! ! ! ! ! Lund University ! Centre for Middle Eastern Studies ! Master’s Program in Middle Eastern Studies ! Spring Semester, 2015! ! ! ! ! Revisiting Civil-Military-Relations Theory! - the Case of the Kurdish Regional Government of Iraq! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Verena Gruber ! Supervisor: Spyros Sofos ! ! ! ! ! Acknowledgements! ! First, I would like to express my gratitude to my immediate supervisor Spyros Sofos, and by extension in an unofficial capacity William Park, who both supported my research with countless consultations and emails. For my field work, my foremost gratefulness goes to Dr. Borhan Yassin, who, in his capacity as a lecturer at Lund University, helped inspire this research and eased my introduction to the region by providing me with a network of his friends who not once failed to support me during my stay. A special thank you also goes to my personal friends and colleagues, Arez and Bahra, who first introduced me to the political and cultural depth of their region, and COL Dennis P. Chapman who never grew tired of answering my many questions on the military institutions.! In Kurdistan, I was able to meet many politicians and personalities on a regular basis. I would like to thank in particular (in alphabetical order and without ranks and titles) Ako Wahbi and his friends, Aziz Barzani, Babakr Draey, Botan Tasheen, Chia Nashirwan Mustafa, Dawn Derkle, Dilshad Sharif, Dlawer Ala’Aldeen, Falah Mustafa Bakir, Farhad Ameen Atrushi, Fazel Hawrami, Galip Dalay, Helgurd Hikmet Mela Ali, Muhammad Ali Ahmed, Mohamad Tofiq Rahim, Hazhar Ismail, Hemin Hawrami, Hoger Hassan, Hoshyar Omar, Khabat Abdulla, Mala Rashad, Mohammad ‘Hama’ Radsh, Mohammed Ihsan, Mustafa Said Qadir, Nahro Asuad, Najat Ali Salih, Nawzad Shwani, Peshawa A. Muhammed, Qadir Hama Jan Aziz, Rizgar Hajy Hama, Said Salah, Said Jawad, Salah Dulmany, Salah Mirani, Saman Jalal, Sheikh Jafar Mustafa, Shorsh Haji, Sardar Aziz, Sarhang Hamid Barzinji, Yusuf Mohammad. ! But apart from the great conversations I enjoyed, my stay would not have been possible without the limitless help from countless people. I would therefore like to pay a special tribute of gratitude to my friends who made my stay memorable experience. In particular and alphabetical order, I thank Alan Salih, Ali Ismail, Ali Jameel, Anthony Sinnott, Bert Stenvert, Chris Weaver, Craig Anderson, Danny Kerbage, Dastan Ali, Dawood Atroushi, Dennis Mabura, Fakhraddin S. Muhammad, Harald Wittmann, Hermoine Gee, Kavout Mahy, Kurdo Hussein, Lokman Atroshi and his team, Malamstafa Khoshnaw, Mèlisande Genat, Miran Burhan, Newroz Mawlood, Peter Jackson Choi, Ranj Salah, Rawezh Kamaran, the team of the Rise Foundation, Silav Ibrahim, Sofia Barbarani Campello, Susan Barwari, Tahsin Kader Ali, Till Paasche, and Zhyar Ahmed Balak.! And last but in no way least i want to thank my family, particularly my parents, for all the innumerable support they have extended to me, for enduring patiently through their fears for my safety, and for trusting and believing in me every step along the way.! ! ! ! ! Abstract! ! The Kurdish Regional Government (KRG) of Iraq intends to unify and institutionalize its politically- divided, party-based military. To conduct a unification successfully, understanding the current system in place is necessary. An analysis of the civil-military-relations (CMR) helps achieve this insight. Deconstructing the KRG case, however, limits of the explanatory power of CMR theory are reached. Identifying these shortcomings and raising questions for future theory revision is consequently the ambition and purpose of this thesis. ! Based on the existing analysis of the political system by Stansfield (2003) and the security sector by Chapman (2011), this research addresses the lack of attention given to the interrelation of those fields. Altering Stansfield’s application of Lijphart’s power-sharing idea by employing the Prisoner’s Dilemma model, the military implications of the political division in the Kurdish region can be explained. The resulting hypotheses provide the basis for deconstructing the civil-military relations of the case study and help determine the deficiencies of prevalent CMR theory. In due course I demonstrate how, instead of a gap between civilian and military spheres, a division between the official institutions (“structure”) and an informal “system” defines the CMR of the region. With the “system” yielding more decision-making power than the “structure”, its conceptualization manages to highlight that prevalent CMR theory is not capable of conceptualizing a system based on personal networks which run parallel to the institutional structure that theory tends to focus on.! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Abbreviations! ! CMR! ! Civil-Military-Relations! ISIS! ! Islamic State of Iraq and Syria, also known as the Islamic State of Iraq and the ! ! ! Levante or the Islamic State! KDP! ! Kurdish Democratic Party! KRG! ! Kurdish Regional Government*! KR-I! ! Kurdish Region of Iraq*! MoP! ! Ministry of Peshmerga! NGO! ! non-governmental organization! OTR! ! Information given “off the record” - details of source will not be disclosed! PDK ! ! see KDP! PUK! ! Patriotic Union of Kurdistan! SSR! ! Security Sector Reform! UPF! ! Unified Peshmerga Forces! UN! ! United Nations! UNSCR! United Nations Security Council Resolution! US! ! United States of America! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! * The distinction between the KRG and the KR-I lies in the first referring to the government institutions while the later refers to the territory.! ! ! Table of Content! ! Introduction ………………………………………………………………………………………………… 1 ! Delimitations and Limitations ……………………………………………………………………. 1 ! Statement of Hypothesis …………………………………………………………………………. 3 ! Disposition …………………………………………………………………………………………. 4 ! ! Civil-Military-Relations Theory: Three Debates …………………………………………………………. 5 ! Institutionalization vs. Convergence …………………………………………………………….. 6 ! Non-Western CMR ……………………………………………………………………………….. 7 ! “New Wars” ………………………………………………………………………………………… 8 ! ! The Case Study ………………………………………………………………………………………….. 11 ! A Contextual Analysis - the History of Kurds in Iraq …………………………………………. 12 ! The Political Division - Two Separate Administrations ………………………………………. 14 ! ! ! ! ! ! Methodology ………………………………………………………………………………………………. 18 ! Design of the Study ……………………………………………………………………………… 18 ! Method and Analysis …………………………………………………………………………….. 20 ! ! Findings and Analysis!……………………………………………………………………………………. 24 ! Structure …………………………………………………………………………………………. 25 ! Goals and Problems ……………………………………………………………………………. 29 ! System ……………………………………………………………………………………………. 34 ! Concluding Statements …………………………………………………………………………. 43 ! ! Revising Theory ………………………………………………………………………………………… 43 ! ! ! ! ! ! ! Conclusion ………………………………………………………………………………………………… 47 ! ! Bibliography ……………………………………………………………………………………………….. 49 ! ! Appendix ………………………………………………………………………………………………….. 52 ! ! ! ! Introduction! ! The Kurdish Regional Government (KRG) of Iraq intends to unify its politically-divided military forces, Peshmerga. Already considered in 1991 (Chapman 2011: 97), the unification of the party- aligned forces got a new urgency with the Islamic State (ISIS) surging through Iraq and towards the Kurdish populated areas. President Barzani marked this urgency by declaring an intent to unify all forces within six month in August 2014 (Mahmoud 2014). ! However, taking two politically and at times administratively separate units, like the Peshmerga, with the intention of merging them into one institutionalized army under the Ministry of Peshmerga (MoP) and the Kurdish government (KRG) requires more than structural consolidation. Accounting for problems of historical mistrust and diverging claims to power, a study of the current system aids the identification of what form of system and structure1 might effectively work for the region. By relating the political and military division of the Kurdish region in Iraq (KR-I), the initial purpose of this thesis was to understand its civil-military relations (CMR) through CMR theory. Deconstructing the case study, however, limits of the explanatory power of CMR theory are reached. Identifying these theoretical shortcomings is consequently the primary ambition of this thesis. ! A dual added value for both a further understanding of the Kurdish region as well as a revision of CMR theory is provided. The emphasis however lies on the revision of theory. In addition to this first accentuation, further delimitations and limitations have to be addressed before formulating the central hypotheses and research questions.! ! Delimitations and Limitations! Given the specialized nature of the presented research, several delimitations had to be taken. With every delimitation for the field work, however, an explanatory limitation of the thesis necessarily follows. Three choices are defended: 1) the decision to view the Kurdish region in a geopolitical vacuum, 2) the focus on the biggest parties within the KRG, and 3) the concentration on the “elite” within parties and military divisions.! Starting with the first point, a distinction between praxis and theory is called for. In praxis, the nature of the KR-I’s geopolitical location in the region denies a contemplation of KRG matters in isolation. Being a mountainous hub of relative wealth and oil richness in the landlocked area between the Turkish border to the north, the Iranian border to the east, and the rest of Iraq and a 1 The differentiation between system and structure is vital in this research. While at times used interchangeably, the distinction of the terms is a core part of the findings and argument of this paper. Structure, in this thesis, refers to a formal form of civil-military relationship. Institutionalized hierarchies, chains of command, and official links of communication are understood to constitute such formal structures. System, in contrast, indicates the informal aspects of contact, exchange, and networks that exist sometimes parallel, sometimes intersecting, and sometimes entirely absorbed within the formal structure. "1 part of Syria to the south and west2, the KR-I is both influenced by and influencing the surrounding political balance, economic interests, and military considerations. However, this research is not practically but theoretically orientated. Therefore the wider context can be largely neglected. Questions such as the effect of the intended Kurdish military transformation on the surrounding Kurdish territories, on the balance of power in the region, or even on the state-creation project of the KRG have to be subject to subsequent research endeavors. The strict focus on the KR-I is the first and most crucial delimitation. Thereby, the KR-I is defined as follows: The KR-I includes the three governorates of Duhok, Erbil, and Sulaymaniyah, which are officially recognized to constitute the territory of the autonomous KRG. In addition, the territorial gains of the Kurdish military forces during the time of the field work are included in the definition3. These territories include the northern parts of the district of Diyala all the way to the city of Jalawla and Sa’adi, the eastern parts of the governorate Kirkuk, and the northern and eastern borders to the province in Ninawa all the way to the Syrian border and Shengal mountain, but excluding the city of Mosul.! A second delimitation is the exclusive focus on the three biggest parties within the KRG: the Kurdish Democratic Party (KDP), the Patriotic Union of Kurdistan (PUK), and Gorran (Change Movement). Their selection is elaborated in Methodology and is based on the political and military division of the region presented in Case Study. The concentration aims at simplifying a complex issue. ! A last delimitation is taken within these parties: The priority lies on the political and military elites because the hierarchical nature of the political and military structure allows for a better assessment of CMR in a top-down perspective. A reasoning for this assessment and the definition of “elite” is provided in Methodology.! Following the delimitations, I now point to the consequentially arising limitations of this thesis. First, the research is unable to incorporate external influences of neighboring states, adjacent Kurdish regions, and the effects of the ongoing war against ISIS into the examination of CMR. Second, this thesis is not analyzing the effects the proposed military unification has on the surrounding area. Third, other security forces, such as the intelligence, Zerevani, and security (Assayish) are also disregarded, as they do not fall under the institutional framework of the Ministry of Peshmerga and are not yet subject to unification.! And last, this thesis is unable to reflect interests, desires, and opinions of the Kurdish population. This perspective is overlooked as this research focuses on the structural and systemic aspects of the civil-military relations in the KR-I.! 2There are other Kurdish regions in Turkey, Iran, and Syria. Their importance can not be underestimated. However, the focus in this study lies solely on Iraq; therefore “Kurdish region” always refers to the KR-I in this paper. 