111 Seiten, Note: 1,0
List of Figures and Tables
2 Literature Review
3 Theory and Hypotheses
3.1 The Democratic Peace Thesis as Basis for Post-Conflict Democratization
3.2 Challenging the Democratic Peace Thesis: ”How Democratization Can Turn Violent”
3.3 The “Institutionalization before Liberalization” Approach as a Solution
3.3.1 The Role of The Rule of Law
3.3.2 The Role of Bureaucratic Quality
3.4 Intervening Factors
4 Research Design
5 Quantitative Analysis
5.1 Statistical Model and Proceeding
5.2 Data and Case Selection
5.2.1 The Dependent Variables
5.2.2 The Independent Variables
5.2.3 Control Variables
5.3.1 Descriptive Statistics
5.3.2 Main Results
5.3.3 Regression Diagnostics
6 Qualitative Analysis
6.1 Case Selection
6.2 The Case of South Africa
6.2.1 Course of Events
6.2.2 Analyzing the Events
6.3 The Case of Namibia
6.3.1 Course of Events
6.3.2 Analyzing the Events
6.4 The Cases of Liberia
6.4.1 Course of Events
6.4.2 Analyzing the Events
Appendix A Source of Variables
Appendix B Intercorrelation Table
Appendix C Regression Diagnostics
Diese Diplomarbeit entwickelt und prüft Hypothesen über den Einfluss von Rechtstaatlichkeit und einer funktionierenden Bürokratie auf den Erfolg von demokratischen Wahlen nach zivilen Konflikten.
Das praktische Interesse an solch einer Untersuchung ergibt sich aus der dramatischen Zunahme an Bürgerkriegen nach Ende des Kalten Krieges. Vermehrt werden Wiederaufbaumaßnahmen in ehemaligen Konfliktstaaten, insbesondere durch die Vereinten Nationen (VN), mit dem Einführen von Demokratie und Marktwirtschaft verknüpft. Wahlen werden bei diesem Vorgehen, welches auf der Idee des demokratischen Friedens basiert, als Endpunkt des internationalen Engagements gesehen und deshalb oft möglichst schnell durchgeführt. Das Scheitern einer Reihe von VN Missionen in den 90er Jahren (z.B. Angola, Liberia), die auf eine schnelle Demokratisierung fokussiert waren, hat zu einer vermehrten Kritik an diesem Vorgehen geführt. Es besteht die Gefahr, dass Wahlen aufgrund ihres kompetitiven Charakters zu einem Wiederaufflammen von Konflikten führen oder undemokratische Akteure an die Macht bringen.
Paris (2004) schlägt als Alternative eine Strategie der Institutionalisierung vor der politischen und wirtschaftlichen Liberalisierung eines ehemaligen Bürgerkriegsstaates vor. Institutionen - wie Rechtstaatlichkeit, Bürokratie und eine Zivilgesellschaft - sollen nach dieser These dazu führen die beschriebenen, negativen Effekte einer Demokratisierung zu kontrollieren.
In dieser Arbeit wird der weite Begriff Institutionen durch die zwei Konzepte Rechtstaatlichkeit und Bürokratie spezifiziert: Für diese beiden werden jeweils Hypothesen bezüglich ihres Einflusses auf den Erfolg von Wahlen nach internen Konflikten herausgearbeitet. Erfolg wird hierbei als das Verhindern eines Wiederauftretens von Konflikten definiert, sowie auch als das Beitragen zu einer positiven demokratischen Entwicklung.
Aufbauend auf den Empfehlungen von Lieberman (2005) werden quantitative und qualitative Methoden zum Testen der Hypothesen kombiniert.
Zuerst erfolgt eine statistische Überprüfung an Hand einer Gruppe von 35 Ländern, die in der Periode von 1982-2003 demokratische Wahlen nach zivilen Konflikten durchgeführt haben. Aufgrund der kleinen Fallzahl wird die Methode der exakten logistischen Regression angewandt, um zuverlässige Ergebnisse zu erhalten. Die beiden erklärenden Faktoren werden zunächst alleine und dann mit weiteren möglichen beeinflussenden Variablen auf ihren Effekt hin getestet.
In einem zweiten Schritt werden drei Fallstudien – Südafrika, Namibia und Liberia – im Detail in Hinblick auf die beiden Hypothesen untersucht. Die Fallauswahl erfolgt aufgrund der quantitativen Ergebnisse und berücksichtigt sowohl Staaten in denen eine Entwicklung entsprechend der Hypothesen sehr wahrscheinlich ist, als auch einen Fall, der die vorausgesagten Mechanismen in Frage stellt.
Die Resultate sind für die beiden Hypothesen leicht unterschiedlich. Für Rechtsstaatlichkeit kann in beiden Analysen klar der erwartete positive Effekt nachgewiesen werden. Für eine funktionierende Bürokratie sind die quantitativen Ergebnisse nicht ganz so robust und auch die Fallstudien lassen keine vollständige Bestätigung des beschriebenen Kausalmechanismus zu. Trotzdem lässt die ganze Arbeit insgesamt den Aufbau von Institutionen vor dem Durchführen von Wahlen in ehemaligen Bürgerkriegsstaaten als sinnvoll erscheinen.
