Order and Justice in the Dayton Agreement : An English School Analysis of the General Framework Agreement for Peace in Bosnia and Herzegovina

(1)

Linköping University | Department of Management and Engineering Master’s thesis, 30 credits| Master’s in International and European Relations Spring 2020| LIU-IEI-FIL-A--20/03498--SE

Order and Justice in the

Dayton Agreement

An English School Analysis of the General Framework

Agreement for Peace in Bosnia and Herzegovina

Rickard Montgomery-Bjurhult Supervisor: Mikael Blomdahl Examiner: Mikael Rundqvist

Linköping University SE-581 83 Linköping, Sweden +46 013 28 10 00, www.liu.se

(2)

1

Abstract

The war in Bosnia and Herzegovina was ended in 1995 with the signing of the Dayton Agreement. However, despite almost 25 years passing since its implementation Bosnia and Herzegovina is still a troubled state with deep internal divisions. This study seeks to analyze the Dayton Agreement using an English School framework together with an examination of the most prevalent approach to peace within it. In Particular, it uses recent English School theoretical developments that place the concepts of order and justice on a spectrum where an ideal mix is sought. The focus is on discovering whether there is a focus on order to the detriment of justice.

To accomplish this the thesis uses a combination of qualitative text analysis and the ADICO grammatical syntax which seeks to analyze statements by breaking them down into their constituent parts. The results of the analysis showed that the initial hypothesis was faulty, and order was in fact the least coded part of the Dayton Agreement. However, it also uncovered a number of problems, and contradictions within the Agreement, most of whom were to the detriment of the parts dedicated to justice and peacebuilding.

Findings indicate that the parts of the Dayton Agreement focusing on justice and long-term peacebuilding have been negatively affected by inconsistencies and problems within it as well as the sheer variety of goals that the Agreement seeks to fulfill. This in turn has had implications for the long-term success and stability of Bosnia and Herzegovina.

Keywords: Dayton Agreement, Dayton, English School, Order, Justice, Peace, Bosnia and Herzegovina, Bosnian War, Balkan history, ADICO, Qualitative text analysis.

(3)

2

List of Abbreviations

ADICO – The Grammatical Syntax by Crawford and Ostrom (1995). It stands for Attribute, Deontic, aIm, Condition, Or else

BiH – Bosnia and Herzegovina

Dayton Agreement – The General Framework Agreement for Peace in Bosnia and Herzegovina

EU – European Union

GFA – The General Framework Agreement for Peace in Bosnia and Herzegovina ICRC – The International Committee of the Red Cross

IFOR – The international implementation force in Bosnia and Herzegovina IPTF – International Police Task Force

IR – International Relations

NATO – North Atlantic Treaty Organization NGO – Non-governmental organization

OSCE – Organization for Security and Cooperation in Europe UDHR – Universal Declaration of Human Rights

UN – United Nations

UNHCR – United Nations High Commissioner for Refugees WWI – World War One

(4)

3

Introduction

Almost 25 years have passed since the General Framework Agreement for Peace in Bosnia and Herzegovina1_{(more commonly known as the Dayton Agreement) was} signed and while the agreement can be considered a success in so far as there has not been a return to violence within BiH, the creation of a stable democratic state has so far been a failure. Annex 4 was intended to function as a temporary constitution until such time as BiH politicians could agree on a new one, but this has yet to happen (Keil & Kudlenko 2015; Korkut & Mulalić). The overarching state created is weak, suffers from low legitimacy, frequent political deadlocks, rampant corruption, and limited democratic structures. Much of this has (justifiably or not) been blamed on the Dayton Agreement itself and it has received widespread criticism for being an obstacle rather than an aid for the further development of BiH into a stable democratic state (Hronesova, 2015; Keil & Kudlenko, 2015), criticism that is prevalent even outside of academic work. This can be seen in The Guardian which describes the Dayton Agreement as having “Frozen ethnic divisions in place” and “bequeathed an extremely complex system of government” upon the state which has “made governance extremely difficult” (Nardelli et al. Oct 8, 2014) while Al Jazeera describes BiH as a “European quasi-protectorate” and unsuccessful democratic project where foreign powers through the Dayton Agreement have tried (and failed) to implement democracy by force (Pasic, 29 April 2013). However, Bieber (2006) and Hronesova (2015) both argue that as much as the Dayton Agreement is flawed the blame lies as much with the actions (and inactions) of local elites and international powers as with the agreement itself (Bieber, 2006; Hronesova, 2015).

The question then becomes what went wrong with the recipe set out by the Dayton Agreement since it contains ingredients both for short-term peacekeeping in order to quickly stop the war, something that was clearly successful, but also for the long-term creation of a democratic state (GFA, 1995). Was the latter part doomed from the start because of flaws in the agreement itself, or was it because of failures in adherence and implementation of the agreement? A combination of the two? There is an abundance of literature on this subject but not much of it examines the agreement through a theoretical lens (Keil & Kudlenko, 2015). This thesis seeks to add to this literature by examining each annex of the Dayton Agreement

1_{The Agreement is available for free from the homepage of the OSCE.} https://www.osce.org/files/f/documents/e/0/126173.pdf

(7)

6

through the theoretical lens provided by the English School as well as by exploring the approach to peace that it takes.

The analysis will be performed from an English School perspective using the ideas championed by Buzan (2001) and Weinert (2011) where the English School concepts of solidarism and pluralism are not locked in a mutually exclusive zero-sum debate but as the extreme ends of a spectrum where movement can occur in either direction and as tools of analysis and assessment used to aid in an understanding of decisions (Buzan, 2001; Weinert, 2011). Galtung’s (1976) theory of the different approaches to peace and their long-term viability will also be used to examine the peace the Agreement sets out. In particular, the thesis will look at the concepts of order and justice, and how these are present within the Dayton Agreement. Order in the English School understanding of the term being related to pluralists and justice to solidarists (Buzan, 2014; Galtung, 1976). Continuing this train of thought the analysis can then focus on whether or not the Dayton Agreement, or its implementation, veers too far towards order (and pluralism) over justice (and solidarism). The assumption here being that one cannot have order without justice, or it is seen as illegitimate and unjust and will not function in the long term (Buzan, 2014). Should a focus on order over justice be evident, then that could aid in the understanding of why the Dayton Agreement and the state that was constructed using it as a blueprint has struggled long-term.

Problem Formulation

This thesis seeks to understand the dynamics between order and justice in the Dayton Agreement. The successful end of hostilities but apparent failure of long-term democratic state-building suggests that there is an emphasis on order at the detriment of justice within the Agreement which will be the driving assumption of the thesis. Long-term peacebuilding in post-conflict states is both an important and difficult problem in contemporary international relations, as recent problems have shown. To be able to improve future peace agreements it is necessary to understand how agreements such as the Dayton Agreement are constructed and how the combination of priorities within it affects its likelihood of success. This leads to the problem formulation used in this paper.