3As analyzed in Gruber (2014), these additional territories comprise the dominantly Kurdish populated areas adjacent to the KRG. They are are sometimes referred to as “disputed territories” (Iraqi Federal Constitution 2005/06). See Appendix. "2 Statement of Hypothesis! Having accessed the basis of this thesis, I now present the guiding hypotheses and research questions. Reviewing the literature on political and military aspects of the KR-I it becomes apparent that the interconnection between the two spheres is largely unexplored. The most prominent analysis of the political situation in the KR-I is conducted by Stansfield (2003). However, while mentioning the Peshmerga in both a historical and timely description, he fails to address the connection between the parties and their forces in detail. Instead his emphasis rests on the accommodation of elite-dominated parties. Lortz (2005) and Chapman (2011) in contrast focus solely on the security sector of the region. Thereby the first takes a historic approach while the later analyzes the current military set-up and capabilities. Both, however, remain dominantly descriptive and miss further analyses of the provided information. ! Reviewing the literature, I conclude that bridging political and military aspects leads to an added value in the comprehension of the region. Due to the interrelated and mutually enforcing relationship between the two, with political mistrust increasing the interest in maintaining party- based security forces which in turn reinforces the already existing mistrust, analyzing the relationship between the spheres leads to further insight in the military transformation process and the elite accommodation advocated by Stansfield. In addition, this perspective manages to overcome an aspect Stansfield misses in his analysis: The process of elite-accommodation as well as the unification of the Peshmerga occurs in an environment of quasi-anarchy. Anarchy, as used in international-relations theory, refers to a situation of there not being an overarching ruling and guiding authority, as it is assumed to be the case in international relations (Morgenthau 2006). While hard to quantitatively prove, the influence of ‘Iraq’ over the Kurdish population and territory has been almost inexistent since 1991. This lack of superior administration, ruling, and governing within the state structure of Iraq (ergo quasi-anarchy) enforces the already existing mutual mistrust by creating additional security concerns. The inclusion of these concerns into the understanding of the political decision-making processes, particularly when unifying the party-based forces, presents an additional lens for explaining the transformation process in the KR-I. One could even argue that both parties are stuck in a constant Prisoner’s Dilemma. In fact, applying the Prisoner’s Dilemma model accounts for the preservation of the military wings of each party.! ! H1a: The merging of military forces in the KRG occurs in a setting of quasi-anarchy.! ! H1b: The preservation of the military wings of each party and potential problems in their ! ! unification can be explained by applying the Prisoner’s Dilemma model4.! In addition, the application of Prisoner’s Dilemma also allows for a first conclusion towards explanatory limits of CMR theory: If two parties merge their military forces in a condition of quasi- anarchy, due to the prevailing security concerns, the involved parties are likely to hold on to the security guarantee of their own loyal military units. By focusing on an institutional separation of 4 For more on the concept of Prisoner’s Dilemma see Case Study. "3 civilian and military spheres, however, CMR theory is unable to account for a plurality of military forces and their diverging loyalties.! ! H2a: The Prisoner’s Dilemma situation effects the civil-military relations of the region.! ! H2b: The existing body of CMR theory is insufficient in explaining the Kurdish case.! With the inability of theory to account for the existing plurality of civil-military relations in the Kurdish case, an in-depth analysis of the system in the region is pursued in order to access currently missing variables necessary to the understanding of non- and low-institutionalized civil- military relations. The guiding research questions for this endeavor, therefore, read as follows: How can the plurality of civilian and military actors in the Kurdish region of Iraq be accounted for in the study of their civil-military relations? And how does this altered perspective affect CMR theory?! ! Disposition! To answer these research questions, the argument of this thesis will be provided along the following outline: First, two preparatory chapters introduce the background knowledge necessary for understanding the case. A review of CMR theory and an overview of the historical and political basics of the KR-I are put forward. In this context, intellectual backing for the stated hypotheses is provided. The subsequent chapters focus on answering the research questions. In Methodology, an account is given of how I set out to identify the system of the region during my field work. Findings and Analysis shows which aspects were abstracted from the data and how these findings support the asserted shortcomings of current CMR theory. At last, I identify concepts and manifest how their combination aids the understanding of the case and the revision of theory, in Revising Theory.! The first chapter Civil-Military Relations Theory provides a literature review structured along three debates. Starting from the classics, the evolution of the discourse is analyzed along the gap debate. The three phases are identified according to relevant shifts in focus: One, when attention moves from a US-based western analysis towards non-democratic settings, and two, when the investigation returns to the drawing board to question seemingly set assumptions on statehood, gap, and definitions of war and conflict. ! In order to fully comprehend the importance of civil-military relations and the gap debate in CMR theory, however, one needs to understand the case itself. Chapter Case Study is consequently dedicated to deconstructing the Kurdish region. A preliminary overview of history serves as an introduction into the political basis of the KR-I. Next, an analysis of the ongoing political and military division of the dominant parties KDP and PUK is provided. Relying particularly on Stansfield (2003) and Chapman (2011), I demonstrate the value of the Prisoner’s Dilemma model and its resulting effects on the study of Kurdish civil-military relations.! Following this first assumption, in Methodology, I introduce the research design and its execution. To find out which aspects current CMR theory is not able to account for, qualitative interviews were "4 conducted with research subjects according to a targeted convenience sampling and the snowball approach during the field work from August until December 2014. Issues of reliability, validity, and ethical considerations are also discussed in this Chapter. ! Subsequently, Findings and Analysis presents the results of the gathered data. Three main conclusions: One, a division between the parties continues to exist. Two, instead of there being a gap between civilian and the military spheres, a differentiation between a formal structure and an informal system is identified5. Three, the informal aspects are more relevant to the understanding of the examined relationship, because the institutions are weak in decision-making capacities. ! At last, in Revising Theory I identify additional concepts that help substitute the shortcomings of prevalent CMR theory. By combining Pion-Berlin’s (2010) identification of the relevance of informality in civil-military relations with Zukerman-Daly’s (2012) assertion of the relevance the geographic distribution of social networks has on the armed forces, a new perspective is gained on the findings. Furthermore, I show how the findings support the application of the Prisoner’s Dilemma model and how their analysis makes possible a first sketch of the necessary revision of CMR theory.! ! ! ! ! Civil-Military-Relations Theory: Three Debates! ! The field of civil-military relations sits in a multi-disciplinary intersection of several fields of study that are too wide to be included in this thesis. Therefore, following Feaver’s (1999) distinction, this thesis focuses on addressing the political science related aspects of CMR theory, as it best serves the design of this study: The elite- and system- based analysis correlates with the political scientists desire to identify structures and to “focus primarily on institutions and political control” (Feaver 1999: 212). The review of political science related CMR theory is thereby structured around the “gap” debate; a discourse along the question of whether there is a distinguishable difference between a civilian and a military sphere (Cohn 1999). Due to its continuous discussion, the gap debate allows to visualize relevant changes in CMR theory. Furthermore, the findings of this research suggest a necessary reconceptualization of this gap, which additionally supports its prominent status in the literature review. Three phases are abstracted from the wider debate based on the most important changes in focus6: A first part discusses the “classical” exchange between proponents of institutionalization- (Huntington) and convergence- (Janowitz) theory. The deviation from the US-american case towards analyzing “non- western”-civil-military relations provides a second category. And the last relevant shift occurs when scholarship starts to question seemingly set assumptions on statehood, gap, and definitions of war 5 For the definition of "structure" and "system" see footnote 2. 