Figure 1: Number of New Started UN Peace Missions per Decade
Figure 2: The Impact of The Rule of Law on Post-Conflict Election Success
Figure 3 The Impact of Bureaucratic Quality on Post-Conflict Election Success
Figure 4: Statistical Model
Figure 5: Output for Case Selection
Table 1: List of Cases
Table 2: Definition of the Dependent Variables
Table 3: Descriptive Statistics
Table 4: Univariate Regression of all Variables
Table 5: Regression of ROL with all Significant Variables
Table 6: Regression of BRQ with all Significant Variables
Table 7: Regression of ROL with all Significant Variables Excluding Outliers
Table 8: Regression of BRQ with all Significant Variables Excluding Outliers
„There is nothing harder than the apprenticeship of freedom“ Toqueville, Democracy in America (1835 : 229)
Since the end of the Cold War the number of civil wars with following peace- and state-building attempts has significantly increased. In line with the third wave of democratization theory (Huntington 1991) more and more states try to establish a market democracy after the end of domestic conflict.
Due to this, the United Nations (UN) has dramatically raised its peace building respectively peacekeeping activities (see Figure 1). During the Cold War the UN main security activity was peacekeeping, which typically involved the deployment of a lightly armed force to monitor a cease-fire or patrol a buffer zone
Figure 1: Number of New Started UN Peace Missions per Decade
illustration not visible in this excerpt
Source: United Nations
between former combatants (see Banerjee 2005: 19). The first mission employed followed the invasion of Egypt by France, Great Britain and Israel in 1956. For several reasons there was no ambition by the UN to further intervene in domestic issues of war-shattered states. Beside the formal prohibition of intervention within the domestic jurisdiction of any state, the two superpowers, the Soviet Union and the United States (US) had no interest on any intervention within the domestic issues of their allies and client states (see Paris 2004: 15).
After the end of the Cold War this changed drastically. With the western liberal democracy as the superior ideology and form of government, the UN started to account for that in its security missions. Recent peace building missions frequently are going far beyond solely monitoring of a cease-fire or peace agreement. Rather is its aim, as the UN Secretary-General Kofi Annan has put it, to “consolidate peace and prevent a recurrence of armed confrontation by various concurrent and integrated actions undertaken at the end of a conflict”(1997). Often the military aspects are connected with an attempt to set up a system of market democracy.
Throughout the 1990’s nearly all UN peace building missions focused on a fast democratization as the one best strategy for conflict-shattered states (see Paris 2004: 151). The UN officials believed that by promoting democracy, sustainable peace would be achieved. Both recent General Secretaries Boutros Gali and Kofi Annan, have argued in this direction. Annan, for example, has pronounced that within many good reasons for promoting democracy, “not the least – in the eyes of the United Nations – is that when sustained over time, it is highly effective in preventing conflict, both within and between states” (2000). Their belief is based on the democratic peace thesis. Dating back to Immanuel Kant, according to this theory, stable market democracies rarely go to war against each other. A general consensus has emerged around this finding. Several analyzes have similarly concluded that market democracies are less prone to intrastate conflicts (see Paris 2004: 42). By applying the assumptions of the democratic peace thesis, tolerably free and fair elections are usually taken as a sign of success by the international community involved in the efforts of reconstruction in former civil war states. Though quite often a successful organized ballot is regarded as endpoint of a peace building attempt (see Lyons 2002: 215; Caplan 2005: 120). Up to this argumentation holding free and fair elections is equated with the final establishment of a democratic environment, which should enable the war-torn state to move towards peace and stability.
But the question is, if a rapid transition towards democracy with immediate election is always the one best way to end domestic violence and enhance long lasting peace? Recent examples in Afghanistan and Iraq show that elections are by no means an endpoint for domestic conflict within war-torn societies. Democracy is obviously a guarantee for prosperity and peace in western countries, which have stable and mature democratic systems. “The western concept of democracy is based on the idea that the loser of an election has the possibility next time round of being the winner. But in the case of an ethnically or religiously divided country, in which minorities do not live peacefully together, this necessary balance can't be properly guaranteed” (Spiegel 2005: 110) as former US Secretary of State Henry Kissinger has phrased it. Thus for emerging democracies in post-conflict states the development might be different.
Several authors (see Hegre et al. 2001; Snyder 2000) have argued that semi-democratic states, meaning states in the transition from autocracy to democracy, are more likely to experience civil conflicts, than both autocracies and democracies. Post-conflict states usually can be counted in this class of semi-democracies. Nominally they behave like democracies, with organizing free multiparty elections, however a real democratic behavior is not visible in many parts of society. In particular, they have not established a system of peaceful conflict solving and bargaining typical of stable democracies. Quite the contrary, such societies often have a long tradition of solving occurring conflicts violently. That is why first post-conflict elections can be a starting point for new violence, fanned by the competitive elements of democracy, or a swing back to an autocratic system rather than a starting point for sustainable peace. Within contemporary history several cases with destructive effects of immediately held post-conflict elections, can be found. For example in Angola, the post-conflict elections in 1992, organized under supervision of the UN, immediately led to a restart of civil war. One of the former warlords, Jonas Savimbi, refused to accept his democratic defeat and returned to fighting. This brought the country another decade of civil war (see Macqueen 1998: 404).
Another problem of too early post-conflict elections is that it might enable largely undemocratic and extreme parties to come to power, and thus endanger a swing back to an autocratic regime. Richard Holbrooke, the special envoy of the Clinton administration to the Balkans, had clearly stressed this problem before the 1998 elections in Bosnia-Herzegovina: “Suppose the election was declared free and fair and those elected were racists, fascists, and separatists who are publicly opposed to peace and reintegration? That is the dilemma.” (quoted in the Economist 2003a: 28). Due to these past drawbacks, the international community is strongly criticized for its “vote and forget approach” (Plunkett 2005: 73) to solve domestic-conflict.
