. Thus, inferences are only tentative. Table 5 suggests that improvements in Freedom House rankings are more prevalent for mission mandates that rely on a combination of several pillars. That is, if emphasis is placed on only one pillar, 50% of the targeted countries reveal positive democratization trends whereas the other 50% of countries experience the reverse. If, in turn, a mandate comprises all three pillars, 67% of the peacebuilding deployments turn out successful in promoting democracy, while none was found to leave behind a country less democratic than at the outset. This finding corroborates the argument put forward by Grimm (2008: 542), according to which the different components of a peacebuilding mandate are intertwined and thus mutually reinforcing.

Table 4.

UN mandate by pillar and democratization

	Pillar
Democracy	Military/security		Political/constitutional		Economic/social
	Frequency	Percentage	Frequency	Percentage	Frequency	Percentage
Decrease	4	19.0	3	11.5	2	11.1
No change	3	14.3	7	26.9	7	38.9
Increase	14	66.7	16	61.6	9	50.0
N (number of mandates)	21	100	26	100	18	100

Source: Authors’ calculation.

Table 5.

Combination of pillars and democratization

	Pillar
Democracy	One pillar		Two pillars		Three pillars
	Frequency	Percentage	Frequency	Percentage	Frequency	Percentage
Decrease	3	50	3	18.8	0	0
No change	0	0	4	25.0	3	33.3
Increase	3	50	9	56.2	6	66.7
N (number of mandates)	6	100	16	100	9	100

Apart from that, Table 6 suggests that increases in democracy rankings tend to be associated with greater capacities of the UN mission. This tendency is most definite for mission duration where the mean for mission duration increases with better democratization outcomes. For the number of troops and the budget of the peacebuilding mission, trends are roughly similar. Mission capacity is higher in countries that experience either no change in democracy rankings or democratization, compared with countries that are exposed to “autocratization” movements. Following this, it can be specified that UN targets receiving more money, hosting more troops and experiencing extended timeframes might do better than countries without such large-scale peacebuilding support. However, it has to be noted that the relationship between higher mission endowments and democratization is not significant in quantitative analyses, which might be partly attributed to the small n of 31. In addition, high mission capacity may reflect extremely challenging and difficult situations on the ground, thereby further compromising the positive effects of greater mission capacity. Therefore, it can be assumed that countries receiving more material and personnel input might also be the most needy and thus those cases where democratization is particularly hard to achieve. Hence, UN peacebuilding missions must be adequately equipped, funded, and well-staffed so as to mitigate the hardship in post-war countries, especially given that UN target countries are likely to have experienced the most severe conflicts, as we have argued above.

Table 6.

Mission capacity and democratization

	Capacity
Democracy	Budget (in billion US$)			Troops			Duration (in years)
	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range
Decrease	39.7	32.9	6.3–80	5138.33	5712.87	876–15,991	2.33	1.21	1– 4
No change	227.0	216	65.0–576	7009.14	7036.25	1090–21,350	5.57	4.43	<1–13
Increase	185.0	247.0	5.3–873	6754.22	6494.54	90–22,016	6.44	4.85	1–17

Source: Authors’ calculation.

While the descriptive statistics in this section corroborate the hypothesis that UN interventions can promote democratization, factors other than peacebuilding deployment have thus far been neglected. Peacebuilding missions are not deployed at random; rather, they tend to be sent to countries where the domestic circumstances are particularly unfavorable for democratization. The following multivariate analysis therefore takes into account these various factors.