6 The “phases” are not be construed as part of linear progression where the later ones are more modern or refined. Instead, they refer to accentuation, normative tone, and the use of set ideas and concepts for the analysis of CMR. "5 and conflict. This third phase is characterized by the idea of lines between different conceptual spheres to be more blurry than priorly assumed. Since the presented research is situated within this last debate, its discussion will be the most extensive one in the literature review.! ! Institutionalization vs. Convergence ! Debates around civil-military relations are primarily concerned with questions on the connection, hierarchy, and influence between a civilian government and the armed forces of a state. A particular interest is the question how civilians keep control over the forces (Burke 2002). This question, however, often comes along with a fixed set of basic assumptions. Particularly at the beginning of modern civil-military relations theory when Huntington’s The Soldier and the State kickstarts a discussion on the manner in which the armed forces should be related to the state they serve. Finer (2006) puts the underlying conundrum of the field in simple terms: In order to survive, the state needs protection which is provided by armed forces. By the nature of their task, however, the forces are well organized, maintain a high symbolic status, and own a monopoly of arms. Because of these advantages the military has over political civilians, one needs to wonder “not why [the military] rebels against its civilian masters, but why it ever obeys them” (Finer 2006: 6). Answering this question, Huntington (1985) develops the theory of institutionalization in which he proclaims that the best way to make sure that a) the military is effective and b) it is does not interfere in the civilian state is to keep the military strictly professional, the civilian side presiding, and both spheres as far apart as possible. ! Morris Janowitz (1960), reacting to Huntington’s assertions, establishes the theory of convergence in which he aims to answer the same conundrum and uses the same basic assumptions but arrives at the opposite conclusion. His theory states that the more the military is in touch with civilian life, the less likely outright interference of the one side into the other becomes as there will be no chance for an ideological difference to develop. ! While these two “classical” authors disagree on a lot of terms7, what they have in common is 1) the belief in a clearcut division between a civilian and a military sphere, 2) the assumption of statehood with the resulting disregard for the sub-state level, 3) the predominant focus on the United States (Huntington 1985, Janowitz 1960, Cohen 1997, Collins and Holsti 1999, Kohn 1999), and 4) the normative postulation of the primacy of civilian control8. Thereby, from the very beginning “the gap” 7 They differ in three ways: 1) Huntington employs an institutional analysis while Janowitz focuses on the sociological relationship between society and the military (Feaver 1999). 2) They differ in their prescription of the ideal size of “the gap”. And 3) they follow different theories of democracy - liberal democracy theory (Huntington) v. civic-republican model (Janowitz) (Burk 2002). 8 Burk explains this normative preference through the ideas of the democratic value that “those with authority ought to be the elected representatives of the people” (2002: 8). Feaver even goes as far as to states “civilians are morally and politically competent to make the decisions even if they do not possess the relevant technical competence in the form of expertise [because] the expert is not in a position to determine the value that the people attach to different issue outcomes [… and therefore] civilians should get what they ask for, even if it is […] wrong” (1999: 215-16). "6 was a central point of discourse. In this first phase, the nature and size of the gap is debated. The differentiation between the spheres derives from Huntington’s assumption that different institutions will lead to different values, which will lead to differing policy preferences and identities, thereby making the spheres distinct. The importance of the gap, according to Huntington, remains with the ultimate goal of assuring civilian superiority and preventing military intervention in domestic affairs (Luckham 1971, Albright 1980, Cohen 1997, Feaver 1999, Burk 2002, Schiff 2004, Michael 2007, Chuter 2009, Angström 2013).! ! “Non-Western” Civil-Military-Relations Theory! This military intervention in domestic affairs and the question of how to avoid it becomes particularly prominent during the time of coups and counter-coups in Latin America. Finer (2006) in The Man on Horseback identifies the different ways and intensities in which the armed forces can influence the political sphere, with an ultimate climax being the overthrow of a civilian government. Davis (2003) differentiates three generations of scholars who fit this second “phase”. The first apply modernization theory to lead former colonies towards military professionalization9, while the second generation sees a split between those that applaud and those that condemn the military intervention in Latin America, and the third try to explain why military regimes also collapsed after a while into their rule. No matter the normative point of departure of the scholars Davis differentiates, the more important issue to notice here is the shift in scholarly attention away from the primary case study of the United States and democracies towards analyzing “non-western” cases. In this process, limitations of existing CMR theory become apparent. While the analysis of the reasons, causes, and effects of military intervention in Latin America serves to support the normative postulation of civilian control being better than military control of a state, the reflection on communist systems provides the insight that a high politicization of militaries does not necessarily end in a coup either (Albright 1980, Perlmutter and LeoGrande 1982). Perlmutter and LeoGrande identify fusion, symbiosis, and coalition as the working modes of CMR in communist states, while Albright clearly states that in communist systems there is “no distinction between civilian and military sphere” and instead finds a system based on personal loyalty structures (Albright 1980: 558-9).! While the debates on both Latin American and Soviet/Post-Soviet cases are much more extensive than presented, the most relevant insight of this phase for the research is the highlighted incompatibility of applying western-standardized approaches onto non-western cases. Details of the different studies are considered to not be further relevant as this phase still maintains the 9According to Feaver (1999) the most important authors in this field are Finer 1962, Huntington 1968, Stepan 1971 and 1988, Welch 1976, Nordlinger 1977, and Perlmutter 1977. "7 underlying presumptions of 1) the assumption of statehood, 2) the division between a civilian and a military sphere, and 3) the norm of civilian control10.! ! “New Wars” ! The last phase of discourse is represented by Kaldor’s (1999) catchphrase “new wars”. While not directly related to CMR, her concept fits the idea this segment represents. In this debate, a questioning of seemingly set concepts, such as statehood, the gap, and the definite line between peace and conflict takes place. ! The central question in a “new war” understanding of CMR theory reads as follows: If a variation in warfare can be traced and existing assumptions about dichotomous concepts of theory are reedited, what consequences do these changed circumstances have on civil-military relations and their analysis? So far, only a limited amount of approaches, to the author’s knowledge, try to break out of the state perspective, the clearcut distinction of spheres, and the normative postulation of civilian- being categorically better than military- control11. In fact, I was only able to find approximations to either one of the presented revisions of concepts, but no proposal that redefined all of the above. The study of the KR-I instead challenges the gap and the state-perspective, by applying a meso-level analysis of the northern region of Iraq and viewing the context of party- territories in a stage of quasi-anarchy. The focus in this “new wars” phase of civil-military relations discourse lies on the reconceptualization of the gap. It is characterized by an investigation of the nature of the gap and an increased differentiation between different types of relationships.! Most known for a first theoretical revision of the assumption of a necessarily given gap is Schiff’s (1995) theory of concordance in which the author declares that three actors -the civilian elite, the military, and the society- need to be d’accord on four variables: the composition of the officer corps, the political decision-making process, the recruitment method, and the military style. If concordance on these issues is given by the actors, stable civil-military relations are achieved (Schiff 1995). Thereby, Schiff does not postulate which CMR work best or ought to prevail (Schiff 1996), but instead she uses her theory as an analytical instrument to identify which system exists: Sometimes this system is one of institutional separation and at other times no distinction is possible. ! With this first revision of the presumed gap, an entirely new angle of the debate becomes visible. In the term of “fusion”, and linking particularly to Soviet examples, arguments for the nonexistence of 10 Although this norm is reviewed more extensively in the study of Soviet-cases. 11The idea that a civilian control over the military is superior to the other way around presents the most prominent normative assumption of CMR theory. These normative expectations often influence policy decisions for states, transitional governments, and Security Sector Reforms (SSR). Chuter (2009) warns: “A theory which describes reality incorrectly or incompletely is unfortunate, but a doctrine which prescribes [sic.] action on the basis of a fallacious theory is positively dangerous” (Chuter 2009: 1). "8 a gap are made. Feaver (1999), however, raises a distinct criticism towards the assumption that the gap could ever be understood in a non-dichotomous manner:! ! “The spheres are necessarily analytically distinct - a distinction that derives from democratic ! theory and the agency inherent in political community […] The spheres are also necessarily ! distinct in practice -it matters whether the policy maker wears a uniform or not- and so ! ! fusionist scholars find that their subjects repeatedly revive the idea of difference even as ! ! they provide evidence of overlaps with the activities of actors from different spheres.” ! ! (1999: 220)! While I support Feaver’s argument that a conceptual distinction makes a conversation about different aspects of theories and case studies easier, I also agree with Schiff (2004) and Chuter (2009) who point out that just because this clear institutional and cultural distinction holds for the United States, one can not conclude on this separation of spheres to work universally12 and simply presupposing it can lead to unwanted results (Chuter 2009). For this thesis, therefore, the distinction between “a civilian” and “a military” aspect is upheld in a terminological, conceptual aspect to ease the understanding of the argument, while it has to be dismissed for the final analysis as the “gap” between clear sphere is anything from institutionally marginal to culturally absent. To further support this argument, the classical conceptualization of the gap between institutional and cultural forms (Huntington 1985) is altered by more recent reviews of the different forms of the relationship between the civilian and the military sphere. A particular highlight is placed on the concepts presented by Rahbek-Clemmensen et.al. (2012) and Angström (2013).! As mentioned above, the gap and its conceptualization exist along a range of intensities between the poles of “fusion” (e.g. Perlmutter and LeoGrande 1982, Feaver 1999) and “separation” (e.g. Huntington 1985). Fusion refers to the impossibility of differentiation between spheres13. while separation refers to Huntington’s (1985) idea of different institutions leading to different values, cultures and resulting policy-preferences and identities.! Relevant to understand these poles as well as the many variations in between is a recent study by Rahbek-Clemmensen et.al. (2012) that identifies four different forms of the gap: cultural, demographic, institutional, and differences in policy-preferences. By connecting their institutional, cultural and policy-preference type of the gap, one ends up in the initial analysis of the nature and causes for the gap as discussed in Huntington (1985). Rahbek-Clemmensen et.al., however, enlarge this institutional difference, by identifying the media, courts, and the education system as a medium of contact and relationship between the spheres (Rahbek-Clemmensen et.al. 2012: 673). 12After all, the dichotomy does not even hold for the US itself as Schiff shows in her analysis of a post- revolutionary America (Schiff 2004). 