Has a focus on order over justice in the Dayton Agreement negatively affected its chances of success?

(8)

7

The emphasis here is on the internal logic within the Agreement itself as this has consequences for its implementation.

In order to accomplish this, an analysis of the Dayton Agreement and all of its annexes will be performed. This will be done using institutional and qualitative text analysis. The results will later be discussed within the realm of academic literature on the subject, different approaches to peace, and through the theoretical lens provided by the English School.

(9)

8

Theory

English School

The English School is the dominant non-American theory of International Relations (IR) and differs from American IR theories in several ways; it accounts for the influence of history, is methodologically pluralist and tries to create “a more holistic and, integrated approach to the study of international relations” (Buzan, 2001 p. 472). The primary focus of English School literature is on international society and much of English School theorizing has traditionally been locked into the pluralist versus solidarist debate about the nature of international and world society “or more specifically [how] interstate society relates, and should relate, to world society” I.e. “how states relate, and should relate, to people.” (Buzan, 2014, p. 83). The most important part of this is the conflict between the idea of universal human rights for all and that of a state’s right to sovereignty and non-intervention. Buzan (2014) explains that it be either seen as the dilemma between “the rights of states versus the rights of people” or “the nature and potentiality of interstate society, and particularly the actual and potential extent and type of shared norms, rules, practices and institutions within interstate societies” (Buzan, 2014 p. 83) The latter having a less conflicting view between state rights and people’s rights. Most important in this conflict is the possibility of humanitarian intervention. A strong belief in universal human rights also opens the door for humanitarian intervention into another state should it not be living up to its responsibility of protecting the human rights of its citizens. Conversely, a strong belief in state sovereignty can lead to seeing humanitarian intervention as a gross violation of a state’s rights (Buzan, 2001; Buzan, 2014). However, the differences between what is considered world versus international society is not always clear. Buzan (2001) tries to clarify the difference by pointing out that the two have a different ontological basis. According to him “world society starts from individuals and is in clear contrast to the state-based ontology of international society,” but he admits that this makes the position of organized international non-state entities, such as international corporations or non-governmental organizations, unclear (Buzan, 2001 p. 477).

Pluralism is similar to the more famous theory of Realism and focuses on state sovereignty, international anarchy, rationality, and distinctiveness. In a pluralist view, states can be compared to people, with different values, ideas, and preferences. This then leads to a minimalistic interpretation of international society that is focused mostly on shared interests such as agreements on sovereignty, non-intervention, and how to ensure order in the anarchy

(10)

9

of the international system (Buzan, 2001; Buzan, 2014; Weinert 2011). Violation of the non-intervention principles is generally only seen as acceptable if it is necessary for the survival of the intervening state or if it is done to maintain a careful balance of power (that is then also seen as necessary for the self-preservation of said state). This is very similar to realist thinking but pluralism does have some important differences. According to pluralists, states both can and do intentionally engage in peaceful and orderly relations as opposed to the realist vision of states engaged in a constant struggle for survival within the international anarchy. According to Buzan (2014), “the basic principle of realism is survival” while “that of pluralism is coexistence” (Buzan, 2014 p. 90). Pluralism also has room for ethical considerations, but its approach to ethics is very practical. There is no black and white description of what is good or bad but rather an understanding that within different communities there is a wide diversity of ideas about ethics and morality and that this is not only acceptable but also something that is worth defending. Thus, pluralist ideas of ethics and diversity lead many pluralists into opposing those ideologies that have a universalist basis and argue that certain ideas and rights are good for all humans, no matter their culture (Buzan, 2014, Weinert, 2011). This leads to an obvious potential for conflict between pluralists and those theories and/or ideologies that promote universal human rights and specifically with those that promote the idea of humanitarian intervention in defense of these rights.

As the necessity of order is an important point for pluralists that also naturally leads pluralists to see disorder as a direct threat to society. Disorder can come from many sources, but most obviously it is from conflicts both within and between states as well as through a lack of existing (or functional) states, i.e. anarchy (Buzan, 2014). The most important goal for international society is thus to provide an option where states can coexist, compete, and manage collective issues in an orderly fashion (Buzan, 2014). This focus on order and ethical pragmatism mean that pluralists by default usually support the existing system both within a state and internationally. Wight puts it thus: “[pluralism] makes a presumption in favor of existing international society” (Qtd. in Buzan, 2014 p. 91).

Solidarism on the other hand is more revolutionary and has a wider understanding of what international society can and should be. Here the state is not considered an actor in and of itself but as an agent that represents all of its citizens. This branch allows for an international society that is encompassed by a wider portfolio and it also opens up for the possibility of internationally shared norms and ideas of what is acceptable and not acceptable within society and the relations between states and their citizens (Buzan, 2001; Buzan 2014). Solidarism is

(11)

10

also universalist with a view that all of humanity is the same and enjoys the same rights but then also has the same responsibilities. Solidarists argue that a focus on the rights and well-being of all human well-beings is important to ensure an international society that is peaceful and stable in the long-term. Thus, solidarists focus especially on Human Rights as a shared moral understanding that has spread across international society. Solidarists are also open to the possibility of humanitarian intervention by third powers into conflicts elsewhere (Buzan, 2001; Buzan 2014; Weinert, 2011). This does not mean that solidarism does not recognize sovereignty as an important concept for states but rather that it allows for more flexibility and convergence of the concept (such as within the European Union) and that it is tied to a responsibility to protect as well as respect the rights of its citizens. A state that does not respect its citizens in this way is thus illegitimate and loses its right to sovereignty and non-intervention (Buzan, 2014).

The complexity of the solidarist position means that there exists a significant variation within solidarism itself. For example, since solidarists view certain values as universal, there is the question of how wide this group of universal values is and how coercive (or not) one should be when spreading these ‘universal’ beliefs to states that do not abide by them (Buzan, 2014). Buzan (2014) distinguishes two different clear strands within solidarism, ‘state-centric solidarism’ and ‘cosmopolitan solidarism’. As can be inferred from the name the former is more similar to pluralism as it has the same ontology as pluralism with states as the central unit of analysis (Buzan, 2014). However, aside from this change of focus, it is still otherwise closely related to the other branch of solidarism as states are assumed to willingly share norms and cooperate beyond the pluralist logic of simple coexistence. It can be seen as less revolutionary, but still firmly solidarist as states are seen as the ones who guarantee and promote universal values (Buzan, 2014). Despite the ontological difference, there is no inherent contradiction between the state-centric focus of this branch of solidarism and the promotion of human rights if one views it as human rights and state rights being on the same level, enjoyed by both and not one superior to the other (Buzan 2014).