Multinominal logistic regression

The findings of the multinominal logistic regression are presented in Table 7. The findings obtained from the analysis indicate that UN peacebuilding is positively associated with post-war democratization. Freedom House rankings denote that an increase in democracy (compared with no change) is significantly more likely where a UN peacebuilding mission is deployed. As shown in model 3, the presence of a peacebuilding mission raises the multinominal log-odds for an increase in democracy relative to no change by 1.801. In more comprehensible terms, when all other variables are held at their means, the probability of an increase on the democracy scale stands at 68% for countries hosting a UN peacebuilding mission, almost twice as high as for post-war countries without UN peacebuilding involvement. Further, the model predicts that a hypothetical country with all of the least favorable characteristics (i.e. IMR and battle deaths at the 95th percentile, favorable domestic conditions such as democratic tradition and democratic neighborhood at 0, oil reliance at 1) would still have a 20% chance of democratization (compared with no change) after a UN peacebuilding intervention, compared with only 7% without UN involvement. Assuming a particularly favorable setting, with IMR and battle deaths at the 5th percentile and conducive domestic circumstances, the likelihood for democratic transition turns out to be 84% with a UN deployment and 58% without. The predicted probabilities for each outcome of the dependent variable (decrease, no change, increase) are displayed in Table 8, broken down by UN peacebuilding deployment.

Table 7.

Multinominal logistic regression coefficients to estimate the impact of UN peacebuilding on democratization

	Model 1		Model 2		Model 3
	Decrease	Increase	Decrease	Increase	Decrease	Increase
UN Peacebuilding	1.001 (0.73)	1.449** (0.58)	0.919 (0.75)	1.502** (0.59)	1.124 (0.82)	1.801*** (0.66)
Battle Deaths	0.026 (0.14)	−0.118 (0.11)			0.033 (0.16)	−0.120 (0.12)
Peace Agreement	0.725 (0.69)	0.382 (0.56)			0.885 (0.75)	0.605 (0.60)
Victory	1.022 (0.69)	0.704 (0.56)			1.353* 0.74	1.000* (0.60)
Oil			0.246 (0.78)	−1.274* (0.68)	0.106 (0.82)	−1.154* (0.70)
Infant Mortality Rate			0.005 (0.01)	0.002 (0.01)	0.004 (0.01)	0.001 (0.01)
Democratic Neighbor			−0.035 (0.63)	0.348 (0.52)	−0.058 (0.65)	0.406 (0.53)
Democratic Tradition			−0.285 (0.68)	0.290 (0.54)	−0.455 (0.73)	0.355 (0.56)
Regime Type	1.349*** (0.50)	0.109 (0.41)	1.610*** (0.56)	0.003 (0.48)	1.624*** (0.60)	−0.086 (0.49)
Non-UN Intervention	1.298 (0.97)	1.048 (0.79)	−0.264 (0.57)	1.113 (0.80)	1.432 (1.00)	1.169 (0.82)
N	102		102		102
McFadden’s pseudo R²	0.09		0.11		0.14
LR χ²	20.91*		24.11**		29.82*

Notes: *p < 0.10, **p < 0.05, ***p < 0.01.

The reference category is no change in democratization. Standard errors of the respective regression coefficients are displayed in parentheses.

Model 1 focuses on war characteristics and model 2 on the domestic setting; model 3 includes all variables.

Diagnostics: for the purpose of detecting outliers or influential data points, we ran separate logit models and used the diagnostics tools on each model. No residual stands out as being extremely large or influential.

Table 8.

Predicted probabilities for democratization by peacebuilding deployment

	UN peacebuilding deployment	No UN peacebuilding deployment
5-Year change on Freedom House Scale	Adjusted predicted probability	Adjusted predicted probability
Autocratization	0.19 (95% CI 0.29, 0 .34)	0.22 (95% CI 0.12, 0 .33)
No change	0.14 (95% CI 0.01, 0.27)	0.43 (95% CI 0.31, 0 .56)
Democratization	0.68 (95% CI 0.49, 0.87)	0.34 (95% CI 0.22, 0 .47)

Note: all control variables are held at their means.

Source: Authors’ calculation.

In addition, there is a clear-cut relationship between a country’s regime type at the outset and the prospects for democratization. A country that is ranked more liberal at the first time point is significantly more likely to experience a decrease in its democracy ranking within a 5 year post-war period. The effect is most salient in model 3, where we would expect the multinominal log-odds of a movement toward a more autocratic regime to be increased by 1.624—when compared with no changes on the Freedom House scale for a one-unit increase in regime type (e.g. from not free to partly free). That is, democratic nations are more likely to experience some transition towards more illiberal regime types in the aftermath of a civil war. In addition, this finding implies that there is some kind of saturation effect at work: the more liberal a country is at the outset, the less likely a further move towards democracy on an index’s continuum becomes. One might therefore reason that UN peacebuilding missions are especially needed in former democratic countries so as to prevent “autocratization”.