How can one avoid such a development? The right timing of the elections seems to be a crucial point. Robert Paris (2004) concludes that before starting the liberalization process with free elections and marketization, a certain level of institutionalization has to be reached in a civil war-torn state. According to this “Institutionalization before Liberalization (IBL) hypothesis” (see Paris 2004: 179) states with stable institutions should have a higher probability of successful peace building. By qualitatively studying several cases from the 1990’s, Paris (2004) gets to the opinion that institutions, like a civil society, a functioning bureaucracy or a rule of law carried out by an independent judiciary, are necessary to manage upcoming disputes caused by the democratization and liberalization process in post-conflict states. In September 2000 a report by an expert committee recommended a doctrinal shift for the UN peace building missions by demanding to regard post-conflict elections only as part of broader efforts to strengthen governmental institutions, including an effective civilian government and the basic acceptance of human rights and rule of law (see Brahimi 2000). Some recent missions by the UN and other international organizations for example in Kosovo, East-Timor and Bosnia-Herzegovina have partly adjusted their strategy to the IBL approach, by setting up an international interim administration for the transition period from civil war to democracy.
For future peace building efforts it is therefore of immense interest, whether to introduce primarily stable institutions before organizing democratic elections, is a promising strategy. If that theory proves true, conflict-torn countries with functioning institutions previous to the first post-conflict ballots should have a higher chance for stability and a positive democratic development afterwards. Hence this is the main question I will aim to answer by this thesis.
In this paper the IBL hypothesis is specified from the wide construct of institutions to two more detailed concepts. One is the rule of law and the other is bureaucratic quality. For both concepts, plausible hypotheses are developed regarding their causal influence on the successful settlement of post-conflict elections. Whereas Paris addresses with his IBL hypothesis economic as well as democratic liberalization, my examination is only focusing on the latter one.
While there has been some qualitative evidence for the IBL hypothesis a quantitative study is lacking so far. This paper tries to close this gap by conducting a quantitative analysis concerning the issue. The lack of a quantitative studies is not amazing due to problems connected with measuring the institutional quality of different countries. For my analysis I rely on data provided by the International Country Risk Group (ICRG) for the period 1982-2003. Out of this time period all countries that experienced a democratic election following a domestic armed conflict, classified as a civil war, will be considered. Computing a logistic regression model will test whether the success of these elections, meaning no recurrence of violence and a certain level of democratic quality three to five years afterwards, depends on the institutional quality. Because of the relative small case number (35) instead of an asymptotic logistic regression an exact logistic regression is conducted. The quantitative analysis also considers other possible explaining factors for successful democratization and peace building that can be found in the literature.
Regarding the current discussion about bridging the gap between quantitative and qualitative methods by combining both in one study (see for instance Collier et al. 2003; Lieberman 2005) and some methodical problems associated with the statistical analysis, I try to strengthen the results by additionally doing a qualitative analysis. For this purpose I will select three cases out of the quantitative results and research them in detail. This is done by looking closely at the events taken place after the election and trying to identify the possible role of the defined institutions.
The quantitative analysis shows strong evidence for the positive influence of the rule of law on the successful democratic settlement of post-conflict societies, even while controlling other explaining factors. Furthermore the effect remains stable when excluding outliers. For bureaucratic quality, the results are less robust. There is a significant influence solely for the lenient definition of success, which partly is not steady when simultaneously considering other factors and eliminating outliers.
The qualitative part finds further confirmation for the positive role of an established rule of law at the time of the first post-conflict elections. For bureaucratic quality there is no final judgment possible, even though the existence of a functioning executive instrument seems to favor the development of peace and stability in war-torn societies.
My paper is organized as follows: in the next chapter I will give a review on the relevant literature on civil war, peace building and democratization to better range my paper in the research context. Afterwards I will lay the theoretic basis and explain how the rule of law and a functioning bureaucracy can help to turn a post-conflict election into success. Out of this theory, my main hypotheses will be developed. Subsequently I will specify my research design. Section 5 is dedicated to the quantitative analysis. I will first present and discuss the statistical model and explain the sources of my data. Thereafter the logistic regression analysis will be worked out. The regression diagnostics and a discussion will conclude this section. Section 6 will deal with the qualitative analysis of the hypotheses. Guided by the “most likely - least likely approach” (Eckstein 1975: 118) three cases - South Africa, Namibia and Liberia - are selected and researched in detail. The last chapter gives the conclusion of my investigation, some outlook on future research, and a discussion about relevant policy recommendations.
The purpose of this section is to give a short overview on the literature about civil war and peace building to range my paper in the research context. I will start summing-up relevant literature on civil war for my research topic. Thereafter I will briefly review the essential literature on peace building and especially post-conflict democratization. Since section 3 will develop the theory and hypotheses out of prior literature findings equally, this section will be quite concise.
Concerning civil wars, a broad range of literature exists. Sambanis (see 2002: 218) classifies it along the three main phases of war: onset, duration and postwar transition.