13As reviewed in Feaver’s (1999) critique above, fusion is almost impossible to conceptualize as the idea of there only being “one” does not serve an analysis of a relationship that by definition, nature, and name presupposes there to be two. However, this dogmatic assumption of duality also leads to misconceptions about actual civil-military relations in certain case studies (Chuter 2009); the KR-I being one example. "9 In addition, they identify a new form of gap in the variable “demography”. This variable addresses the question whether the demographic set up of the military represents the according population. While their study focuses on the US example, the demographic set up of military forces is seen to be a central variable in the literature and praxis of merging military forces (Licklider 2008, Licklider 2014). Particularly after civil wars, building demographically representative armed forces seems to provide a first trust-building measure for the reintegration of society. In the KR-I, a similar pattern can be observed. As shown in the findings, all Unified Peshmerga Forces (UPF) are 50-50 divided between the parties KDP and PUK. Therefore, Rahbeck-Clemmensen et.al.’s differentiation of four different “kinds”, natures, and reasons of gap, comes in handy for explaining the case study. However, I would like to add a fifth variable: the gap of legitimacy. Following the underlying normative tone of CMR analyses and SSR prescriptions (Chuter 2009), most academics assume there to be a gap between the two spheres due to the democratic value of legitimacy remaining with those who are elected - the civilian sphere. This assumption of legitimacy on one part and not the other also creates a distinction and gap between the spheres that ascends the conceptual differentiation between the two domains. However, the attribution of legitimacy to one side only does not always hold. Finer (2006) already asserts a symbolic value to the armed forces which hints towards a degree of perceived legitimacy. And while most studies assert only a temporary increase of legitimacy for the military14, the literature on the army as the “school of the nation” and some prominent examples like Egypt or Turkey might force to revisit the dismissal of none or short- term military legitimacy entirely. In either case, the variable “legitimacy” is a relevant addition to the conceptualization of the gap, but does not yield wide consequences for the Kurdish case, as there is no visible differentiation between “the military” and “the civilian” present.! Applying these five kinds of gaps on the Kurdish case, the research shows that there is a marginal gap when analyzing the institutions (structure), but there is no distinction between the spheres possible when addressing the other four kinds of the gap (system).! While Rahbeck-Clemmensen et.al. attend to the kinds of the gaps, another recent approach categorizes the different forms of the relationship. Ångström (2013) develops five ideal-type models of civil-military relation: a civilian dominance, a military dominance, parity between the two, intertwining, and dissolution (Ångström 2013: 225). This differentiation in itself is not relevant to the presented research, but the underlying idea Ångström presents in his work is instructive. While both researchers and policymakers have a tendency to treat the dichotomous categories of civilian and military as a technical matter of differing degrees of conflict and cooperation, Ångström identifies the distinction to be based on political norms of how society should be organized:! ! “Rather than understanding civil and military as fixed categories, I suggest that they should ! be understood as norms. Norms, following Peter Katzenstein, include both regulative and 14 For example Finer (2006) sees this as one of the opportunity that can incentivize military interference in politics, while noting that the legitimacy is only temporary, as even military leaders have to change into civilian clothes eventually. "10 ! constitutive elements[:] Norms ‘describe collective expectations for the proper behavior of ! actors with a given identity’. Critically, the norms of civil and military are inherently linked to ! political order and hierarchy [… and] both civil and military tend to be heterogeneous ! ! categories.” (Ångström 2013: 226)! According to Ångström, the most relevant variable for comprehending the difference lies in the state’s interest to differentiate between different acts of killing: murder and the “necessary” violence used by armed forces15. The later, therefore, is a legitimized form of violence that is bound to the norm of civilian guidance. Following this understanding of normative ideas building the basis for the categorization, definition, and understanding of the distinction between the civilian sphere, a new array of questions appears: If a civilian wears a uniform, is he/she military? If not, what constitutes the difference? Is the difference visible or not? As Ångström shows in the case of Afghanistan, these questions are relevant when occurrences of “seasonal soldiering” are observed. For the KR-I a similar relevance can be asserted: Peshmerga on the front lines wear Kurdish traditional clothes as well as different, largely unrelated prints of uniforms. Sometimes civilians join for a day of fighting or two in either their civilian clothing or in military print. While this point will be further discussed in the section Findings, it can be noted that there is no continuous visual distinction between civilian and military spheres in the Kurdish case. ! Another aspect able to be addressed due to Ångström’s argument is captured in the following scenario: Consider a person who concludes a military career, retires from the military, then enters the political arena and takes over the Ministry of Defense. Is this person “civilian” or “military”? Depending on which norm prevails in the according society (Ångström), which concord has been reached between the dominant actors (Schiff), and which perspective of the five identified forms gaps is taken (Rahbeck-Clemmensen), the answer to this question will be different. It is this understanding, that “what it means to be civilian and military is a political choice […] that is inherently linked to a particular form of international and domestic political order” (Ångström 2013: 227), that alters our understanding of civil-military relations and the approach to the presented case. ! ! The Case Study ! ! Having discussed the literature, a next step introduces the Kurdish region in order to understand how this case study highlights deficiencies in CMR theory. I will first present a historical overview over the region before sketching the existing literature on the political and military aspects of the Kurdish Region. ! 15“If we did not have ‘civil’ and ‘military’, the elites would not be able to uphold the distinction between war and peace.” (Ångström 2013: 227) He further connects this separation of “public-private” as well as “civil- military” with a norm inherent to the idea of the Westphalian state; also see Wimmer (2003). "11 ! A Contextual Analysis - the History of Kurds in Iraq !! ! ! The Kurdish region in Iraq as it stands today is a product of Britain's geo-strategic interests on the trading route to India. Having secured the region around Basra and Baghdad, the area surrounding Mosul and covering the Kurdish populated areas in the northern mountains was included in the entity called Iraq due to three factors: 1) a political consideration - including the Ottoman-district “Mosul” and its Arab and Kurdish Sunnis helped to keep a Sunni domination over the Shia south, 2) a military strategy - Iraq was easier defended through the Kurdish mountains, and 3) an economic interest - the oil surrounding the area of Kirkuk (Anderson, Stansfield 2004: 162-3).! Once part of Iraq, the Kurdish population fought for national self-determination in cooperation with other Kurdish groups in Turkey, Iran, and Syria. To maintain security in Iraq, the British forces tried to co-opt the Kurds by working closely with tribal chiefs (ibid.). With this propping up of tribal forces on the one hand and within the greater developments of a spread of marxist ideologies on the other hand, an already existing division was reenforced. With the conservative tribes led by the family of Mustafa Barzani, and the urban, left-wing intellectuals led by Ibrahim Ahmed and Jalal Talabani being separated by ideology, it was the necessity to resist the center in Baghdad that forged the two into an “uneasy alliance” (ibid.: 164), called Kurdistan Democratic Party (KDP), in 1946. ! Skipping ahead in time, the next relevant point is posed by the ousting of the monarchy in Iraq in 1958 by Abdel Karim Qassim that concluded, after a series of military turn-overs, in the last coup led by the Baath Party in 1968. During this time, the two wings united in the KDP regularly changed their position of relative dominance within the party. By the time of the last Baath Coup in 1968, Mustafa Barzani had established himself as the main force. Concluding the March Agreement in 1970, that established certain rights and recognition for the Kurdish people (ibid.), a first peace was found in the region. However, all achievements were destroyed 1975 by an Iran-Iraq alliance in the Algiers Agreement that ended Iran’s support to the Kurdish resistance. The Kurdish revolution was at an end and finalized by Mustafa Barzani fleeing to Iran. The political vacuum left was filled with an ever stronger division in the KDP. In 1975, the left wing under Ahmed and Talabani split and created the Patriotic Union of Kurdistan (PUK). Since then, intra- and inter- party strives continued parallel to both sides fighting the regime, participating in the Iran-Iraq war, and living through the Anfal-campaign and Halabja in 1988. ! During the Gulf War of 1991, with the Kurdish uprising on March 4 being retaliated against so brutally on March 28 by the central government, the United Nations Security Council Resolution (UNSCR) 688 demanded an end of the repression and established a no-fly-zone above the 36th parallel. This safe-haven provided a first chance of self-rule for the Kurdish population (ibid.: 171). ! In the political vacuum of quasi-anarchic circumstances, the parties KDP and PUK stepped in to provide for security and basic services. Thereby, a basic cooperation between the parties was "12 maintained and a common, democratic government structure was constructed. The first elections commenced on May 19, 1992, forming the Kurdistan National Assembly (in 2009 renamed to parliament). In a 50-50 power sharing agreement, the two parties (a threshold of 7% inhibited other parties from joining the government) split the control of the area. But neighboring and international powers exploited the instability between the parties16. At the same time the parties did little to lower their mutual antipathy and competition for resources, influence, and prestige (ibid.: 173). Instead of sharing, mutually exclusive areas of influences were created17. ! Starting in 1992 already, with political violence erupting (Chapman 2011), a civil war broke out in 1994, which saw the PUK enforcing its hold on Erbil and Sulaymaniyah and the KDP retaining control over Duhok and Germian. During the conflict, the PUK aligned with Iran, while the KDP cooperated with Baghdad on the mutual interest of the oil smuggling route to Turkey and invaded Erbil on August 31, 1996. The PUK Peshmerga retreated to Sulaymaniyah and the eventual cease- fire line created from Koysinjaq to Haji Omran became the dividing line of the political map.! From the cease-fire in 1996 until 2006, the Kurdish north was separated into two parallel structures, with both building up a party centered administrative system. These zones where so isolated from another, one can speak of two quasi-states in the Kurdish region (McDowall 2007: 386) with the KDP settling in Duhok and Erbil, and the PUK remaining dominant in Sulaymaniyah and Kirkuk. ! A normalization of the relationship between the divided parties and areas was reached in a step by step process through the Washington Agreement on September 1998, a renewal of this agreement in October 1999, a meeting of Masoud Barzani and Jalal Talabani in Frankfurt in 2002, and the 2003 US invasion of Iraq. In this invasion, the chance to gain rights and liberties in an Iraqi framework brought together the parties to speak in one voice for the Kurdish cause and population. Since then, the Iraqi Federal Constitution 2005/06 granted the KRG the status of a federal unit with relative autonomy in the governorates Duhok, Erbil, and Sulaymaniyah. Since, the KRG held