Turning to the impact of war characteristics, the picture is less clear. The β-coefficients highlighted in model 1 reveal that neither the outcome nor the costs of civil war significantly alter prospects for democratization. For higher costs of war, Freedom House estimations point in the hypothesized direction but are not statistically significant. In addition, the analysis does not support the conventional wisdom that democracy emerges as a consequence of a negotiated agreement. Lastly, the influence of a clear victory is ambiguous: model 3 displays significant positive effects on both an increase and a decrease in democracy. It is conceivable that this pattern relates to the political claims of the dominant conflict party. Where a victory is achieved by a pro-democratic force, democratization becomes more likely. In contrast, if the victorious conflict party is antidemocratic, the inverse link becomes more likely. Yet, this analysis does not allow for a definite conclusion in this respect, as conflict parties were not classified according to their political agenda. Future research is needed to analyze this relationship in more depth.

Concerning the influence of domestic factors (model 2), the pattern is equally unclear. With regard to the general expectation that socio-economic development promotes democratization, the data tell a different story: the IMR variable is statistically negligible in both estimations. As a robustness check, the IMR measurement was replaced with GDP per capita, but standardized coefficients indicated an equally small and insignificant effect. This result might validate the claims of Limongi and Przeworski (1997: 177), who found that socioeconomic development does not determine the emergence of democracy, but rather increases the chances of the survival of democracy once it has been established. Further, although a democratic neighborhood displays a positive coefficient for increases in democracy and a negative coefficient for decreases, the confidence interval is too large to allow any interpretation of the effect with sufficient confidence. The same holds true for democratic tradition where the effects point in the hypothesized direction but fail to reach significance. Although the overall impact of domestic factors remains rather doubtful, the effect of oil is robust, substantiating the popular theory that resource wealth inhibits a country’s prospects for democratization. In model 3, oil emerges as significant predictor, indicating a 1.154 decrease in the ordered log-odds of experiencing democratization, ceteris paribus. Holding all other variables at their mean values, the likelihood for democratization decreases by roughly 50% for countries relying heavily on oil exports in comparison to countries not affected by the “resource curse”.

Finally, we turn to the influence of non-UN interveners on post-war regime transformation. The estimation uncovers an inconsistent impact of non-UN actors. Although there is no definite relationship, the findings strengthen the presumption that the UN is more effective at promoting democratization than other interveners

If we disaggregate the dependent variable according to both components of the Freedom House index, namely civil liberties and political rights, we reveal a positive and significant impact of UN peacebuilding missions on both dimensions (see Table 9). The negative impact of oil on the prospects for democratization only holds true for political rights, similarly, regime characteristics at the outset apparently only play a significant role for the aspect of political rights. A higher number of battle-related deaths was linked to significant decreases in the odds of democratization for the civil liberties dimension. Surprisingly, we found that UN peacebuilding was also associated with significantly higher odds of a decrease in democratization, yet significance was borderline.

Table 9.

Multinominal logistic regression coefficients to estimate the impact of UN peacebuilding on the disaggregated freedom house index: civil liberties and political rights

	Civil liberties		Political rights
	Decrease	Increase	Decrease	Increase
UN Peacebuilding	1.697* (0.86)	1.558*** (0 .61)	0.236 (0.80)	1.455** (0.62)
Battle Deaths	−0.106 (0.19)	−0.205* (0.12)	0.030 (0.15)	−0.064 (0.12)
Peace Agreement	0.419 (0.84)	−0.094 (0.56)	0.776 (0.72)	0.196 (0.58)
Victory	1.406 (0.87)	0.546 (0.55)	0.857 (0.71)	0.935 (0.58)
Oil	1.190 (0.96)	−0.298 (0.68)	−0.032 (0.82)	−1.673** (0.78)
Infant Mortality Rate	0.001 (0.01)	−0.001 (0.01)	0.009 (0.01)	−0.003 (0.01)
Democratic Neighbor	0.050 (0.803)	0.182 (0.51)	−0.646 (0.64)	−0.160 (0.52)
Democratic Tradition	−1.959 (1.20)	0.228 (0.54)	0.660 (0.67)	0.691 (0.58)
Regime Type	0.869 (0.73)	−0.566 (0.45)	1.374** (0.55)	−0.139 (0.49)
Non-UN Intervention	−13.518 ( 858.07)	0.469 (0.69)	1.209 (0.98)	1.107 (0.82)
N	101		102
McFadden’s pseudo R²	0.13		0.15
LR χ²	25.82		31.33**