For the first category, onset and causes of civil war, several mainly economic and quantitative studies subsist. For instance, as prominent examples Collier and Hoeffler (2004) as well as Fearon and Latin (2003) both use quite similar models to explain the start of a rebellion by rational decision-making of the actors. The demand and start of an insurgency is influenced by the expected utility of violence, including opportunity costs of a rebellion. Agreement has emerged around several variables, which decrease the utility of rebellion and therefore lessen conflict risk. These are educational level, per capita income and domestic growth rate. Discussion is ongoing on whether also grievance factors, like ethnic fractionalization, inequality or political rights have an influence on civil war occurrence, and people engage in civil war to seek justice (see Collier and Hoeffler 2004: 588). Particularly the point of ethnic diversity is a highly discussed one. Some authors (Reynal-Querol 2002; Ellingsen 2000) find a clear positive relationship between ethnic and religious cleavages and prevalence of civil war, whereas others (see Fearon and Latin 2003) do not find any statistically relevant association. Collier and Hoeffler (see 2004: 581) rather identify ethnic dominance than ethnic fractionalization as an incisive factor. Similarly no consensus exists about the role of identity wars, meaning conflicts of religious or ethnic origin, and other types (nonidentity) of war. There is a continuing discussion in the literature about the theoretical validity and empirical applicability of these classifications (see Sambanis 2002: 234). Recently two other factors are researched for their influence on civil war onset and duration: geography and natural resources. For the latter Ross (2004) offers a good summary of all principle quantitative studies. Generally there are only a few consistent effects observable – oil is linked to the onset of civil war and lootable goods are associated with war duration - and the causal mechanisms are quite vague. The underlying theory is that looting of natural resources is a way in which rebels can finance their insurgency. The research about geographic reasons is even more in its infancy. Buhaug and Gates (2002) have published a first systematic inquiry about geographical and other factors that influence the scope and location of civil-conflict. Fearon and Latin (see 2003: 85) assume that rough terrain favors the start of an insurgency, and find empirical evidence by analyzing mountainous terrain in their study. In turn Sambanis (see 2005: 311) mistrusts that argumentation.
Most interesting for my study are the papers that also account for the regime type as a reason for civil war outbreak. Agreement has emerged around the finding that stable democracies are considerably less likely to experience civil war than any other kind of regime type. For example, Gurr (2000) proves that the risk of violence because of grievance of ethno politic groups is less probable in well-established democracies, because minorities may use a form of protest instead of rebellion in such systems. Equally Ebaldwi and Sambanis (2002) find, with an econometric model a negative significant association between democracy and the outbreak of civil war. Rummel gets to a similar conclusion and assumes that in democracies, social conflicts that might become violent in other circumstances are resolved “through voting, negotiation, compromise, and mediation” (1995: 4). Thus, democracies are less likely to experience a wide range of domestic disturbance, such as revolutions, bloody coup d’états, guerrilla warfare and civil wars. Applying the democratic peace thesis to domestic conflict is in line with this argument, a point that I will discuss in detail in the theory section. Admittedly there is also some contradiction to the domestic democratic peace thesis. Collier and Hoeffler (2004) for instance find consistent with their theory, that grievance variables do not matter; no significant effect of democracy on civil war outbreak.
Other scholars, alongside solely looking at established democracies, also consider the process of becoming a democracy and its implication on domestic conflict. For example Hegre et al. (2001) find in their much noted paper firstly the same evidence as the preceding investigations: well settled democracies are considerably less likely than any other type of state to experience civil war. However, during the process of transformation in both directions, either from authoritarian rule to a democratic system or vice versa, states have the highest probability to experience civil-conflicts compared with both stable democracies and pure authorial regimes. Therefore their findings suggest that there is a u-shaped relationship between regime type and civil war, and so-called semi-democracies have the highest likeliness of conflict onset. This theory, even though much discussed, was recently confirmed by other studies (see for example Fearon and Latin 2003). Concerning international conflict, Mansfield and Snyder (2005) have quantitatively examined that states undergoing a transition from authoritarian rule to democracy have a high chance to get involved in international war. According to their results, war arises mainly in those transitional states that lack effective political institutions. If these institutions are absent, politicians will often resort to nationalistic ideology, to discredit their opponents as enemies of the nation in order to succeed in electoral competition. Therefore, states risk nationalist conflict if they try to turn into a democracy without institutions of public accountability (see Mansfield and Snyder 2005: 2). Snyder (2000) applies this theory to domestic conflicts and gets large confirmation by examining several case studies. For example, the bloody ethnic conflicts in Rwanda and Burundi in the mid-nineties were largely caused by too fast democratization attempts under pressure from international donors (see ibid.: 300). Similar conclusion is drawn for the case of Sri Lanka where elections held in an ethnic diverse society lead to repeated failure and violent conflict (see ibid: 280).
Referring to postwar transition significantly less literature exists compared to civil war onset and duration. The most comprehensive quantitative study is done by Doyle and Sambanis (2000). Their theoretical assumption is that the chance of peace building success is a function of local capacities, international support and the intensity of war related hostilities. The authors find that successful peace building is more likely to be successful after nonidentity wars, after long and not very costly wars, in countries with high development level, after a signed peace treaty at the end of the war, and if UN peacekeeping missions are conducted (see Doyle and Sambanis 2000: 795). The latter result is the most important of their research. UN missions, set in place after signing a peace treaty by both conflict sides foster the development to peace and stability. Particularly promising seem to be multidimensional peacekeeping operations, that is to say operations including civilian function, economic reconstruction, institutional reform and economic oversight, which are extremely significant and positively associated with peace building success, even defined as a development towards stable democracy alongside no recurrence of violence. However, UN missions classified as traditional peacekeeping are not at all significant (see Doyle and Sambanis 2000: 791). Fortna (2004) confirms the results of Doyle and Sambanis (2000) with a hazard rate analysis. Interestingly her results do not differ over the various types of UN missions. She discovers significant positive effects for all three types: multidimensional peacekeeping, traditional peacekeeping, and observer missions. Conversely to Doyle and Sambanis (2000) there are no significant effects for identity wars and a signed peace treaty identified. Additionally Fortna (2004) discovers significant influences of economic development, cost of war, and partly for wars that ended with a decisive victory by one side. This latter result is in accord with Licklider (1995) who explored a higher chance of renewed violence, if there was a negotiated settlement instead of a clear victory by one side for wars over identity issues. His explanation is that after a military victory, there is no major opposition group left, who could behave as a veto player in the transformation process and restart violence (see Licklider 1995: 685). Walter (2002) focuses her research on the conditions necessary for successful settlement of peace treaties after civil wars. According to her credible commitment theory negotiated peace treaties will only turn into success, if the warring parties receive a security guarantee from an outside country or an international organization that ensures their safety during the transition period, in which their demobilization usually takes place. By studying seventy-two civil wars between 1940 and 1992 she gets strong statistical evidence for this hypothesis (see Walter 2002: 90). Hartzle and Hoodie (2003) likewise research how peace-settlements influence the durability of peace. Their largely confirmed theory is that power-sharing settlements increase the likelihood of peace in civil war-shattered states.