three elections, one popular vote for the Kurdish presidency, and the political arena has further diversified. In the latest elections on September 21, 2013, the KDP gained 38 seats, PUK 18, the Gorran Movement 24, the Islamic Union (Yekgrtu) 10, the Islamic Brotherhood 6, the Islamic Movement and two Communist parties won one seat each, and 11 seats are allocated to the minorities by quota. In this election, the second strongest party is Gorran. While not having been addressed in the historic review so far, this party is the third civilian actor identified to be valuable to the research. Its importance derives from their strength in government, as well as from their current position of holding the Ministry of Peshmerga (MoP). Therefore, the party shall be further 16The most prominent example is a meeting between Iraq, Iran, Turkey and Syria in 1994, discussing “the de facto Kurdish entity and how to ensure its demise” (Anderson, Stansfield 2004: 174) 17 By that time the ideological difference had been reduced to a minimum. Instead, it was the goal of undisputed leadership over the territory and personal rivalries that drove the competition (Anderson, Stansfield 2004) "13 introduced: Gorran is a 2009 break away from the PUK. Founded and led by Nawshirwan Mustafa, it constituted the first opposition party after the 2009 elections by refusing to take its 25 seats into the government. In 2013 the party attained 24 seats and joined the government in the Ministry of Peshmerga (MoP), Ministry of Finance, and the Speaker of Parliament. ! Today, the Kurdish Regional Government maintains extensive legal, executive, and judicial liberties granted by the Iraqi constitution. This high level of autonomy has been constantly expanded, both in a bargaining process with Baghdad (Natali 2007, 2012) and a conflicting interpretation of the constitution (Holland 2012), up to a level where the region is considered to constitute a de-facto state (Anderson and Stansfield 2004, Chapman 2011, Gruber 2013). Internally, however, the KR-I remains divided even today. While the US-led invasion 2003 and the thereby triggered changes for the Kurdish region increased the effort by the leading parties to merge and institutionalize, the actual structures of the Peshmerga remain divided (Stansfield 2003, Bengio 2009, Chapman 2011). In order to access this disunity, an account will be given to the political landscape of the region and the nature of the division.! ! The Political Division - Two Separate Administrations! This overview of the literature existing on the KR-I limits itself to issues related to the current political and military context. Several scholars have picked up on different aspect already, but the interrelation of the aspects is largely unexplored. Stansfield focuses on the Kurdish region and its role within Iraq (2013) as well as on the political situation within (2003). Others take note of how changes in the north of Iraq are effecting the relationship to surrounding neighbors (Olson 2005, Bengio 2009). Denise Natali (2012) analyzes economic factors and questions of statehood in the bargaining process with Baghdad (2007). And while some focus on the question of Kurdish nationalism and nation-building (Bengio 1999, Ahmed 2012), Kurdish identity (Aziz 2011, Van Bruinessen 2004), every day life (King 2013), natural resources (Zedalis 2013) and other issues, the role of the armed forces remains, at best, a side note of history. To the author’s knowledge, it is only the work of Chapman (2011)18 that extensively covers the Kurdish security sector. However, Chapman, provides a solely descriptive perspective. Therefore, while his presentation serves as a good point of departure, there is a lack of connection between the parties and their military forces. It is this gap that my proposed research aims to fill. Combining Chapman’s military understanding with Stansfield’s analysis of the nature of politics in the KR-I, I aim to show the relevance of this connection and how it serves as a precondition for conducting this research.! ! In his 2003 analysis of the political situation in the Kurdish region of Iraq, Stansfield provides several insights, including the existence of two different administrative systems split along 18Lortz (2005) analyzes the Peshmerga in a historical perspective until 2003. However, the more important period post-2003 is examined by Chapman. "14 territorially distinguishable lines. He also notes the identifications of the parties with the institutions government. More importantly, however, are the consequences deriving from such a situation. Having been separated in all but nationality, the Kurdish system of government is at times more comparable to two one-party-states trying to merge into one rather than two parties unifying their interests in cooperation and across territory (McDowall 2007: 386). Therefore, Stansfield’s third conclusion that the party-based rivalries need to be managed with a process of elite accommodation in a consociational- type of administrative system can only be seen to reflect half of the solution. What he fails to point out, by focusing on the issue as if it was solely about accommodating two parties in one territory, is the relevance that derives from the historical, territorial, and administrative division. By additionally viewing the case from a perspective of two separate units merging into one under an umbrella of relative anarchy, continuing mistrust and rivalry stemming from the inherent insecurity of the transformation process can be accounted for. As explained above, anarchy refers to a setting where an overarching authority is missing. A situation which is most often assumed for the international sphere, anarchy can be employed for state-internal environments if the central government is not effectively able to exert sovereignty in the form of legislation and execution over its territories. Sarah Zukerman-Daly (2012) uses the post-civil war environment as one example to support this assumption19. For the KR-I, a similar argument holds as the region still resides within a landscape of post-civil war, and administrative- and territorial-division, with the central Iraqi government having very little influence over the region. Therefore, having established that there is an aspect of quasi-anarchy under which the Kurdish parties operate (H1a), Stansfield’s idea of elite accommodation has to be adapted accordingly. ! The particular aspect deriving from a situation of quasi-anarchy, is the factor of insecurity that comes with it. Without a functioning, superior level of security provision, vigilantism and group formation creates an alternate structure of safety and stability20. In the Kurdish region the parties provide this regulatory function for society. In addition, they also distribute economic benefits, lands and other services to their clients (see Findings). Parallel to this party influence, there exists cooperation between the dominating parties in the form of the Kurdish Regional Government. Due to the surrounding threat potential arising from all sides around the region, the Kurdish parties are pressured to accommodate their interests and cooperate despite their security concerns. However, that pressure alone does not yet negate the existence of diverging interests, claims to power, and the potential intent to shift the existing balance in ones favor. It is particularly in the security sector that one can observe this party interest in maintaining a vehicle for personal security, insurance of influence21, and future bargain. Therefore, I argue that Stansfield’s application of the consociational 19“The post- civil war landscape resembles the international system [as] there exists no overarching government able to enforce agreements” (Zukerman-Daly 2012: 2). 20 “You just have to have a weapon. If you do not defend yourself, no one will” (OTR). 21“For much of their history the KDP and PUK have relied upon armed force as a principle means of pursuing their political aims.” (Chapman 2011: 4) "15 model has to be altered by the model of Prisoner’s Dilemma to account for the continuing insecurity that arises between the parties when loosening their control over their own forces by unifying them under a governmental institution (H1b). ! Reviewing a basis of literature on the concept in Axelrod (1980), Andreoni and Miller (1993), Nowak and Sigmund (1993), Ellison (1994), a definition confirms how the model fits to the research case: A Prisoner’s Dilemma refers to a scenario where two actors are involved in a game with each other, without ever being able to react to the other player as they are either unaware of the other’s choice or have to play simultaneously. Three ideal scenarios appear: Either both cooperate -which ends up relatively good for both, both defect -which turns out very badly for both, or one defects while the other cooperates -which ends up very good for the one that defects and very badly for the one who cooperates. The crux in the game is the following:! ! “The distinguishing feature of the Prisoner’s Dilemma is that in the short run, neither side ! can benefit itself with a selfish choice enough to make up for the harm done to it from a ! ! selfish choice by the other. Thus, if both cooperate, both do fairly well. But if one defects ! ! while the other cooperates, the defecting side gets its highest payoff, and the cooperating ! side is the sucker and gets its lowest payoff. This gives both sides and incentive to defect. ! The catch is that if both sides defect, both do poorly.” (Axelrod 1980: 4)! This Prisoner’s Dilemma is considered to be vital for the case because 1) it incorporates the inherent insecurity of all sides, and 2) it minimizes the complex situation to a simple game scenario. Both the KDP and the PUK are involved in a game of trust and dilemma when approaching the question of placing the party-forces under one central command. After all, if either one of the two parties controls the unified forces, a potentially negative consequence in political and military security for the other party, its territorial stronghold, and its clients may arise. As with the Prisoner’s Dilemma, the two parties are stuck in the situation of being forced to play (unify the forces) infinitely22, while their individual options (handing over control or remaining in control behind the scenes) are interrelated with those of the other party. Therefore, the question is not just how to unify the military forces, but also how to remain a degree of security for the parties. Particularly with the history of having unified the forces already in 1992 and them breaking apart along the party-polarization during the civil war, it is recognized that a structural unification alone is not enough to decrease the party-based influence. While these questions are vital to be asked for the Kurdish government, answering them is not subject of this thesis. Instead, I will now demonstrate how the situation of Prisoner’s Dilemma effects the ability to comprehend the complex, dynamic, and multi-layered civil-military relations of the Kurdish region.! One of the main findings of this research is the differentiation between “structure” and “system”. Further demonstrated in Findings and Analysis, the qualitative assessment of the civil-military 22 The idea is that KDP and PUK do not play the Prisoner’s Dilemma once or twice, but infinitely, with the consequence that each decision is seen as “one” game in a sequence, in which players will gain “learning”, “forgiveness”, and “(mis)trust” (Axelrod 1980). "16 relations of the KR-I leads to conclude that next to the formal structure of official institutions and lines of command, there exist informal channels of influence that create parallel and sometimes intersecting relations between the actors.! This central finding of the research connects to the picture gained of the Kurdish region and its sought after transformation, when applying the model of Prisoner’s Dilemma. By accepting the basis of quasi-anarchy and the resulting level of insecurity, the model helps explain as to why there is an informal system of influence maintained parallel to the increasing institutionalization of the structure: The influence exerted through the system provides