The likelihood-ratio chi-square statistic indicates that the overall model is statistically significant, in comparison to the null model with no predictors. As a further robustness check, we estimated the model with Polity IV data. While signs and statistical significance of coefficients changed for some control variables, the effect of UN peacebuilding deployment on the likelihood of democratic transformation (compared with no change) remained significant and positive when democratization measures were based on Polity IV scores instead of Freedom House (see Online Appendix 5). Therefore, the results of our analysis can be interpreted with sufficient confidence. When comparing models 1–3, it becomes apparent that the effects generally point in the same direction, while the magnitude increases with controlling for additional variables. Since the analysis allows no indication of the impact of domestic factors (except for oil), we must conclude that structural theories are not very suitable for explaining short-term (eruptive) democratization (see also Teorell, 2010: 10). That is, although socio-economic development, democratic tradition and a democratic neighborhood are strong predictors for the general level of democracy/autocracy, they are not adequate to track movements towards and away from democracy after a civil war, as captured here with the 5-year delta rankings.

Conclusion

We investigated the relationship between United Nations peacebuilding missions and democratization trends in post-civil war countries. Based on statistical analyses, we attempted to clarify the link between peacebuilding and democratization since 1989 by examining 103 post-civil war countries, among them 31 that were subject to a UN peacebuilding mission. Overall, the analysis depicts a positive and significant connection between the two concepts. For both Freedom House and Polity IV data, we reveal an increased likelihood of democratic transformation, compared with no change, where the UN is deployed with a peacebuilding mandate that includes democracy-promoting components. Further, oil dependence was found to significantly decrease the likelihood of democratization success and a more liberal regime type at the outset turned out to facilitate movements away from democracy. A clear victory of one conflict party can result in both decreases and increases of democracy, presumably depending on the party’s political stance. Other characteristics of the civil war, domestic factors and mission capacity displayed no statistically validated impact on post-war transitions to democracy. However, as illustrated by descriptive statistics, there is moderate evidence for a positive relationship between greater UN mission capacity and better outcomes in a target country’s democratic track record.

In drawing inferences from the results of the analysis, a number of limitations need to be considered. Given the multidimensional character of democratization, we are aware of the fact that it is problematic to translate democracy into a common metric such as the Freedom House scale (Munck and Verkuilen, 2002: 23). The index has been repeatedly criticized for its non-transparent coding scheme. Moreover, Freedom House has revised its criteria several times, and comparisons of rankings over time may be thus distorted (Bernhagen, 2009: 34f). Yet, despite the ambiguities and shortcomings associated with measuring democracy, we agree with Pickering and Peceny that the said democracy indices can still provide a “rough sense of movement from more autocratic to more liberal political structures” (Pickering and Peceny, 2006: 545). Substantially, a little “more democratic” might solely mean a little “less authoritarian”, to follow what Levitzky and Way (2010) call “electoral authoritarianism”.

While our paper provides evidence of a positive association between peacebuilding and democratization, it was beyond the scope of this study to further explore the causal chain running from UN mission deployment to political transformation. Future research should therefore focus on mediators using time series analyses in order to determine the structural link between the two. Furthermore, case analyses might help to fully capture the causal mechanisms between UN democracy promotion and regime transformation, clarifying why some UN targets accomplish successful democratic transitions while others fail to make progress or even deteriorate.