Cosmopolitan solidarists on the other hand see all humans as part of a greater whole and that there are inalienable universal rights, moral considerations, and natural laws that are superior to laws negotiated and created by states. According to Buzan (2014), cosmopolitan solidarism originated in a “desire to establish foundations of a moral critique of the state and the logic of interstate society. This is accomplished by taking the community of humankind as a whole as a moral referent against which to judge the behavior and purpose of states” (Buzan,

(12)

11

2014 p. 118). Within cosmopolitan solidarism, there is a clear focus on human rights as universal and superior to state rights which leads the door open to humanitarian intervention against states that violate these rights. When pluralists argue against the danger to international order posed by solidarism it is in most cases cosmopolitan solidarism that is the source of their fears (although some do not know or recognize the difference). However, most of solidarisms moral and ethical considerations that take the English School as a whole further than the cold rationalism of the theory of realism have their root in cosmopolitan solidarism (Buzan, 2001; Buzan, 2014, Weinert, 2011). Interestingly, despite the focus on pluralists on arguing against this more extreme version of solidarism, most solidarist thinkers actually belong to the state-centric branch of solidarism and even within cosmopolitan solidarism, there is little interest in pushing for the abolishment of states or the state-centric international system (Buzan, 2014).

Another way of thinking about the pluralism versus solidarism debate is as the concepts of order and justice and it is in fact this dilemma from which the pluralist versus solidarist discussion originally emerged. Using this angle then the solidarist/pluralist debate is about how order and justice interact in today’s globalized international system and what combination of the two is ideal in any given situation. (Buzan, 2014). Pluralism represents order and pluralist writers argue that in order for a society to have justice it must first have order, or a just system is impossible, although they do recognize that a system with a complete lack of justice is unstable. Solidarism on the other hand then represents the side of justice in this debate and many theorists that follow this branch argue that the relationship is reversed and justice is a prerequisite for order both within a single state and on the international stage (Buzan, 2014). According to Buzan (2014), the issue of order and justice and ensuring the presence of both even outside the state itself is according to the English School “the basic mechanism or driving force that explains how and why things work the way they do” (Buzan, 2014 p. 26).

There has been much debate between pluralists and solidarists which has been framed in a way where they are both mutually exclusive positions and locked in a zero-sum argument focused on whether the individual or the state should be the priority in international society. In this way, it has been similar to the more famous realist versus liberalist debate within international relations theory, and similar to that debate it is also equally unproductive (Buzan, 2001). However, recent English School writers, most famously Buzan (2001), have argued that the two are not mutually exclusive, instead, pluralism and solidarism are linked, and both represent the extreme ends on a spectrum. On this spectrum, movement is both possible as well as probable. The further division of solidarism into state-centric and cosmopolitan solidarism

(13)

12

further support this view, as state-centric solidarism can be considered to not be as far towards the extreme end of the solidarist side as cosmopolitan solidarism is (although still firmly within the solidarist part of the spectrum) and most arguments framing them as opposites do so by focusing on the cosmopolitan branch. As mentioned before this can also be thought of as a spectrum with one side representing order and one side representing justice. The debate would then be about where on this order/justice spectrum is the ideal place to be in a specific circumstance (Buzan, 2001; Buzan, 2014; Weinert, 2011). This thesis will make use of this more modern take on the English School when it analyses the Dayton Agreement.

Approaches to peace

The main goal of the Dayton Agreement was to stop further bloodshed within BiH. However, in addition to that, there was also a focus on peacebuilding, on creating the groundwork for a long-term stable peace not only in BiH but also in the wider Balkan region (Keil & Kudlenko, 2015). While there is an abundance of scholarly work covering the Dayton Agreement itself there is little theoretical exploring of what type of peace the agreement sets out to create (Keil & Kudlenko, 2015).

Peacebuilding as an academic term dates back to Three Approaches to Peace (1976) written by Johan Galtung. In this article, he explores the different approaches to peace that are possible when a third party intervenes in a foreign conflict. He argues that three different approaches can be used in this type of situation: ‘peacekeeping,’ ‘peacemaking,’ and ‘peacebuilding’ (Galtung, 1976; Keil & Kudlenko, 2015). All three approaches will be explored below.

Peacekeeping is the classical and more realist approach to ensuring peace and Galtung (1976) refers to it as “the dissociative approach” (Galtung, 1976 p. 282). Within this approach, peace is guaranteed by keeping the conflicting parties separate. This separation can be done in a number of different ways. The most obvious way is by keeping them geographically separate, but it can also be done through social (such as by fueling xenophobic tendencies between the different groups) or technological means (creating barriers that are difficult and/or dangerous to bypass). This separation is often also accompanied by either the threat of a foreign military presence being deployed on peacekeeping operations or the actual presence of such a force. This can potentially include the complete destruction of the warring parties’ capacity to engage in warfare (Galtung, 1976; Keil & Kudlenko, 2015). While this approach is efficient in the

(14)

13

short term as the conflicting groups become unable to further their conflict it has a number of weaknesses, in particular in the long term. Peacekeeping operations by military units can (intentionally or not) lead to further devastation within the conflicting region and the approach can lead to further fear within the local population as there is now another armed threat for them to fear. This fear is necessary for the approach as without peacekeepers being a credible threat, they would not serve the purpose of dissuading conflict, but at the same time, this fear does not encourage reconciliation and recovery in the region (Galtung, 1976). Nor does it necessarily contribute to world peace as it involves an armed foreign power intervening in another sovereign nation by force, thus it can lead to further arms proliferation in other nations where the leadership is afraid that a similar fate could happen there (Galtung, 1976). Another weakness of this strategy is that many of its methods are weakened or trivialized by modern technology, physical distance or geographical obstructions are trivial to bypass in modern society and technological hindrances can be met with further technology. Furthermore, warfare can be conducted impersonally at long-distance through missiles and drones and many of these weapons can cause catastrophic damage. Maintaining the groups in isolation is similarly difficult in today’s age of globalization (Galtung, 1976).

A further problem of peacekeeping identified by Galtung (1976) is that by only stopping the conflicting sides from engaging in warfare it fails to deal with the root of the conflict nor with whether or not it is justified or what type of actors are involved in the conflict. Therefore, peacekeeping can lock in an already present system of oppression and/or legitimize aggressive annexations performed by one side or the other. Thus, by enforcing a status quo and keeping either side from engaging in conflict the intervening power would then end up supporting the side of the oppressor and/or aggressor, either directly through functioning as another police force or indirectly through legitimizing the system and/or territorial gains gained through violence. This can occur even if it is the actual intention by the intervening power to be neutral in the conflict (Galtung, 1976). Peacekeeping is thus an example of what Galtung refers to as a ‘negative peace,’ this type of peace is characterized simply by an absence of violence with no reconciliatory forces and relations between the conflicting parties are similar to when they engaged in warring conflict. This also means that peacekeeping can turn into a very long-term engagement for the peacekeeping forces as there is a high likelihood that the conflict will flare up again should the peacekeepers leave (Galtung, 1976, Keil & Kudlenko, 2015).