Besides these quantitative studies, there is a sizable qualitative literature about post-conflict peace building. For example Dobbins et al. (2005) examine eight peace building missions by the UN – Congo, Namibia, El Salvador, Cambodia, Mozambique, Eastern-Slavonia, Sierra Leone, and East Timor – out of which they define seven as successes. That led them to the conclusion that nation building by the UN is effective in means of termination of conflict and promoting democracy. For missions led by the United States (US) Dobbins et al. (2003) finds less positive results. That can be partly caused by the more difficult missions of the US which required a higher personnel and financial commitment. Factors that play a major role for success are economic development, ethnic homogeneity, prior democratic experience and the effort made by the external actors. In general, according to Dobbins et al. there is no fast way of nation building, “five years seem to be the minimum amount of time to enforce an enduring transformation to democracy” (2005: 166). Two recent publications by Chesterman (2004) and Caplan (2005) examine the cases where the international community set up an interim administration after civil wars. Caplan (2005) researches the mission in Kosovo, Eastern-Slavonia, East Timor, and Bosnia Herzegovina where the UN and other international actors did not just keep the peace, but embarked on acting as de facto government to rebuild political authority. Such missions contain the risk that the transitional administrations feel more accountable to their donors then to the people they should serve for (see Caplan 2005: 195-196). The author identifies several variables crucial for the success of these missions, but his key conclusion is that “success, where it occurred, owes as much if not more in some cases to contextual factors than to operational practices” (Caplan 2005: 13). Chesterman (2004) focuses on the similar issue, however going back with his research to past missions as far as the time of the League of Nations. Chesterman (2004) likewise discusses the question whether the UN and other international actors should undertake this sort of function at all. His conclusion is, especially for the period from 1991-2003, that most of the UN peace building attempts were inadequate, inconsistent and irrelevant. For the future, he advises to use the instrument of transitional administration more selectively and with more emphasize to such basic concerns as economic stability and the rule of law (see Chesterman 2004: 256). The aforementioned publication by Paris (2004) draws similar conclusions as Chesterman (2004). By a qualitative analysis he assesses most post cold war peace building attempts by the UN as inadequate and proposes the IBL approach as a solution for the future. His argument will be thoroughly explained as base for my theoretical approach in the next section. Covey et al. (2005) argue analogically, by doing a detailed analysis of the Kosovo conflict that viable peace is closely related to sufficient domestic institutions, which permit a peaceful conflict-solving.
The role of institutions for democratization processes in transitional states is a topic already discussed in the wide literature of democratization. In general, the main conclusion of classical authors (see for example Huntington 1976; Linz and Stepan 1996; Diamond et al. 1990) is that institution play a central role for the successful implementation of democracy in transformational societies. I will discuss that literature in detail regarding my research question in section 3.
This short literature review highlights that the topic of civil war is mainly studied with regard to the causes for onset and duration, largely in a quantitative manner. In contrast, for post conflict reconstruction, quantitative studies are rare and have elided the issue of post-conflict democratization so far. However, there are several qualitative studies presuming that democratization could be the cause of renewed conflict. Institution building is assumed to be a solution of this problem. My thesis will therefore try to enlarge the research area by conducting a quantitative analysis on the issue of post-conflict elections and the role of institutions followed by case studies selected out of the quantitative results.
This section will start with a short introduction to the democratic peace thesis that is often used as the main argument for immediate democratization strategies in post-conflict states. Afterwards the doubts about this approach to post-conflict reconstruction will be presented, out of which the IBL theory will be developed. In the following paragraph, I will specify the IBL theory as applied to the concepts of the rule of law and bureaucratic quality. The section concludes by introducing other explaining factors for successful post-conflict democratization, which should be used as control variables in the quantitative analysis.
As pointed out in the introduction for the international community, post-conflict elections often serve as an endpoint for their peace building involvement. Keeping that in mind, it is not surprising that in many former civil war states, a strategy of fast democratization and liberalization is carried out. Less attention is often paid to set up stable institutions. An astonishingly high number of commentators, not only in the west, seem to support this principle (see Miall et al. 1999: 204). The theory behind this fast democratization strategy is, as mentioned above, the democratic peace thesis.