In the end we must keep in mind that the liberal blueprint is is not a panacea for democratic ills. In certain cases, democracy promotion and ambitious electoral programs might even do more harm than good. Illustrative examples are Angola and the Democratic Republic of Congo, where elections incited the recurrence of fighting. However, although mission mandates sometimes seem unrealistically ambitious, our study has demonstrated that UN peacebuilding interventions can give rise to a flourishing democracy and help encourage post-war populations to embrace democratic institutions, thus paving the way for sustainable peace. UN peacebuilding efforts in Mozambique, Croatia, Burundi and Liberia have exemplarily demonstrated how the seeds of democracy can be sown from the outside and how war-torn countries can be supported on the path towards democracy. UN missions might not be able to perform miracles. However, we should allow ourselves to be optimistic enough to acknowledge that the blue helmets can help post-war countries—particularly those most in danger of relapsing into civil war, authoritarian rule, or state fragility—become (semi-)democratic and stable.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Notes

References

Advisory Committee on Administrative and Budgetary Questions (n.d.) UN ACABQ. Available at: http://www.un.org/ga/acabq/secretariat.asp.html (accessed 3 March 2014).

Allen

Yuen

(2013) The politics of peacekeeping: UN Security Council oversight across peacekeeping missions. International Studies Quarterly (OnlineFirst): 1–12.

Aslaksen

(2010) Oil and democracy: More than a cross-country correlation? Journal of Peace Research 47(4): 421–431.

Barbé

Costa

Herranz

Johansson-Nogués

Natorski

Sabiote

(2009) Drawing the neighbours closer… to what? Explaining emerging patterns of policy convergence between the EU and its neighbours. Cooperation and Conflict 44(4): 378–399.

Beardsley

Schmidt

(2012) Following the flag or following the charter? Examining the determinants of UN involvement in international crises, 1945–20021. International Studies Quarterly 56(1): 33–49.

Bernhagen

(2009) Measuring democracy and democratization. In: Haerpfer

Bernhagen

Inglehart

Welzel

(eds) Democratization. Oxford: Oxford University Press, pp. 24–41.

Bertelsmann Transformation Index (2012) Mozambique Country Report. BTI. Available at: http://www.bti-project.de/laendergutachten/esa/moz/2012/ (accessed 10 March 2014).

Bollen

Jackman

(1989) Democracy, stability, and dichotomies. American Sociological Review 54(4): 612–621.

Börzel

(2011) When Europe hits … beyond its borders: Europeanization and the near abroad. Comparative European Politics 9(4): 394–413.

10.

Boutros-Ghali

(1995) Democracy: A newly recognized imperative. Global Governance 1(1): 3.

11.

Boutros-Ghali

(1996) An Agenda for Democratization. New York: United Nations.

12.

Bueno de Mesquita

Downs

(2006) Intervention and democracy. International Organization 60(3): 627.

13.

Burnell

(2000) Democracy assistance: The state of the art. In: Burnell

(ed.) Democracy Assistance: International Cooperation for Democratization. New York: Routledge, pp. 339–361.

14.

Burton

Higley

(2010) Elite settlements. American Sociological Review 52(3): 295–307.

15.

Call

Cook

(2003) On democratization and peacebuilding. Global Governance 9(2): 233–246.

16.

Callaghy

(1993) Vision and politics in the transformation of global political economy: Lessons from the Second and Third Worlds. In: Slater

Schutz

Dorr

(eds) Global Transformation and the Third World. Boulder, CO: Lynne Rienner, p. 244.

17.

Carothers

(2011) Aiding Democracy Abroad: The Learning Curve. Washington, DC: Carnegie Endowment for Peace.

18.

Carothers

(2004) Critical Mission: Essays on Democracy Promotion. Washington, DC: Carnegie Endowment for Peace.

19.

Casper

Tufis

(2002) Correlation versus interchangeability: The limited robustness of empirical findings on democracy using highly correlated datasets. Political Analysis 11(2): 196–203.

20.

Collier

Adcock

(1999) Democracy and dichotomies: A pragmatic approach to choices about concepts. Annual Review of Political Science 2(1): 537–565.

21.

Collier

(1991) Critical Junctures and Historical Legacies. Shaping the Political Arena: Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Princeton, NJ: Princeton University Press.

22.