Peacemaking instead focuses on resolving the current issue that is causing the conflict. Ideally, the third party will then be a neutral mediator helping the conflicting parties to resolve

(15)

14

this issue. This should then eventually lead to a situation where the underlying cause will disappear and there will be no need for weapons or conflict between the parties. While it is more enlightened and long-term viable than peacekeeping there is still a number of difficulties and weaknesses of this approach (Galtung, 1976; Keil & Kudlenko, 2015). The current tension can simply be a symptom of an underlying incompatibility between the interests of the different conflicting groups, should that be the case then it is only a matter of time before another issue causes the conflict to flare up again. In addition to this, resolving even this is no guarantee of peace as hatred between groups can endure even if only out of habit or as an imagined source of one group’s internal woes. Furthermore, an abundance of war materials and fighters can turn into a source of tension on their own and these do not simply disappear because the original tension was solved (Galtung, 1976). Thus, for peacemaking to be successful it is also necessary to either remove the underlying incompatibilities outright, arrange for a compromise, or at the very least distract from them in such a way that they cease to be pursued by the groups, ideally, this distraction would force the different groups to cooperate. Such an event could be an external threat or a natural disaster, but both of these simply replace one problem with another, and finding such an event that does not is difficult. Peacemaking is also more suited to horizontal wars where the different sides are similar in composition and make-up (such as two nations) as opposed to vertical ones (for example an oppressive absolute monarchy and its oppressed lower classes) where anything short of the destruction or disintegration of the oppressors would only continue the oppression (Galtung, 1976). It is also important to note that peacemaking does not necessarily involve less potential force than peacekeeping and the two sometimes overlap. Aiding one of the conflicting groups to complete a total victory over the others or completely destroying the ability for one side (or all) to engage in conflict or pursue their incompatible goals are also ways to solve the issue. That said, these are obviously not ideal methods of doing so and in such a case the conflict could flare up again later after rebuilding (Galtung, 1976).

Peacebuilding is the last of these three approaches and as can be inferred from the name it centers around building structures that aid in ensuring long-term peace and stability, these structures should not only solve any incompatible goals or interests between the different conflicting groups but also prevent any from occurring in the future by offering attractive and credible alternatives to conflict. Thus, peacebuilding efforts focus not simply on peace in the now but also justice, reconciliation, and cooperation in the future (Galtung, 1976; Keil & Kudlenko). Another important part to eliminate the potential for future conflict is the removal

(16)

15

of all types of domination and oppression and a general societal embrace of equity, both between and within groups. In order to accomplish all of this, structures of interdependence between the different groups must be created, structures that lead to positive exchanges between groups. These exchanges should not simply be restricted to the economic aspect but all other aspects as well (Galtung, 1976). Ideally, this would lead the former conflicting groups to be so interdependent that the very idea of armed conflict between the groups disappears. Naturally, this also requires peaceful alternate ways for solving disputes as they are bound to happen on occasion. Eventually, the different groups could even integrate to a level where there is no observable difference between the groups, but this is not a necessity for this approach (Galtung, 1976).

Peacebuilding is then an example of a ‘positive peace’ focused on the long-term viability of the peace project with healing and rebuilding as vital parts as opposed to the negative peace of peacekeeping which is characterized simply by an absence of violence with no reconciliatory mechanism (Keil & Kudlenko, 2015, Galtung, 1976). However, it is possible to combine the different approaches to peace, this would then involve an initial peacekeeping operation to stop the conflict as fast as possible with a gradual move towards peacemaking and eventually peacebuilding. This has the advantage of more immediate results but the disadvantage that this switch from one type of approach to the other is difficult and the legitimacy of the peacebuilding process can be damaged by actions performed in the earlier phases (Galtung, 1976).

Contemporary peace efforts tend to favor either peacebuilding or a combination of the different advances with peacebuilding being a common term employed within UN discourse and seen as important by the UN to ensure the long-term viability of a peace project. However, according to Keil & Kudlenko (2015), the UN’s approach to peacebuilding is more practical and focuses on “three essential features . . . national ownership, national capacity, and common strategy” (Keil & Kudlenko (2015).

(17)

16

Literature review

Gilbert and Mujanović (2015) examine the Dayton Agreement from the perspective of its long-term effects on the inhabitants of BiH to see whether or not it is a success story worth emulating in later peace deals or not. If one looks purely at the fact that there has been no return to armed conflict, then it can definitely be interpreted as such. However, they argue that for a vast majority of the country’s inhabitants Dayton also ushered “in a political-economic order of inequality and dispossession, not only of the means of dignified livelihood but of a future and the agentive capacities to shape that future” (Gilbert & Mujanović, 2015 p. 606). In fact, justified or not, the Dayton Agreement has to many citizens of BiH turned into a symbol of all that ails the country, and a desire for a new one to solve all of these issues is commonplace. Similar examinations and conclusions can be found in Korkut & Mulalić (2009), Perry (2005), Keil & Kudlenko (2015), Hronesova (2015), Gordy (2015), Tuathail et al. (2006), and O’Brien (2010). That said, Gilbert & Mujanović point out that even should a new agreement be negotiated there is a risk that little would change as there is little widespread agreement of what needs to change. Furthermore, the main beneficiaries of the current state of the Dayton Agreement are the same political elite that would be involved in the new negotiations. Since these elites are part of the problem there is a risk that, even if the country received or negotiated an agreement that improves on all the flaws present in the Dayton Agreement, the country might still end up in the same situation that it is in today (Gilbert & Mujanović, 2015; Korkut & Mulalić, 2009). They also note how internationally the Dayton Agreement is seen as a successful achievement of international cooperation while the current state of BiH is viewed as the fault of an endemic culture of ethnic nationalism. This glosses over the fact that the Dayton Agreement gives widespread power and responsibility to international (and internationally appointed) actors. Actors that all have played a part in the development of contemporary BiH (Gilbert & Mujanović, 2015; Korkut & Mulalić, 2009; Keil & Kudlenko, 2015; Gordy, 2015).

Gordy (2015) is one of the harshest critics of the Dayton Agreement and all the institutions it was involved in creating. He describes BiH’s state institutions as “dysfunctional and parasitic while imposing a counterproductive ethnifying dynamic on political activity”, a corrupt political elite that has been allowed to profit off of its population while being legitimized in doing so through the Dayton Agreement (Gordy, 2015 p. 611). According to him the Agreement, in particular the Constitution, sets up is an obstacle to further reconciliation and democratic progress in BiH. He sees the Constitution as a monument to stubbornness,

(18)

17

incompetence, and the unsuitability of the forces that were involved in its creation and application (Gordy, 2015). While Gordy is one of the most vocal critics of the Agreement and its institutions, similar criticism can be found in all previously mentioned literature, especially towards the Constitution found in Annex 4 of the Agreement (Gilbert & Mujanović, 2015; Korkut & Mulalić, 2009; Keil & Kudlenko, 2015; Tuathail et al. 2006; O’Brien, 2010).