a certain security guarantee to the parties aside the official channels of government. The core of civil-military relations therefore is not an issue of balancing the mutual influence between the civilian and the military sphere, but between the system and the structure, both of which maintain levels of civilian and military aspects within them (H2a). In addition, the concept of Prisoner’s Dilemma allows to predict that this level of system influence is likely to be maintained during the process of Stansfield’s elite accommodation and therefore has to be accounted for in future research.! With this first demonstration of how the existing body of CMR theory has to be re-conceptualized in order to comprehend the Kurdish case, I also want to point towards other reasons why classical CMR theory does not work for the KR-I. ! For one, prevalent theory starts from a state perspective and sets out to analyze the relationship between one civilian and one military actor. The state perspective provides the preliminary assumption of one center of monopoly of legitimate violence and the resulting gap of legitimacy. In the Kurdish region, however there is neither a clear distinction between the spheres, nor is there, even if we assume a conceptual distinction, one actor on each side of the relationship. Instead there are three: The two dominant parties KDP and PUK maintain their own loyal forces unit 80 (KDP) and unit 70 (PUK), and in addition there are Unified Peshmerga Forces (UPF) under the Ministry of Peshmerga. This existence of several civil-military relations next to each other already defies the prevalent body of theory.! More than that, the internal relations of these six actors also do not fit into the idea of clearcut distinction between spheres. Stemming from a guerrilla movement that was later institutionalized under a strict party (and territorial) division, the Peshmerga force even today largely resembles more the idea of “nation/party in arms”. Classical “distinctive” insignias, for example, such as the wearing of a uniform or the carrying of a weapon are as common in the streets of kurdish villages as they are on the front line to ISIS. Another example is that most soldiers also work as taxi drivers, and generals can also be business men and politicians, just as well as politicians can sometimes call the shots in military combat. Further examples will be provided in the Findings section. In conclusion, it becomes obvious that the existing body of CMR theory lacks explanatory power for the Kurdish case (H2b). This research therefore aims to provide new perspectives of CMR to fill "17 the void where the current theory lacks a deeper understanding: in the case of hardly- institutionalized, party-based-politicized, and transforming military establishments. ! Following this analysis of the case study and the question of why CMR theory does not work to explain the case, the question of what I did to figure out the actual system of civil-military relations arises. In the next section, I will therefore present the design of the study, the method employed, and a critical revision to the reliability, validity and biases of the research. ! ! Methodology! ! As Feaver (1999) defined, the key task of an empirical-descriptive analysis is “distinguishing between reality and rhetoric, between what appears to be the case and what is in fact the case” (Feaver 1999: 216-17). Adhering to this philosophy, the presented research sets out to identify what the civil-military relations of the Kurdish region actually are. Therefore, first, I identify a way to access the relationship by determining its most relevant actors. Second, I employ qualitative interviews, which allow for an understanding of the meaning held by research subjects (cf. Arjimand 2013). Third, by using the snowball principle, additional relevant subjects are identified. And fourth, interviews are transcribed, analyzed by hand, and the examples are used to find which aspects of the relationship can not be explained by the existing body of theory. Only after concluding from the findings, I set out to find additional concepts that can help bridge these shortcomings.! ! Design of the Study! Before presenting the method, the terminology used to differentiate between the many layers in the design have to be outlined.! I. Actor: The term refers to the organizations identified to be relevant for the research. The civilian actors are the parties KDP, PUK, and the MoP (Gorran). The military actors are the Peshmerga Divisions 70, 80, and the UPF.! II. Subject: A subject is the individual person within an actor that is determined to be valuable for the research. ! III. Distant Subject: A distant subject refers to those people surrounding the subject, such as press secretaries, prior commanders, relatives, and friends. Distant subjects are still considered valuable research subjects.! IV. Research Subject: Contrary to a subject, a research subject refers to the person that has been interviewed for the research. Research subjects can be subjects or distant subjects.! V. Value for the research: The value for the research is identified by the actors’ and subjects’ level of influence in the civil-military relations.! ! "18 The first step in the field work identified relevant actors for my research. While trying to avoid a presumption of “a gap”, the terminology of civil-military relations leads to conclude that in order to access the relationship that has to be analyzed, a tackling from both sides, the civilian and the military, is necessary. Actors were determined with the help of Dr. Burhan Yassin and Dennis Chapman (2011). The relevant civilian actors are determined to be the dominant parties, KDP, PUK, and Gorran, and the Ministry of Peshmerga23. Since Gorran currently holds the MoP and has no party-owned military forces, the research considers three civilian actors however. The political influence (see Case Study) and military relevance for the KRG (see Chapman 2011) of each party guided the selection process. As these political actors were chosen for their military relevance, the military pendants were identified accordingly: the KDP unit 80, the PUK unit 70, and the MoP in its declared strength of 14 Unified Peshmerga Forces (UPF).! Within these identified actors, it is necessary to identify subjects who are relevant to access the structures, influences, and proposed changes for the future. For this research an elite-focus is proposed. This emphasis derives from a combination of two factors: Stansfield’s (2003) assertion of the relevance of the elite in the region, and the hierarchical nature of the observed actors. Therefore, within the political actors, subjects are sought in the positions of politburo-members, and for the military actors the officer corps are targeted. While it is a valuable objection that there might be people outside the identified actors that are influential, most of the time, the powerful personalities (and therefore interesting subjects) are found to be within the according position of the party/military24. Within each actor there is a different number of subjects that are suitable. Their value for the research is determined by their level of influence in the system, which in term is accessed through in-depth conversations with other people of influence (snowball principle) and the analyses of local academics, journalists, and other actors in the region.! Once the most relevant subjects have been identified, the hardest part of the field work is to gain access to the according personalities. Given the aim of this research to access the highest political and military level of the region, the chances of getting appointments was relatively low. Therefore, in order to make the research project more realizable, it was defined that also people close to the identified personalities can be handled as valuable research subjects. These personalities are referred to as “distant subjects”. ! At this stage, in addition to the stringent target sampling of the selection process, an element of convenience sampling necessarily has to be added to the design of the study.! ! ! ! 23 Other parties are neglected due to their lower relevance in military and political terms. The Ministry of Interior and the Security Council of Kurdistan are disregarded because the focus of this research lies only on the unification of the Peshmerga. 24 Not to be confused with influential positions in the government. "19 Method and Analysis! Once access to research subjects was established, due to the goal of gaining in-depth understanding of the civil-military relations in the region, the method of qualitative interviews was applied. While reservation might be voiced that employing a qualitative analysis on the assessment of a structure comes with the risk of another interviewee providing an entirely new outlook on prior evaluations, in addition to it not providing any quantifiable data or clues towards the organigram of the relationship, there are several counter-arguments presented: For one, there already exists an analysis of the set-up of both the parties and their according military forces (see Chapman 2011). Second, the relationship between the actors is in flux as it is affected by the ongoing conflict and the increased unification process by the MoP. And third, within the Kurdish region there is no census which renders quantitative data just as much a speculation towards reality as qualitative research can achieve. In this context, the methods opted for are deemed to be appropriate, valid, and as reliable as possible. As for the first aspect, while I am not able to exclude the eventuality of someone providing me with an entirely different perspective on the analyzed relationship, I aimed at countering this chance by interviewing a maximum amount of research subject possible during the four months on the ground. In addition, this thesis does not claim to reflect anything else than the information obtained through the qualitative interviews with the research subjects during the field work. ! Qualitative interviews were conducted with 25 research subjects. “Categorizing” these 25 into “civilian” and “military” or into the six actors is almost impossible due to the infinite inter-linkages of positions. The prime example to this fact is Sheikh Jafar Mustafa. Currently the military leader of unit 70, he is also a politburo-member of the PUK, the military advisor to KDP leader Masoud Barzani, and the former Minister of Peshmerga. Therefore, what is instead relevant to understand is that the goal of the field work was to access as many subjects and distant subjects within and between the six actors as possible. ! The interviews themselves were in-depth with open-ended questions, non-standardized and unstructured in order to suit the nature of in-depth inquiry. At the same time, however, every interview was accompanied by a protocol (see Appendix), as it aids the focus of the research. There are two different protocols; one for the civilian and one for the military sphere. This differentiation of protocol does not arise from an inherent difference of the civilian and the military sphere but from the context the interview was conducted in. Due to the current war, most interviews with military officers had to be held at the front line. This unique opportunity enabled access to the civil-military relationship from an additional practical perspective. By asking for the lines of command and supply for example, I was able to access the differences between the formal structure and the unofficial system. The therefore resulting difference in the line of inquiry varied in such that political actors were addressed more firmly on the topics of goals, plans, ideas, political programme, and potential problems, while the focus rested more on the nature of the structure of "20 command, as well as incidences and the set up of the unions on the front with military personnel25. In general, both protocols were constituted by the main topics and were sometimes adapted whenever a specific case managed to highlight a certain aspect of questioning26. ! The unstructured interview was chosen as it allowed me to adapt to my research subject in a manner to create a conversational atmosphere. This