Coppedge

Reinicke

(1990) Measuring polyarchy. Studies in Comparative International Development 25(1): 51–72.

23.

Dahl

(1971) Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press.

24.

Department for Peacekeeping Operations (n.d.) Peacekeeping Operations Timeline. UN DPKO. Available at: https://www.un.org/en/peacekeeping/documents/operationslist.pdf (accessed 17 March 2014).

25.

Diehl

(2008) Peace Operations. Malden: Polity Press.

26.

Downes

(2004) The problem with negotiated settlements to ethnic civil wars. Security Studies 13(4): 230–279.

27.

Doyle

Sambanis

(2000) International peacebuilding: A theoretical and quantitative analysis. American Political Science Review 94(4): 779–801.

28.

Doyle

Sambanis

(2006) Making War and Building Peace: United Nations Peace Operations. Princeton, NJ: Princeton University Press.

29.

Durch

Holt

Earle

Shanahan

(2003) The Brahimi Report and the Future of UN Peace Operations. Washington, DC: Henry L. Stimson Center.

30.

Fearon

Laitin

(2003) Ethnicity, insurgency, and civil war. American Political Science Review 97(1): 75–90.

31.

Fortna

(2004) Interstate peacekeeping: Causal mechanisms and empirical effects. World Politics 56(4): 481–519.

32.

Fortna

(2008) Peacekeeping and democratization. In: Jarstad

Sisk

(eds) From War to Democracy: Dilemmas of Peacebuilding. Cambridge: Cambridge University Press, pp. 39–80.

33.

Freedom House (2012) Freedom in the World. Country Rankings: 1972–2012. Freedom House. Available at: http://www.freedomhouse.org/report-types/freedom-world (accessed 3 March 2014).

34.

Gartzke

D-J

(2002) The Affinity of Nations Index, 1946–1996. Available at: http://dss.ucsd.edu/~egartzke/data/affinity01242010.zip

35.

Geddes

(1999) What do we know about democratization after twenty years?. Annual Review of Political Science 2(1): 115–144.

36.

Gilligan

Stedman

(2003) Where do the peacekeepers go? International Studies Review 5(4): 37–54.

37.

Gleditsch

Ward

(1997) Double take: A reexamination of democracy and autocracy in modern polities. Journal of Conflict Resolution 41(3): 361–383.

38.

Gleditsch

Christiansen

Hegre

(2007) Democratic jihad?: Military intervention and democracy. Policy Research Working Paper 4242: 43–46.

39.

Grimm

(2008) External democratization after war: Success and failure. Democratization 15(3): 525–549.

40.

Grimm

(2010) Erzwungene Demokratie: politische Neuordnung nach militaärischer Intervention unter externer Aufsicht. Baden-Baden: Nomos.

41.

Grimm

Leininger

(2012) Not all good things go together: Conflicting objectives in democracy promotion. Democratization 19(3): 391–414.

42.

Grimm

Merkel

(2008) War and democratization: Legality, legitimacy and effectiveness. Democratization 15(3): 457–471.

43.

Gromes

(2007) Demokratisierung nach Bürgerkriegen: Das Beispiel Bosnien und Herzegowina. Frankfurt am Main: Campus.

44.

Gurses

Mason

(2008) Democracy out of anarchy: The prospects for post-civil-war democracy. Social Science Quarterly 89(2): 315–336.

45.

Hartzell

Hoddie

(2003) Institutionalizing peace: Power sharing and post-civil war conflict management. American Journal of Political Science 47(2): 318–332.

46.

Hartzell

Hoddie

(2007) Crafting Peace: Power-sharing Institutions and the Negotiated Settlement of Civil Wars. Philadelphia, PA: Penn State University Press.

47.

Hearn

Robinson

(2000) Civil society and democracy assistance in Africa. In: Burnell

(ed.) Democracy Assistance: International Co-operation for Democratization. New York: Routledge, pp. 241–263.

48.

Heldt

(2007) International Peacekeeping: The UN versus Regional Organizations. Center for International Development and Conflict Management, University of Maryland.

49.