Bieber (2006) is the main dissenting voice against a negative view of the Dayton Agreement itself. His conclusions provide a more positive view of the Agreement, especially of what he refers to as the ‘spirit of Dayton,’ i.e. the intentions behind the agreement that shine through but aren’t always the same as the agreement as it is written (Bieber, 2006). He argues that the agreement itself is very flexible and that while progress has been slow it is still present, both from internal BiH actors as well as driven by international actors, most visibly the Office of the High Representative. Even parts that have been completely imposed by international actors, such as the flag of BiH and its (purely instrumental) national anthem have both been accepted and internalized by the population at large (Bieber, 2006). He points out that not only has the Dayton Agreement ensured peace within BiH but there has also been a sharp decrease in violence and crime with ethnic motivations, as well as a decrease in people that fear a return to conflict even though peacekeepers are now mostly absent (Bieber, 2006). That said, he does not ignore the fact that the country is deeply troubled and identifies the same belief among the local population that the Dayton Agreement is the root of all of BiH’s problems. However, he argues that the Dayton Agreement is not to blame for this, rather most of the problems are characteristic of a nation that is recovering from a devastating war and that they are an obstacle that would have to be overcome no matter the peace agreement signed (Bieber, 2006). He also blames the actions of local elites who encourage ethnic tension by using any available loophole within the Dayton Agreement itself to either avoid implementing changes or simply make the changes on a cosmetic level (Bieber, 2006). Many of these elites have shown little investment in either the state or the implementation of the agreement, which has led to a weakening of both. According to him, the biggest flaw of the agreement is not the institutions and rules that it set up but rather the institutions that it recognized and legitimized, referring here to “the two entities (and by extension cantons) and their claim to ethno-territorial predominance of particular communities” (Bieber, 2006 p. 17). While he is unique among previously mentioned authors for his confidence in the Agreement, the criticism of local elites can also be found in a number of the previously mentioned papers (Korkut & Mulalić, 2009; O’Brien, 2010; Keil & Kudlenko, 2015; Tuathail et al. 2006).

(19)

18

Following its implementation in 1995, there were a number of academic articles released which analyzed the structure and content of the Dayton Agreement, primarily from a legal point of view. Among these are Aolain (1998), Szasz (1997), and Gaeta (1996). All of them were critical of the Agreement and identified a number of sections that were either lacking, contradictory, overly complex, or a combination of all of these. They also all provide negative predictions for the future of BiH based on the Dayton Agreement, predictions that concerned many of the problems that later authors have identified in BiH today (Aolain, 1998; Szasz, 1997; Gaeta, 1996). What has not been found in the literature are papers that analyze the Dayton Agreement from a clear theoretical point of view, nor papers that examine the type of peace that the Agreement creates and it is in this area where this thesis seeks to make a contribution.

(20)

19

Methodology

This chapter covers the research design and methodological approach that is used in this thesis. Since the thesis aims to examine and understand a single case in the form of the Dayton Agreement then the natural choice would be to make use of a single-case study. Specifically, what Punch (2014) refers to as an ‘instrumental case study’ this type of study is used to gain “insight into an issue, or to redefine a theory” (Punch, 2014 p. 121) This matches with the goal of this thesis which is to analyze the Dayton Agreement with the aid of theory to gain insight into the Dayton and its effects on BiH. As the focus is on subjective

interpretation and exploration of a phenomenon to gain insight the thesis will also make use of qualitative research design. The analyzed statements within the Dayton Agreement will be counted but there will not be a focus on quantitative data (Punch, 2014). To perform the analysis the Thesis will make use of the Grammar of Institutions developed by Crawford & Ostrom (1995) which is a framework used to analyze institutional statements by breaking them down into strategies, norms, and rules and examining their grammatical structure (Crawford & Ostrom, 1995; McGinnis, 2011). Furthermore, a qualitative text analysis will also be conducted in order to separate the different statements into categories on the English School Order and Justice spectrum envisioned by Buzan as well as into the different

approaches to peace by Galtung (Galtung, 1976; Buzan, 2001; 2014; Kuckartz, 2019). A unique feature of the English School is that it is “theoretically and

methodologically eclectic” (Buzan, 2014 p. 23). This can for example be seen in the previous example where different positions in the pluralist/solidarist debate indicate different

ontological assumptions about international society. It seeks to avoid the stagnant zero-sum game that is rife within especially traditional American IR and instead seeks a holistic understanding of IR. The spectrum by Buzan can be seen as a further development of this. The English School is also unique in the importance it places on history (Buzan, 2014). This necessitates a broad approach to the analysis as well as the inclusion of a significant history chapter covering recent history in BiH and the Balkans.

Materials

This thesis relies exclusively on data analysis of documentary data from documentary sources. This type of data usually comes in the form of written down documents, which is the case here, but this does not have to be the case (Punch, 2014; Finnegan, 2006). The major source of documentary data is the Dayton Agreement itself, but the thesis also makes use of a

(21)

20

wide variety of other literature covering the Dayton Agreement, BiH, and/or the Balkans as well as covering the theories and methods that are being utilized. The Dayton Agreement represents a primary source of data, the original source, and data being analyzed. The other sources are secondary sources with research that have been performed and later published by other authors. While many of them involve an analysis of the same primary data source, it still involves the repurposing of that data to suit the purpose of this thesis (Punch, 2014; Finnegan, 2006). Fortunately, there is a wide variety of literature published that covers the Dayton Agreement and recent developments in BiH.

A Grammar of Institutions – ADICO

Sue Crawford and Elinor Ostrom (1995) developed the grammar of institutions in an effort to provide a clear and consistent language for the analysis of institutions within political science. The institutions are broken down into what they refer to as “institutional statements” and these are then analyzed using the grammar of institutions (Crawford & Ostrom, 1995; Schlüter & Theesfeld, 2010; McGinnis, 2011). According to Crawford and Ostrom, there are three different types of institutional statements. These are strategies, norms, and rules. They argue that it is important to differentiate between them as they have different foundations, treating them all as the same “fails to clarify the difference between shared advice based on prudence, shared obligations based on normative judgments, and shared commitments based on rules created and enforced by a community” (Crawford & Ostrom, 1995). The three different types represent different parameters that explain why an actor either does or does not follow the institution and provide for more options than the simple threat of being punished for not following it (Schlüter & Theesfeld, 2010). For this analysis, Crawford and Ostrom provide five different sub-parts that form the grammatical syntax of the grammar of institutions. These are Attribute (A), Deontic (D), aIm (I), Conditions (C), and Or else (O). This is then referred to as the ADICO syntax. Not all institutional statements contain all parts of the syntax and it is the presence (or lack) of these different components that determine if it is a strategy, norm, or a rule. A strategy contains only an attribute, aim, and condition, and is therefore referred to as AIC within the syntax. Norms include the attribute, deontic, aim as well as condition (ADIC) while rules contain all components of the syntax; attribute, deontic, aim, condition, or else (ADICO) (Crawford & Ostrom, 1995; Basurto et al. 2010; Shlüter & Theesfeld, 2010; McGinnis, 2011). For the sake of clarity, the different components will be explained below.