This thesis was first formulated by Immanuel Kant (1795 ), who believed that most of the people would never vote to go to war, unless in self-defense. Therefore, if all nations were republican democracies, it would end war, because there would be no aggressors. US president Woodrow Wilson has first brought the theory in political practice. At the end of World War I he proclaimed that by spreading democracy throughout the globe would promote peace in both domestic and international affairs. For him, democracy was “unquestionable the most wholesome and livable form of government the world has yet tried” (quoted in Notter 1965: 109). Only by promoting democracy throughout the world, peace and stability can be achieved. Although his proposal mainly focused on interstate conflict, Wilson also believed that these principles are essential to domestic conflict as well (see Paris 2004: 40-41). Dean Babst (1972) was the first to research the theory on a scientific basis. Since the mid-eighties, the research on the connection between liberal forms of government and the impact on both domestic and international conflict is one of the most studied topics in political science. Starting with Michael Doyle’s path-breaking analysis in 1983, there have been various studies proving that interstate war between consolidated democracies is improbably compared with the great number of conflicts between other types of regimes. While a general consensus has emerged around these findings, when it comes to explaining there is considerable disagreement (see Farnham 2003: 397). There are several underlying explanations on why democracies are reluctant to fight war against each other. One hypothesis focuses on political norms as the cause for democratic peace. According to this explanation, due to the democratic culture, political leaders are more used to negotiation and compromises and usually settle mutual conflicts peacefully (see Weart 1994: 311). Another idea is that the political structures hinder democracies to start fighting each other. Russet (see 1993: 30) argues that democracies give voting power to those most probably to be injured and killed in wars. Thus there is a low probability that the population’s majority in democracies decides to go to war. While the democratic peace theory derives from the idealistic school of international relations, there is also a realistic approach challenging to see democracy as the sole explanation for the observed phenomenon. Spiro (see 1994: 79-80) for instance suggests that peace between democracies can be rather explained by the intervening variable of close alliances, and the democratic peace is furthermore a peace between allied democratic states.
So far, much of the research effort to the democratic peace thesis has been put on the correlation between democracies and interstate conflict. But several analyzes on civil violence have similarly concluded that democracies are less probable to experience intrastate conflicts as well (see Paris 2004: 42). Stable democracies normally neither participate in domestic violence, like mass murder and genocides, nor in civil wars (see Rummel 1995: 24). Hegre et al. (2001) find in their much noted study that stable democratic systems are in comparison, unlikely to experience any civil conflict, and thus find support for a democratic civil peace theory. Such settled democracies should have the ability to manage all upcoming disputes peacefully. The institutions and processes of democracy should defuse revolutionary violence by diverting popular discontent in electoral competition and nonviolent protest (see Mason and Quinn 2006: 25).
Considering these findings it is not amazing that the UN has identified the spread of democracy as its favored strategy to end international as well as domestic conflicts. Besides his agenda for peace (1992) the former UN general secretary Boutros-Ghali has also developed an agenda for democratization (1996). Therein he defines the UN’s support of democratization as the base for a cooperative and peaceful international system. For Boutros-Ghali “peace development and democracy are inextricably linked” (1996: 118).
But as mentioned in the introduction, serious doubts exist on whether a rapid democratization is the one best way to end domestic conflicts in every case. These doubts and where they derive from, I will discuss in the next paragraph.
All the arguments mentioned above seem reasonable to foster a further spread of democracy, especially in post-conflict situations. But all arguments on the democratic peace thesis have one important error in reasoning: although well established market democracies may be more peaceful in their domestic and international acting, the policy of promoting democracy usually involves transforming a country in a market democracy (see Paris 2004: 44). No state becomes a consolidated democracy overnight. On the contrary, all states have to go through an often rocky phase of transition. Creating a stable market democracy is usually a tumultuous, conflict-ridden and lengthy process, especially under the fragile conditions of war-shattered states (see Paris 1997: 57). During this phase the chances to experience new conflict is high, not only compared with stable democracies, but also to autocracies (see Hegre et al. 2001: 34). Huntington (see 1991: 192) also finds evidence that political violence is frequently connected with democratization. Ohlson and Söderberg (see 2002: 35) likewise describe the process of democratization in post-conflict states as often revolutionary and conflict-causing. Democratization involves dramatic shifts for the volatile society in civil war-torn states. Indeed one can discover a reasonable number of post-conflict elections which turned into new domestic violence. Examples are widespread throughout the globe and range from Sri Lanka, Rwanda to the already described case of Angola. In Burundi the 1993 elections – although assessed as democratic and fair by international observers – intensified ethnic polarization between Hutu and Tutsi ethnic groups leading to more than 20.000 deaths (see Mansfield and Snyder 2005: 200).
Arguments for possible damaging results of elections can already be linked to one classical thinker of political theory: Thomas Hobbes (1682 ) describes democracy as one reason for the civil war in England between 1640 and 1648. Democracy encourages battle of opinions and thus can lead to a destroying war between different doctrines. His conclusion is the necessity of an unlimited ruling Leviathan. Snyder (2000) finds qualitative empirical evidence by showing that there is a strong relation between the process of democratization and the start of nationalistic and ethnic conflicts. Nearly all states undergoing bloody ethnic conflicts during the 1990’s were taking initial steps toward democratic transition, like holding free elections in advance (see Snyder 2000: 28). His explanation is that in early phases of democratization, the elites often appeal for nationalistic issues to compete for popular votes. If a state has already experienced a domestic armed conflict, the risk of repeated violence due to ethnic and national resentments caused by public elections might be even larger. The general problem of multiparty elections is that they are based on competition. By gaining political support, different political parties stress their different approaches, which can lead to polarization and conflicts. Therefore elections can serve as a focal point for destructive and harmful competition in countries going through a phase of transition (see Paris 2004: 163). Especially in countries where potential conflicts were delivered by means of weapons in the past, elections may have often a quite destabilizing effect. If ballots are held too early “politics becomes the continuation of war by other means” (Chesterman 2004: 207). If one former conflict party does not like to give political power to one of its former opponents, who won popular elections, it might return to violent actions. In contrary to mature western democracies, the losing parties in post-conflict democracies have no guarantee that there will be another election offering them a new chance to compete for office. That should lower the costs of ballot losers to start new rebellions (see Strand 2005: 8).