Heldt

(2011) Peacekeeping and transitions to democracy. In: Fjelde

Höglund

(eds) Building Peace, Creating Conflict?: Conflictual Dimensions of Local and International Peacebuilding. Lund: Nordic Academic Press, pp. 47–73.

50.

Humphreys

Sachs

Stiglitz

(2013) Escaping the Resource Curse. New York: Columbia University Press.

51.

Jaggers

Gurr

(2013). Polity IV Project: Political Regime Characteristics and Transitions, 1800–2011. Polity IV Project. Available at: http://www.systemicpeace.org/polity/polity4.htm (accessed 3 March 2014).

52.

Jarstad

(2006) Dilemmas of war-to-democracy transitions. Paper prepared for presentation at the Conference State, Conflict and Democracy, 12–13 May, Lund University.

53.

Jarstad

Sisk

(2008) From War to Democracy: Dilemmas of Peace-building. Cambridge: Cambridge University Press.

54.

Joyner

(1999) The United Nations and democracy. Global Governance 5(3): 333–357.

55.

Knight

(2003) Evaluating recent trends in peacebuilding research. International Relations of the Asia-Pacific 3(2): 241–264.

56.

Koubi

Spilker

Böhmelt

Bernauer

(2013) Do natural resources matter for interstate and intrastate armed conflict? Journal of Peace Research 21(3): 1–17.

57.

Krasner

(2005) The case for shared sovereignty. Journal of Democracy 16(1): 69–83.

58.

Kumar

(2006) Promoting Independent Media: Strategies for Democracy Assistance. Boulder, CO: Lynne Rienner.

59.

Lavenex

Schimmelfennig

(2011) EU democracy promotion in the neighbourhood: From leverage to governance? Democratization 18(4): 885–909.

60.

Levitsky

Way

(2006) Linkage versus leverage. Rethinking the international dimension of regime change. Comparative Politics 38(4): 379–400.

61.

Levitsky

Way

(2010) Competitive Authoritarianism: Hybrid Regimes after the Cold War. Cambridge: Cambridge University Press.

62.

Limongi

Przeworski

(1997) Modernization: Theories and facts. World Politics 49(2): 155–183.

63.

Lindberg

(2006) The surprising significance of African elections. Journal of Democracy 17(1): 139–151.

64.

Lipset

(1959) Some social requisites of democracy: Economic development and political legitimacy. American Political Science Review 53(1): 69–105.

65.

Long

(2007) Regression Models for Categorical and Limited Dependent Variables. College Station, TX: StataCorp LP.

66.

Ludwig

(2004) The UN’s electoral assistance: Challenges, accomplishments, prospects. In: Newman

Rich

(eds) The UN Role In Promoting Democracy: Between Ideals And Reality. Tokyo: United Nations University Press, pp. 169–188.

67.

Lyons

(2004) Post-conflict elections and the process of demilitarizing politics: The role of electoral administration. Democratization 11(3): 36–62.

68.

Lyons

(2009) Peacebuilding, democratization, and transforming the institutions of war. In: Dayton

Kriesberg

(eds) Conflict Transformation and Peacebuilding: Moving From Violence to Sustainable Peace. New York: Routledge, pp. 91–107.

69.

Manning

(2001) Competition and accommodation in post-conflict democracy: The case of Mozambique. Democratization 8(2): 140–168.

70.

Manning

(2007) Party-building on the heels of war: El Salvador, Bosnia, Kosovo and Mozambique. Democratisation 14(2): 253–272.

71.

Marshall

(2012). Major Episodes of Political Violence 1946–2012. Center for Systemic Peace. Available at: http://www.systemicpeace.org/warlist.htm (accessed 3 March 2014).

72.

Marten

(2004) Enforcing the Peace: Learning from the Imperial Past. New York: Columbia University Press.

73.

Morrison

(2009) Oil, nontax revenue, and the redistributional foundations of regime stability. International Organization 36(1): 107–138.

74.

Munck

Verkuilen

(2002) Conceptualizing and measuring democracy: Evaluating alternative indices. Comparative Political Studies 35(1): 5–34.

75.

Paes

W-C

(2005) The challenges of disarmament, demobilization and reintegration in Liberia. International Peacekeeping 12(2): 253–261

76.