(22)

21

A (Attribute): The attribute (A) is the person (or persons) or organization, e.g. the agent(s), to whom the institutional statement is addressed. The attribute also usually includes descriptive elements to further specify the subject of the statement. If it is an individual then this description could include parameters such as age, education, sex, position in society/workplace, etc. For example, an attribute could address ‘all high school-educated men over the age of 18 who are unemployed.’ For organizations, the description would instead focus on organizational descriptions, such as size, location, or type of organization, e.g. ‘all organizations with more than 25 employees engaged in international shipping’. If there are no descriptive elements for the individual or organization then the default value for the attribute is usually everyone. The attribute could also be a state (Crawford & Ostrom, 1995; Basurto et al. 2010; Siddiki et al. 201; Schlüter & Theesfeld, 2010).

D (Deontic): The deontic (D) is the operator of an institutional statement. A strategy (AIC) does not have a deontic, but a norm (ADIC), and rule (ADICO) does. A deontic either permits, obliges, or forbids an action that the attribute then can, should, or is forbidden from doing. Crawford & Ostrom (1995) focus on the words ‘may, must and must not’ as the indicators of a deontic within an institutional statement but Basurto et al. (2010) and Siddiki et al. (2011) all argue that they can come in many different forms as long as they have the same meaning of either permitting, obliging or forbidding something. Furthermore, different word use can imply different levels of prescriptive force. E.g. should or should not could imply less prescriptive force than must or must not (Crawford & Ostrom, 1995; Basurto et al. 2010; Siddiki et al. 2011, Shlüter & Theesfeld, 2010). Of the three different operators, the most ambiguous one is ‘may’ as an institutional statement that simply allows for something (rather than obliging or forbidding it) might not necessarily change much. That said a ‘may’ deontic still has important implications in that it indicates that an action is both not forbidden but also promoted. For example: ‘Junior employees may criticize the board’ is an example of a deontic that both permits something but also has important implications for the relationship between junior and senior employees (Crawford & Ostrom, 1995). ‘Fishing companies must obtain a license before engaging in fishing’ is an example of a must deontic while ‘Farmers are prohibited from dumping waste into the local lake’ is a must not deontic.

I (aIm): The aim (I) refers to the action or outcome of the institutional statement. What does the statement indicate that it wants the agent to do (or not do)? Thus, the aim must be something that both can be done as well as can be avoided. The aim can also include descriptions of what the action is, what its goal is, and how it is performed should they be

(23)

22

necessary. In the two previously mentioned examples the aim is that ‘fishing companies must obtain a license’ (I) and ‘farmers are prohibited from dumping waste (I) into the local lake’ (Crawford & Ostrom, 1995; Basurto et al. 2010; Siddiki et al. 2011; McGinnis, 2011).

C (Condition): The condition (C) concerns under what conditions the institutional statement is applicable. Most often they appear in the form of ‘when’ and ‘where,’ but other words can also apply as long as they involve either preconditions or limits relating to the aim of the statement. In addition to being temporal, a conditional ‘when’ can also deal with special circumstances (such as a weather event or catastrophe) or during a legislative or judicial process. Similarly, a ‘where’ can be more than a geographical determinant (such as jurisdictional). Similarly, to the attribute, the condition is always present, even if it has not been written out. According to Crawford and Ostrom (1995) if conditions are not mentioned then “the default value for the condition is at all times and in all places” (Crawford & Ostrom, 1995 p. 585; Basurto et al. 2010; Siddiki et al. 2011; McGinnis, 2011). E.g. if we change the former example to ‘Farmers must not dump waste’ then even though a condition is not mentioned it is present and defaults to at all times and in all places while in the original example the condition is into the local lake.

O (Or else): The ‘Or else’ (O) part of an institutional statement explains how non-compliance with the statement will be punished. The ‘Or else’ is only present in a rule (ADICO) and not in a strategy (AIC) or a norm (ADIC). An Or else can wary wildly in the severity of the punishment for breaking the rule, from a small fine to imprisonment or worse. In order for an or else to function, there must also be another actor or actors charged with monitoring compliance or non-compliance with the rule, the most obvious example of this is a state but the monitor does not have to be connected to a state and monitoring duty can also be left to everyone. Furthermore, there must also be actors in charge of delivering the punishment that is stipulated by the or else. The monitor and the one delivering the punishment can in certain situations be the same actor(s) while in others the monitor reports to someone else (Crawford & Ostrom, 1995; Schlüter & Theesfeld, 2010). Finally, the rights and responsibilities of those who monitor and punish should themselves be covered in other norms or rules that have been agreed upon through some form of arena for the purpose of processes for collective action, especially since many punishments are actions that are normally prohibited in other rules. Once again this can be connected to a state (such as a parliament or a court) but it does not have to be, a private organization or similar could have its own form of rules that have been agreed upon (Crawford & Ostrom, 1995; Shlüter & Theesfeld, 2010; McGinnis, 2011). For example:

(24)

23

‘Drivers must follow the speed limit or be fined by police’ (O). Here the ‘Or else’ is be fined by police and the police is here both the monitor and the one delivering the punishment. The rights and duties of the police in this situation are covered in other norms or rules.

Since strategies only contain an attribute, aim, and condition they have no real form of sanctioning inherent to them and their enforcement thus rely mostly on an individual’s prudence and the moral values that said individual has internalized. A norm with the added deontic relies on a common understanding within a group and enforcement through social means of what is considered the right or wrong thing to do in a given situation. Thus, it requires an analysis of the belief system of a group in order to understand the normative obligations an individual within said group has. Finally, rules with the added ‘Or else’ are based on a group’s understanding that certain actions that go against what is demanded or forbidden should be punished through material means if discovered and that those that enforce said punishments themselves are governed by mechanisms that allow them to do so (Crawford & Ostrom, 1995; Shlüter & Theesfeld, 2010).

Later comments, criticisms, and further developments of the syntax

Following its initial release by Crawford and Ostrom (1995) the grammar of institutions has received comments and criticism from other academics, several of whom have suggested further developments to it. Some of this is of interest to this thesis and will be explored here. To begin with, there are the problems that the syntax expects a certain level of clarity from the institutional statement that can often be missing. For example, Crawford and Ostrom (1995) recognize that the attribute and condition may be missing from the statement and suggest that in those cases the default attribute is everyone and the condition is at all times but other parts of the syntax can also be missing which leads to the problem of implicit versus explicit components (Crawford & Ostrom, 1995; Basurto et al. 2010; Siddiki et al. 2011; Schlüter & Theesfeld, 2010). For example, the attribute could be missing not because the default value is at all times but because it has already been stated in an earlier connected statement, (institutional statements are often part of a larger system) this can also apply to all other components of the syntax. In that case, the component is implicit rather than explicit. Basurto et al. (2010) argue that a deontic can be implicit and if one does not recognize this then there is a risk that the analyst counts too many strategies. However, determining an implicit deontic can often be difficult to do which can also lead to the reverse issue. They also argue that there can be a confusing overlap between the aim and the condition, “when does the description of the action shift to qualifiers of the action?” (Basurto et al. 2010 p. 535).