The previous point is closely related to the second possible threat that can emerge from post-conflict elections besides new violence: the danger to swing back to an autocratic status. If elections are held too early in civil war-torn states, democratic legitimacy may be transferred to non-democratic actors (see Chesterman 2004: 234). Often, one can note that freely elected leaders do manipulate their countries transition to democracy to avoid facing a democratic challenge again. In recent history we can observe numerous such examples, in which the outcome of elections was what Diamond calls a “pseudo-democracy” (1996: 25). Hence as Walzer puts it “the real quest for a democracy is not the first election, but the second and the third” (see 1996: 1): There is no guarantee that the winners of democratic elections are convinced democrats. Ottaway (see 2003: 193) describes the rise of what she names as semi-authoritarian regimes, as the result of the coercion by the international community to organize elections in states that are not domestically prepared for democracy in the post Cold-War period. Particularly in post-conflict states we have an increased chance that anti-democratic parties take part in the first election. Former warlords, are at high risk to turn to autocratic leaders if they come to power, even though freely elected.
As a result of this paragraph, one can conclude that elections, although conducted in a free and fair manner, do have the potential to cause violence and to impede democratization. There is no guarantee that formally free elections lead to stable and peaceful consolidation of a conflict-shattered state. Exceedingly in the context of a previous armed conflict within a state, there is a high risk that upcoming elections either foster a recurrence of violence because of the competitive character of democracy, or undemocratic parties, working against the establishment of a stable democratic system, come to power. And if the worse comes to worst, both scenarios could take place consecutively. That is therefore the basic assumption for the theory and the following hypotheses that will be developed in the next paragraphs.
The question now is, how to avoid such developments? Is there a strategy through which the destabilizing side effects of post-conflict elections might be weakened or in the best case, totally prevented? The next section will deal with a possible solution.
There are several reasons why usually no conflicts evolve in stable democracies, although its democratic system has the same competitive elements that I identified as a threat for peace and stability in post-conflict states. The main cause is that competition in these mature democracies takes place in a “framework of institutions” (Paris 2004: 159). These institutions are capable in managing disputes and to translate public debate into government policy. The democratic peace literature tends to take the existence of a functioning state with stable institutions as given. But in many conflict-shattered states we find rather different conditions: the absence of even the most rudimentary governmental institutions (see Paris 2004: 46). That is why Paris (2004) is arguing, as noted above for an Institutionalization before Liberalization (IBL) strategy, as a possible solution to stabilize, and in the end successfully democratize post-conflict countries. In war-torn states, there are three common problems according to Paris (see 2004: 163-173): intense social conflict, weak conflict dampeners – meaning no culture or tradition of peaceful conflict solution – and ineffective institutions, meaning constitutions, legislatures, bureaucracies, courts and the like. Therefore peace building attempts should focus on building a framework of stable political institutions prior to encouraging political and economic competition. Paris (see 2004: 187-188) designs a comprehensive strategy for post-conflict peace building: firstly, to wait until the right conditions for elections; secondly, to design a suitable electoral system that rewards moderation; thirdly, to support a good civil society; fourthly, to prevent hate speech; fifthly, to adopt conflict-reducing economic policies and sixthly, as a common denominator to rebuild effective state institutions. Thus the hypothesis of Paris is that the institutions are the key factor for success of post-conflict reconstruction. With stable institutions at the start of a democratization process, a conflict-shattered state should be able to deal with threats to a sustainable democratic development. Whether elections further the development of a stable democracy and diminish the risk of renewed violence, does not solely rely on a ballot conducted in a free fair manner, but rather on the institutional setting within the state.
Arguments and theoretic outlines for an IBL approach can be equally found in several other sources. Huntington (see 1976: 4) for example assumes that it is often a strategic mistake of institutionally weak states to handover power to the mass public, because such states face a gap between broad participation in politics and lacking institutions to manage increasing demands. Cowen and Coyne (see 2005: 40), who construct a game theoretic model for post-conflict reconstruction also think that to set up stable institutions is a key element for success. This opinion is shared by Doyle and Sambanis (see 2000: 780), too, who argue that sustainable peace needs state authority as a starting point to overcome security concerns. Out of his research results, Snyder (see 2000: 41) advises leaders to implement a strategy of civic institution building before starting a democratization process. At least a gradual development of the rule of law, a working bureaucracy, civil rights and an independent media should be achieved before holding free elections (see ibid.: 41). Prezeworski (see 1988: 64 et sqq.) develops the theory that democracy can only be established if there are institutions that guarantee for every potential party that their interests are not affected in a highly adverse manner in the course of democratic competition. Democracy will only be sustainable if institutions are capable in reducing some of the uncertainties connected with democratic elections. Since no one can be sure that their interests will ultimately triumph, one needs at least some certainty that he is not totally overthrown by a winning majority. Walter (see 1997: 140) makes a similar point while she finds the weak institutions in many civil war states as one reason for the failure of many democratization attempts. If there are no basic democratic institutions set in place, it is presumable that the loser of the first election will be permanently excluded from power. Chesterman et al. (see 2005: 382) conclude that elections should be regarded as just one small part of a transformation to democracy besides setting up respect for the rule of law, establishing a market economy and a civil society. Finally Coyne (see 2005: 325) argues that a reconstruction success, meaning a self-sustaining liberal political, economic and social order that does not rely on external support, is constrained by institutional prerequisites.