Paris

(2002) International peacebuilding and the “mission civilisatrice”. Review of International Studies 28(4): 637–656.

77.

Paris

(2010) Saving liberal peacebuilding. Review of International Studies 36(2): 337–365.

78.

Pickering

Peceny

(2006) Forging democracy at gunpoint. International Studies Quarterly 50(3): 539–560.

79.

Przeworski

Alvarez

Cheibub

Limongi

(1996) What makes democracies endure? Journal of Democracy 7(1): 39–55.

80.

Pushkina

(2006) A recipe for success? Ingredients of a successful peacekeeping mission. International Peacekeeping 13(2): 133–149.

81.

Ramsbotham

(2000) Reflections on UN post-settlement peacebuilding. International Peacekeeping 7(1): 169–189.

82.

Reilly

(2004) Elections in post-conflict societies. In: Newman

Rich

(eds) The UN Role in Promoting Democracy: Between Ideals and Reality. Tokyo: United Nations University Press, pp. 113–135.

83.

Rich

(2004) Crafting Security Council mandates. In: Newman

Rich

(eds) The UN Role In Promoting Democracy: Between Ideals And Reality. Tokyo: United Nations University Press, pp. 169–188.

84.

Ross

(2001) Does oil hinder democracy? World Politics 53(3): 325–361.

85.

Russett

Oneal

(2001) Triangulating peace: Democracy, Interdependence, and International Organizations. New York: Norton.

86.

Sachs

Warner

(2001) The curse of natural resources. European Economic Review 45(4–6): 827–838.

87.

Sachs

Warner

(1995) Natural resource abundance and economic growth. National Bureau of Economic Research Working Paper Series (5398): 1–47.

88.

Santiso

(2002) Promoting democratic governance and preventing the recurrence of conflict: The role of the United Nations development programme in post-conflict peace-building. Journal of Latin American Studies 34(3): 555–586.

89.

Sarkees

Wayman

(2010) Resort to War: 1816–2007. Correlates of War Project. Available at: http://www.correlatesofwar.org/ (accessed 13 March 2014).

90.

Sartori

(1987) The Theory of Democracy Revisited. Chatham, NJ: Chatham House.

91.

Sartori

(1991) Comparing and miscomparing. Journal of Theoretical Politics 3(3): 243–257.

92.

Šimunović

(1999) A framework for success: Contextual factors in the UNTAES operation in Eastern Slavonia. International Peacekeeping 6(1): 126–142.

93.

Sisk

(2008) Peacebuilding as democratization: Findings and recommendations. In: Jarstad

Sisk

(eds) From War to Democracy: Dilemmas of Peacebuilding. Cambridge: Cambridge University Press, pp. 239–260.

94.

Tansey

(2009) Regime-building: Democratization and International Administration: Democratization and International Administration. Oxford: Oxford University Press.

95.

Teorell

(2010) Determinants of Democratization. Explaining Regime Change in the World, 1972-2006. Cambridge: Cambridge University Press.

96.

Tsui

(2011) More oil, less democracy: Evidence from worldwide crude oil discoveries. Economic Journal 121 (551): 89–115.

97.

Ulfelder

Lustik

(2007) Modelling transitions to and from democracy. Democratisation 14(3): 351–387.

98.

United Nations (1992) An Agenda for Peace. New York: United Nations.

99.

Walter

(2002) Committing to Peace: The Successful Settlement of Civil Wars. Princeton, NJ: Princeton University Press.

100.

Weinstein

(2005) Democracy Advantage. How Democracies Promote Prosperity and Peace. New York: Routledge.

101.

West

Kloeck-Jenson

(1999) Betwixt and between: “Traditional authority” and democratic decentralization in post-war Mozambique. African Affairs 98(393): 455–484.

102.

Zeeuw

J de

Kumar

(2006) Promoting Democracy in Post Conflict Societies. Boulder, CO: Lynne Rienner.

103.

Zürcher

Manning

Evenson

Hayman

Riese

Roehner

(2013) Costly Democracy: Peacebuilding and Democratization After War. Stanford, CA: Stanford University Press.