(25)

24

Similarly, results from other analysts have shown that most legislation is often written as a strategy or norm rather than a rule, i.e. there is rarely an ‘Or else’ stated within individual statements. However, most legislation should be viewed as a rule rather than a norm because breaking a law usually involves an ‘Or else’ even if not stated. Said ‘Or else’ is then present in an institutional statement elsewhere in the legislative system. Thus, an ‘Or else’ can also be implicit and within a legislative or legal system, it usually is, more often than not. This leads to a problem that is only worsened by the (often) intentional vagueness of legislative statements (Basurto et al. 2010; Siddiki et al. 2011; Schlüter & Theesfeld, 2010, Novo & Garrido, 2014). However, accounting for implicit an ‘Or else’ is difficult as it requires finding it within other statements of a legislative framework, that is often enormous, and leads to a potential system where everything becomes classified as a rule. Therefore, it can often be useful to ignore the implicit ‘Or else,’ especially within legislative and legal institutional statements and pay special interest to the statements that do have an explicit ‘Or else,’ but this is an area of the syntax that could use further development (Basurto et al. 2010; Schlüter & Theesfeld, 2010).

Schlüter and Theesfeld (2010) also point out that what Crawford and Ostrom refer to as strategies (AIC) also contain norms, similar to what the authors refer to as norms (ADIC), the difference is whether they are internal or external. A strategy relies on norms that an individual has internalized in some manner, such as from parents, education, or the extended social environment where they have grown up. Thus, even strategies can contain sanctioning, but it is based on internal pressure (Schlüter & Theesfeld, 2010). What Crawford and Ostrom refer to as norms on the other hand involve social norms and external pressure, i.e. what the surrounding people consider the right or wrong thing to do. Depending on when and where someone has been raised, by whom and where they have lived before, the internal and external norms can mirror each other or they can be completely different, especially for an individual that now resides in a completely different culture from where they are from (Schlüter & Theesfeld, 2010).

On a similar note, it can be said that all types of institutional statements can involve some type of ‘Or else,’ in the case of a strategy (AIC) there could be an internal or else based on an individual’s internalized norms and it could be either positive or negative. The individual does this because they feel that it is the right thing to do and doing so makes them feel good about themselves, alternatively not doing it makes them feel bad. The same can be said for the norm (ADIC) here the social norm can lead to the presence of an external ‘Or else’ that is either positive and/or negative. If the person lives up to the social norm of what is the right thing to

(26)

25

do then there would be positive reinforcement from the group while there would be a negative reaction if they do not live up to it (Schlüter & Theesfeld, 2010). The ‘Or else’ involved in the full rule (ADICO) on the other hand, can only ever be negative and relies on material punishments (although the severity varies) should non-compliance with the statement be discovered. While for most cases it can be simplified in such a manner as seeing the ‘Or else’ in a rule as more powerful than one in a norm which is stronger than the one in a strategy it is not always true. For example, an individual with considerable financial assets might consider a negative reaction from their social group to be a more powerful ‘Or else’ than a fine of what is to them an insignificant sum. There is also the possibility that the different varieties of sanctions work in opposing directions, a rule that is considered particularly unjust within a social group might lead to a situation where an individual feels good and gets positive reinforcement from their social group for breaking said rule while still risking the material punishment for breaking the rule (Crawford & Ostrom, 1995; Basurto et al. 2010; Siddiki et al. 2011; Schlüter & Theesfeld, 2010).

Qualitative Text Analysis

Different forms of qualitative data can differ hugely from each other and can be highly complex, but in this thesis, the focus is on qualitative text analysis. Like the data, there is a large variety of different methods to analyze qualitative content, but one aspect of it that is important no matter what method is chosen is that of categories or codes. According to Kuckartz (2019) categories are “of crucial importance for effective research, not only in their role as analysis tools but also insofar as they form the substance of the research and the building blocks of the theory the researchers want to develop” (Kuckartz, 2019 p. 183; Kuckartz, 2014; Mayring, 2014). According to Kuckartz (2019), a variety of categories can be found within research on social science. These include, but are not limited to analytical categories, evaluative categories, factual categories, formal categories, natural categories, thematic categories, and theoretical categories (Kuckartz, 2019). Of special interest to this paper are analytical, evaluative, and thematic. A thematic category relates to a certain theme, such as a topic or event. For example, xenophobia, the American civil war, order, or justice. An evaluative category is related to a scale where a thing is evaluated, this can be anything from press freedom to autism. The researcher is the one who has to determine the criteria and then evaluate the data accordingly. Finally, analytical categories are a product of the process of analyzing the data and go beyond a simple description of what is in the data. However, there is not a clear divide

(27)

26

and the different categories can blend into each other or assist each other, for example, data that has been assigned a theme or otherwise evaluated can then be considered a part of the analytical (Kuckartz, 2019).

According to Kuckartz (2019), there are different ways to create these categories. The most used ones are ‘concept-driven’ (also known as deductive), ‘data-driven’ (inductive), or a mixed method of the two (Kuckartz, 2019). If the categories are concept-driven, then they originate from the concept of the paper, so either the theory that is being tested, the literature that is being investigated, the problem formulation or research question of the paper itself, etc. (Kuckartz, 2019). Data-driven categories, on the other hand, are developed during the research and categorization process. Initially, no categories are assumed, and the coding process should then naturally form the categories, that can also be re-organized and changed during the research process. Finally, mixing the two usually implies a pre-existing framework for categories that have been formed using the concept-driven approach but then all data is instead evaluated using the data-driven approach. (Kuckartz, 2019, Kuckartz, 2014, Mayring, 2014). This thesis will utilize concept-driven categories.

Limitations and Validity

There are a number of limitations inherent to the approaches used in this paper. Using the grammar of institutions means that the researcher must correctly identify the presence (or absence) of the different subcomponents in the institutional statement (attribute, deontic, aim, condition, or else) in order to classify the statements as strategy, norm, or rule. A failure to do so can lead to incorrectly classifying a statement as something else which can have a negative effect on the reliability of the results. Ideally, when using this ADICO syntax a paper should use multiple coders and have them all code the same sections individually, if there is a high percentage of inter-coder reliability (i.e. the coders agree on the different components in the statements) then it is a sign of reliability while a low percentage is a problem for the paper’s reliability (Basurto et al. 2010; Siddiki et al. 2011). As this thesis has a single author then any results based on the author’s coding should be seen as an indication of a result that would need to be examined by more coders to ensure its reliability. The thesis will also try to account for implicit subcomponents in addition to explicit ones which further increases the difficulty of operationalizing the syntax. (Basurto et al. 2010; Siddiki et al. 2011).