Given these literature findings, I assume that in post-conflict states there should be a higher probability of democratization success depending on the institutional quality at the start of the process. If the IBL theory proves true, post-conflict countries with better functioning institutions at the first post-conflict elections should have a higher chance to make these elections a success, meaning no recurrence of violence and a sustainable democratic development. That is therefore the general hypothesis for my analysis.
Since the term institution is quite imprecise and encompasses a wide range of different ideas which makes an exact examination complicated, I will focus my analysis on two sub-concepts: the rule of law and bureaucratic quality. A detailed theory about their influence on the success of post-conflict elections will be developed and presented in the next two sections.
In the following section I will show how a positive influence of an established rule of law on the success of post-conflict elections can be explained.
Since the literature shows ambiguity about the exact definition of the rule of law (see Kleinfeld Belton 2005: 3), I will first clarify what I understand by the latter concept. For my paper I define the rule of law as the supremacy of law for all parts of a society. Every person, and especially the political leaders, should be bounded by common statutes. The laws, with a constitution as the central element, are supposed to be applied by an independent judiciary. Such an independent judiciary is able to limit government arbitrariness and power abuse.
In the past, the UN and other international actors have falsely assumed that the rule of law will automatically follow succeeded democratic elections. But without first recreating the rule of law, a war-shattered state faces great risk to slipping back to a collapsed status, which will prolong conflict and fail to achieve the aim of sustainable peace and democracy (see Plunkett 2005: 77).
A settled rule of law executed by an independent judiciary can be an important factor to manage new disputes in succession of post-conflict elections. Elections are often connected with disputes about the official results afterwards. Especially if a new electoral system is set in place for the first time, discussion about its right execution are likely to occur. Also, fraud attempts and irregularities are probable to arise if a post-conflict election takes place. A judiciary, accepted as independent by all parties, can play the role of a mediator in case of such disputes. Paris (see 2004: 205) describes a constitutional court to resolve disputes surrounding elections as a central element for success of post-conflict elections. Linz and Stepan (see 1996: 10) define the rule of law embodied in a spirit of constitutionalism as well as essential precondition for a successful democratization. This spirit of constitutionalism must be based on a hierarchical law system interpreted by an independent judiciary (see ibid.). If a constitution and law system based on it is widely accepted by all electoral participants and rooted within most parts of society, a return to violent as a means of conflict-solving should be less probable, since all parties can be sure that the democratic rules of the game will be accepted by their opponents. Such an embedded constitution could be reached best if it is not exclusively imposed by external actors, but rather derives out of negotiations of the relevant social groups of a society.
 Article 2 (7) of the UN Charter
 For example in Kosovo, besides the UN, the Organization for Security and Co-operation (OSCE) the European Union (EU) and the North Atlantic Treaty Organization (NATO) are taking part in the international reconstruction efforts.
 For an short overview see Farnham (2003: 397-398).
 Similar arguments can be found at Gagnon (1994) who highlights the case of Serbia.
 For example: Cameroon, Burkina Faso, Equatorial-Guinea, Tanzania, Gabon, Kenya, Armenia, Azerbaijan, Georgia, Kirgiztan, Kazakhstan, Belarus and Mauritania are among those states (see Carothers 2002: 13).
 The misuse of power by an elected leader is often one of the reason for his opponents to return to the weapons as described above.
 In general, definitions can be divided in two categories: One focusing on the ends of the rule of law, meaning for instance to uphold law and order and to make fair and predictable judgments, while the other puts more emphasis on the institutions necessary to actuate the rule of law (e.g. comprehensive laws, well functioned courts) (see Kleinfeld Belton 2005: 3).
 Sawyer (see 2005: 133-135) even goes to such lengths describing those constitutional rules and arrangements in emerging democracies should be implemented upon grassroots discussions. A process, that might be hard to work out in many post-conflict states with very low educational level.
Wissenschaftlicher Aufsatz, 21 Seiten
Examensarbeit, 38 Seiten
Seminararbeit, 26 Seiten
Essay, 10 Seiten
Studienarbeit, 32 Seiten
Hausarbeit (Hauptseminar), 21 Seiten
Wissenschaftlicher Aufsatz, 21 Seiten
Examensarbeit, 38 Seiten
Seminararbeit, 26 Seiten
Essay, 10 Seiten
Studienarbeit, 32 Seiten
Hausarbeit (Hauptseminar), 21 Seiten
Der GRIN Verlag hat sich seit 1998 auf die Veröffentlichung akademischer eBooks und Bücher spezialisiert. Der GRIN Verlag steht damit als erstes Unternehmen für User Generated Quality Content. Die Verlagsseiten GRIN.com, Hausarbeiten.de und Diplomarbeiten24 bieten für Hochschullehrer, Absolventen und Studenten die ideale Plattform, wissenschaftliche Texte wie Hausarbeiten, Referate, Bachelorarbeiten, Masterarbeiten, Diplomarbeiten, Dissertationen und wissenschaftliche Aufsätze einem breiten Publikum zu präsentieren.
Kostenfreie Veröffentlichung: Hausarbeit, Bachelorarbeit, Diplomarbeit, Dissertation, Masterarbeit, Interpretation oder Referat jetzt veröffentlichen!