Similar limitations are present in the text analysis since the researcher has to correctly classify the data into the different categories and failing to do so or having a weak category

(28)

27

system affects the reliability of the results. Thus, using multiple coders would once again increase the reliability of both the results and the categories themselves (Kuckartz, 2019; Basurto et al. 2010) The lack of multiple coders could, therefore, affect the validity of the paper and in order to ensure that the results of the research are valid, the data would need to be coded by other researchers. Since the analysis is based on a single-case study one can also question the external validity, i.e. the generalizability of the results. However, as a qualitative single-case study a deep understanding of the single-case in question is valued above generalizability (Mayring, 2014). It is also important to note the limitations of focusing on the Dayton Agreement as the primary source of data. The Agreement was developed through a lengthy peace process that involved many negotiations conducted behind closed doors. All these negotiations have not necessarily made their way into the Agreement but could still have affected its implementation or the actors involved. Similarly, the implementation of it may vary for a number of different reasons. Thus, if an issue in the Dayton Agreement is identified by this analysis that does not mean that this issue was implemented in practice.

Implementation

This thesis will analyze the Dayton Agreement using the previously mentioned ADICO institutional grammar and qualitative text analysis. Initially, every annex will be analyzed individually, and its paragraphs will be analyzed as institutional statements. For the text analysis, there will be concept-driven (deductive) categories and these will be: an evaluative category based on the order and justice spectrum where the statements will be evaluated as mostly relating to order, justice, or a mix of both, and an analytical category where they are placed as either, peace-keeping, peace-making, or peace-building based on the approaches developed by Galtung (Galtung, 1976). Every statement of each annex will also be analyzed using the ADICO syntax. This combination of methods is used to discover differences (if any) between the concepts of order and justice in the Dayton Agreement. The results will then be used to determine if there is an indication of a problem of order and justice within the agreement itself which has led to the mixed results of its implementation. Findings will then be discussed in the discussion section.

(29)

28

Historical background: BiH and the Balkans

To understand the roots of the conflict in BiH it is necessary to examine recent history in that region of the world which is the purpose of this chapter.

As the tenets of nationalism spread through Europe in the 1800s the Balkans quickly proved how dangerous these ideas could be in the wrong circumstances. At the onset of the 1800s, most of the Balkans still belonged to the Ottoman Empire, but the powerful days of the Ottomans were far behind them and now the Empire was in decline. During the Ottoman period until this period the area had been no stranger to violence and war but hostility between different ethnic groups was rare (Glenny, 2012). The Empire consisted of a vast amount of different cultures and ethnicities, this was especially pronounced in the Balkans were different faiths, ethnicities, and cultures lived side-by-side, and in many regions, there was no clear ethnical majority, BiH was one of these. By this point, the Ottoman Empire had a system where its population was divided into millets, (millet translates to nation), but it did not resemble what is today considered a nation. The millet system was instead based on religious grounds if the roots of the religion could be traced back to the old testament (Glenny, 2012). Thus Muslims, Orthodox Christians, Catholics, and Jews belonged to different millets. Because of this, a Bosnian Muslim and Turkish Muslim were considered a part of the same millet and so was a Serb Orthodox and Pontic Greek Orthodox even though culturally, ethnically and linguistically they were nothing alike while the Bosnian Muslim and Serb were similar on all those levels (Glenny, 2012). Each millet had its own rights, responsibilities, and religious leadership. Intermarriage and interbreeding between millets were forbidden unless one converted but within the millet, it often crossed cultural and linguistic boundaries. Thus, the main attribute governing one’s identity in the Empire at this time was religion and the most favored millet was the Muslim (Glenny, 2012).

Conversions were rarely pushed within the Ottoman Empire and the millets were often left to their own devices as long as they were loyal. However, because the Muslim millet had more rights and was more favored, an individual of the Empire could rise in status and increase their opportunities by converting to Islam (Glenny, 2012). The border regions were the exception to the rule of no conversions and locals here were encouraged to convert. In addition to this, Muslim settlers were encouraged to settle in these regions. This was done with the logic of ensuring maximum loyalty to the Empire in the most vulnerable areas (Glenny, 2012). The biggest Muslim groups in the European part of the Empire could be found in Albania and BiH, both of which had mostly formed through conversion. Being on the border with Austria, the

(30)

29

Muslims in BiH were considered extra important and thus, were also allowed considerable autonomy. However, there were still considerable communities of Orthodox Serbs and Catholic Croats in BiH, but the millet system ensured that the groups were segregated. Most Muslims lived in the cities and were on average wealthier and better educated, while the Orthodox communities dominated the countryside and were poor. The Catholic community lived mostly on the border with Croatia where they enjoyed good relations with the ruling Catholic Austrian Empire (Glenny, 2012).

This confusing tapestry of nationalities, cultures, and religions meant the potential for conflict in the region was high when ideas that each nationality deserved their own land and nation started to spread. In addition to this, the region was a border area for three different great powers of the time, the Ottoman, Austrian, and Russian empires (Italy would also start to involve itself later) so rebellious and/or nationalist groups in the region had no difficulty in finding a foreign sponsor, either from a power desiring the land they were on, or one that wanted to weaken another (Glenny, 2012).

To reverse its decline, the leadership of the Ottoman Empire eventually embarked on a system of modernization and centralization, the tanzimat (reorganization), modeled on those already done elsewhere in Europe. Since the Empire lacked a clear nationality to promote, it also invented a new supra-national identity, Ottomanism (Glenny, 2012). However, this was seen as a threat to much of the already existing establishment, especially the Muslim millet, the Janissaries, and Muslim landowners, thus, the Empire became plagued by unrest, coups, and rebellions (Glenny, 2012). BiH contained some of the most independently minded Muslim landowners, who were already loosely connected to the Empire’s Muslim leadership as they were culturally and linguistically more similar to the Christian peasants that surrounded them than the leadership in Istanbul. This tinderbox leads to the Empire being rocked by rebellions in the mid-1800s (Glenny, 2012). Rebellions without a nationalist character would often get one later in order to either mobilize more fighters, justify territorial gains, or get sympathy for their struggle among the great powers. The outcome of these rebellions showed that the easiest way for the rebels to succeed was to obtain a strong foreign power as a backer for their cause so most of these movements started courting great powers, who happily deepened their involvement in the region (Glenny, 2012).

The modernization process in combination with the revolts and some disastrous wars meant that the Ottoman Empire was in dire need of money, a problem that foreign banks,

Order and Justice in the Dayton Agreement : An English School Analysis of the General Framework Agreement for Peace in Bosnia and Herzegovina