sort of ambience was identified to aid the research as it proved to increase the trust and ease of my research subjects. At the same time, the open-ended nature of the questions allowed to avoid the pitfall of asking along predefined concepts of CMR understanding. Specific questions were only asked along the provided information of the speaker. In addition, the tactic of anonymously crosschecking already received information to every new research subject was applied to gain a multi-reflective perspective on the system I tried to access27.! In the interview, I always began with introducing myself and the research I am doing, why I am doing it, what I am doing it for, and what sort of topics I was interested in discussing. I also provided an explanation of why the particular research subject was chosen, informed the person that I would like to use his/her comments within the realms of my research, and offered the option of ending the questions or taking information off the record at any point of inconvenience. Once the basics were cleared, I always started with a very general question of “can you explain the situation”28. This open approach was chosen due to the fact that civil-military relations can not be directly asked for29 and it is easy to ask the wrong question: Sometimes, problems can derive out of a translation issue. I found, for example, that it is contra productive to employ a translator who is not aware of military terminology as this can lead to confusion. Sometimes the problem arises because the term itself is not familiar, and at other times the same term is used but understood in a different manner. I got around these issues by describing the term or asking indirectly. ! The interviews were recorded and transcribed. Due to the familiarity with the data, in the analysis, I was consequentially able to define broad categories beforehand. The impression sticking out the most was that of there being a certain “system” next to the official structure that seemed to be more relevant to explaining “what is the case” (Feaver 1999). Reoccurring statements along the line of “Officially…, but actually…” enforced this idea. Therefore, the category “System” was likely to become the core piece of this research. In addition, I chose categories along topics that helped highlight the relevance of studying this system. These topics are History, Structure, Goals, and Problems. I will now further address the analysis of the data.! 25 See Appendix Protocols. 26E.g. the elections in Sulaymaniyah, the budget discussions with Baghdad, the Shengal incident, the deployment to Kobane. 27 This is obviously a cumulative process, as the knowledge increases with the number of research subjects. 28 See Appendix Protocol Note 1. 29 This was confirmed by Gareth Stansfield in a personal Skype conversation on September 12th, 2014. "21 Since the division between KDP and PUK is one of the main characteristics of the political landscape of the region, an inquiry on the reason and nature of this division, despite the already existing knowledge, suited as a smooth entrance point towards more specific questions in the interviews. These questions, however, were answered by an analysis of history. Therefore the topic “History” developed. Analyzing the findings, I realized that while the basic dates concur, the emphases and connotations of history are distinctly different between the parties. Hence, while not repeating the dates of history again, some examples are presented to this “connotation” by differentiating the data between the sub-categories of “what is agreed upon” and “where there is disagreement”. The analysis then shows how this category proves there to be an ongoing party- division and tension between them.! While usually situated at the end of the conversation, in order to access the system and its many dimensions, I always inquired about the goals, steps, and potential problems of getting to the proposed unification. Hence the topics “Goals” and “Problems” were developed. It is interesting to note that the identification of goals and problems are similar in all parties. All sides seem to agree than a unification of the forces is the goal and problems along the way are the historical mistrust, the interference of foreign powers, the lack of institutions, and the mentality. However, when asked for the intended steps of getting towards this goal and around the pronounced problems answers were largely vague or completely absent, with the exception of Gorran which produced a political program for the implication30. ! During the interviews, and through the research subject’s identification of current deficiencies of the status quo, insight was gained into the formal way of doing things (structure) on the one hand and the unofficial way of doing things that actually decides matters (system) on the other hand. This is how the topics “Structure” and “System” were created31. Since the structure is available in prior analyses, Stansfield and others, as well as it is available online (KRG.org) this information will only be addressed in the form that is relevant to highlight the underlying system. The interest of this research lies in the analysis of how things work “aside the rhetoric” (Feaver 1999). ! At this point, some biases have to be pointed out. The system is at times a politically sensitive topic. While issues like corruption, nepotism, closed-door-deals, and shadow elements are more and more spoken about today, they are still sensitive when reaching the point of calling names. 30 This general lack of vision towards the intermediary steps also fits into the assessment of the Peshmerga from a foreign military advisor (non-disclosable) who stated that “Kurds are reactive. There is no prior or subsequent analysis of events. There is no analysis of patterns, prediction of future, calculation of plan etc.” (OTR). A research subject supported this, arguing: “We need to create a chief of staff: high ranking officers who build a high level strategic view for the country as opposed to right now where you have local forces defend one part and you do your thing and I do mine.” (i17) 31 During the interview process I did not use the terminology as I had not yet come up with the way to capture “this” and “that”, but I did guide my questioning towards an emphasis of “how does that work in practice”. "22 Therefore, for the safety of my research subjects, I maintain the right to not disclose sources or names in the examples32.! Other biases of the research are a cognitive bias, the tricky part of language, as well as the personal background, psychological and behavioral factors, and the setting of the interview.! Even with disregarding existing models and theories of military science on structure, organization, or recruitment, there remains one bias that is hard to get around to: thinking happens by comparing new information to something familiar. Hence, even without designing the study by applying a purporting model, the task remains of understanding the observed system without trying to fit whatever there is into preexisting ideas of what it “should” be. Language is a hindrance in that33, as is the thought process of trying to understand what is actually going on34. Instead, I tried to approach the field work in the metaphor of building a puzzle of which I neither know what it is supposed to show in the end nor how many parts there are to be put together. I tried to judge the system and new information about the system only in the terms of what I have come to know about the system itself and I tried to stop myself deliberately every time I started comparing what I saw to what it ‘should’ be or what I knew of other military systems. The goal, after all, is to understand whatever system and structure there is in the relationship between the civilian government and the Peshmerga forces in the KR-I.! The background of the interviewed person also played a role in the interview process. A recognizable difference between Kurds that returned from the diaspora and those who had lived all their life in the region was recognized. The later category was less critical in their statements and more prone to tell the “official line”35. ! When referring to psychological and behavioral factors, I found that there was an effect on the research subject depending on whether or not they trusted my translator or the setting in which we met. Most of the time I addressed officials in their office and military generals on the front lines. I found, however, there to be an even more critical tone of voice when meeting some of my research subjects again in unofficial settings and off the record. This can be partially ascribed to there being other people in the rooms during interviews, or the distrust displayed agains the translator. It is an invaluable asset to have a translator who the research subject knows and trusts, in order to hear 32 All interviews are listed alphabetically in the Bibliography, coded by “i” (for “interview”) and the increasing number (i1 - i27). This coding decreases immediate association of the statement with a name. Information provided “off the record” as well as sensitive information is cited with “OTR”. A disclosing of ORT is possible for the review of the thesis but is concealed for any other reading or publication. 33I experienced that sometimes we used the same words but meant different things. When I realized that, I took specific care to find out what persons meant by using particular words. 34 This problem was tackled by triangulating the given information. For example, I repeatedly asked informants “Do I understand correctly that…” and rephrased issues in the terms I had understood them. 35 “The official line is something they tell us [meaning foreigners] but it does not exist.” (OTR) "23 the “true” story36. Furthermore, this research has to be understood at the unique point in time it is undertaken. On August 17th, 2014, ISIS advanced towards the Kurdish territories, penetrating an area dangerously close to the capital Erbil. And my field work commenced on August 25th. This situation of war influenced the research in several ways: One, the war made Kurdish officials realize the need for military efficiency and coordination, which in turn triggered an increased political cooperation in working towards unifying the forces. This intensified attention to the subject produced changes in the structure that might likely have taken more political bargain at another point in time. In addition, this wide political interest opened a lot of doors for with people supporting my research topic. And two, the war provided the unique possibility to observe the actual workings of the system ‘in action’ on the front lines in addition to the ‘rhetorical’ and political reflection on the Kurdish civil-military relations.! ! Findings and Analysis! ! As explained, the analysis of the qualitative interviews to this research is concentrated around five topics: History, Structure, System, Goals, and Problems. Each of these topics produced one or more findings, which are accordingly clustered into categories. “History” shows the ongoing divisions in the KR-I, “Structure” shows there is a system (because of weak and corruption prone institutions), “Goals” and “Problems” allow to identify money and mistrust to be central variables in the system, and “System” led to identify the divisions (found in History), a certain culture (split in sub-categories: loyalty, personalized system and civil-military-gap), and the weakness of the institutions (found in Structure) to be additional variables of relevance. The categories were developed in a step by step process. First, the transcribed interviews were read and quotes were attributed to the five topics. In a second step, each topic was reviewed and within it categories were clustered along the different issues raised. Thereby, the identification of topics assists the sorting of findings, while the structure of categories aims at aiding the reader’s ability to follow the argument more than it helps the value of the discoveries. Therefore, the presentation of the findings is oriented along the main argument of the thesis and intersects topics and categories in order to avoid repetition: First, topic “Structure” addresses the institutional set-up of civil-military relations in the region. It becomes apparent that the official structure of institutions is weak in decision-making capacities (category Institutions, with a special focus on military institutions). Instead, the power is located in the parties (category: Party Politics). Topics “Goals” and “Problems” focus on the deficiencies of the current system and thereby provide a first glimpse into its nature: Corruption, weak institutions, and mistrust are identified to be core issues. Others are foreign power interference, the dependency on Iraq, and a specific mentality. Last, the topic 36For example, the expression “Kurdish truth” has been used to differentiate the “official line” from what is “actually” going on. "24 “System” introduces the central aspects for comprehending the system: Topic “History” is presented in the category Divisions and demonstrates the many divides in the KR-I (historic- political, territorial, and military) while category Culture allows to identify a centralized localism and personalization as well as a non-existent gap between civil-military spheres.! The findings presented are chosen examples of quotes that best display the information provided during the interviews37. After their presentation, each category closes with an analysis of the presented data. In “System” preliminary-analyses after each cluster are provided to aid the reading process.! ! Structure! While the differentiation between structure and system is the core part of this thesis, structure is already relatively well accessed in prior research. Therefore, findings such as the internal party structure and military set up are ignored, as they are well documented (Stansfield 2003, Chapman 2011). Instead, the examples presented are chosen along how they highlight the differences between “the official38” and “the actual”. ! • “The official structure is more political rhetoric to motivate the popular state of working towards a united Kurdish front and to give the impression, especially 2003 in Baghdad, that we are Kurdish people and show a united image. But this official structure is just the surface.” (i17) ! • “There are unwritten rules here. If you come in and you don’t know anything about the society it looks proper. But there is a division deep down. They need to reform the system from the roots.” (OTR)! • “In Kurdistan, democracy and elections have not become the game in town. You don’t have a strong judiciary, you don’t have a strong legislative, all you have is a very strong executive and those in the executive are also the persons and members of politburos of those two parties.” (i19)! • “Okay, when there is law, I can say that the executive they try to implement it and the police and other parts of the government they try to do it. But that is only because by the time it is passed as a law, it has already reached an agreement between the parties, only then parliament passes a law and executive (ministers) are all politburo anyways.” (OTR)! • “If I, and I am [party member], commit a crime and no one sees it, [party] has to agree before I am arrested. If a police or a security officer commits a crime, unless it is public, the head of the Ministry must agree before an arrest can be made.” (i19)! • “When Gorran was founded, 5.000 people were sacked because they voted for Gorran. PUK and KDP have different tricks [to find out who you voted for]. They make you swear by putting 37 Due to the word limitations of this thesis, only a few quotes can be shown as examples to each topic. 38The “official line” is presented to be a unified, institutionalized, democratic government, with the matter of unifying the forces being a mere “technicality” (see “Problem”: Budget). "25  your hand on a Q’ran, or swear on your wife. They also force people to take a photo of their ballot paper with their phone.” (i19)! • “The problem is now the political parties and the institutions are interrelated. And the problem is that most of the important decisions are made outside the government. For example, yesterday there was a meeting of Nechervan Barzani with heads of all political parties on how to deal with the Iraqi government. All of these parties are in the government. Why don’t they decide this in the parliament? They meet outside the institutions and except for Nechervan Barzani and Qubad Talabani nobody has a government position. Yet they make such decisions.” (i2)! • “[Party leader] has not been in any government position since 1991. He always has power but he has not been in any government position.” (i2)! • Institutions! • “In our region, the legislative is not powerful like in other countries. … They don’t send the powerful persons into the legislative. In KDP for example, the most powerful persons are Barzanis. So that time four years ago, no one from the Barzani family was in the legislative power. All the powerful men, the ones that have power, were in the party. … The parties they see […] the institutions [as] just a formality, not a principal they believe in. It is good because they can tell other countries we have democracy.” (i9)! • “The power comes from the social and political system. […] Here to be successful in your election the senior people in the party must support you. So as long as you are loyal to the person who appointed you, you stay in your position.” (i23)! • “The legal system is very politicized. The judges are selected by the political parties. KDP region has KDP judges. PUK region has PUK judges. So they are controlled by the political parties. And of course the parties pick judges who have no ethical stands in their job so they work for the political parties.” (i17)! • “We have a draft of a constitution now […] if you look at the draft it is mind boggling. It says the system in Kurdistan is republican parliamentarian. But if you look at the powers of the president, it is a presidential system… [Example impeachment:] 2⁄3 of the parliament, according to this draft, must vote in favor that the president yes has devalued the constitution and should be impeached. It does not end there. It goes to the judicial council, which has 7 members, these 7 members are all appointed by presidential decree […] Out of these 7, 5 must agree that yes he truly indeed has preached the constitution or the law […] And they cannot vote unless everyone is there […] So if a judge does not show up, it does not happen. […] In addition to all of this, this draft-constitution was written by a committee formed by KDP and PUK. Not parliament. Not people elected by people. And it was drafted before 2009 where KDP and PUK had 82% of the votes. When it was brought in front of parliament for a vote […] they had 48h to read and vote. This constitution in Kurdistan is tailor made for a person who has absolute power.” (i19)! "26  • Military Institutions:! • “The system in theoretical, in framework, is there. The minister is there. Everything is there. But how to make a system function correctly […] it is not easy. Because these two political parties they have created their own armies before when there was a conflict in 1990s […] They have an ideology, even if you mix them, the one who belongs to PUK still is PUK member.” (i5)! • “Yes I [the Minister of Peshmerga] still feel the influence of the political parties. But I also find that the parties must be willing to truly unify because I have experienced no one putting stones in my way so far.” (i27)! • “There was never the case in the PUK for example of [former Minister of Peshmerga] actually controlling the MoP. He is in control of the PUK interest in the ministry.” (i17) ! • “In November 2009 there was a meeting by the US officials who met the KDP and PUK officials to discuss how to unify the Peshmerga forces. The US would provide the money, but they would not accept […] any other militias outside the Ministry. They never reached an agreement in this process because in the second meeting, Talabani and Barzani […] were objected in keeping no militias of the Peshmerga outside the MoP. The KDP wanted to maintain a special force. PUK wanted to maintain a special forces. - Why? - It is obvious. It is because they see themselves above the KRG. For example, Masoud Barzani attended in the parliament just I think twice so they consider themselves above the institutions of the KRG. And the second because there has always been mistrust between KDP and PUK because of their history. They don’t trust each other and they do not trust that one day maybe one of the forces attack another one.” (i2)! • “The president’s wife for example has two brigades. Two brigades she controls. She is not a military commander. She is not an official. But there are two brigades who are supposedly for her protection, but she controls these brigades. They get their wages from her directly. They are loyal to her.” (i17)! • “The party use their forces for their self-interest. Example: During the demonstration of February 2011 in Sulaymaniyah [PUK territory], Masoud Barzani and the KDP forces, it was around 3.000 Peshmerga fighters, they were coming from Erbil and they were planing to stop that demonstration. … President Barzani also asked Kosrat Rasul [PUK Politburo member] to help with his three brigades but he said no.” (i2)! • Party-Politics! • “A lot of the temptation of the [party] has always been: they have guns, they have money. They can shoot you, arrest you, or pay you.” (i17) ! • “The big decision makers of this region are still the leaders of the parties. … Just because they are decision makers does not mean they make the decisions together, that is why PUK is heading towards Iran and KDP is heading towards Turkey.” (i2/ i21)! "27  • “The parties get their budget from different sources: KDP and PUK used to split the entire KRG budget between the two, now the budget goes to the KRG and is distributed according to the seats in parliament, but more than that, every party has TV stations, newspapers, radios, they get money from clientalism, patronage, and corruption in the big business sectors such as oil, cement, telecommunications, infrastructure projects etc.” (OTR)! • “A lot of these people, a lot of their power did not come from their greatness of individual person, it came from the political background that they had. The personal recognition comes from the politicized nature of the military, comes from the fact that an institution, that is not linked to a party, does not exist yet.” (i17) ! • “They [the parties] see the government institutions as a kind of personal property. Like a car of mobile or chair. Example: Where is the building of the president? It is the Headquarter of KDP.” (i2)! • “Here [a party] can pick up a phone and fire the head of an office, the head of the students, or the teachers union. Why? Because the party is everything. The party puts people in position, on the condition that they listen [meaning ‘follow loyally’].” (i19)! • “Party affiliation is more important than capabilities. Right now you get a position not because of merit but because of a letter of recommendation by a politburo or central party member.” (i21)! • “You can not be accepted into Qalachulon Military Academy if someone from the politburo does not support you. And the same goes for Zakho college in Duhok.” (i21)! ! In the presented findings, the first line of statements refers directly to the claim this thesis started with: There exists an official structure, a democratic image, different institutions, and elections while, parallel to it, a system of party-domination is maintained. There are laws, but the parliament is not the decision-making institution and the judiciary is party-controlled. The Peshmerga show an institutional divide in the MoP and the UPF, but large parts of the forces are still based on the parties where the lines between political and military influence are blurred. And category Party- Politics highlights the extent of control the parties still maintain over their territory, their resources, and to large extent their people. ! ! In CMR theory, the analysis of civil-military relations primarily and traditionally focuses on the institutions of the civilian government and the armed forces of the state. When deconstructing the Kurdish case, these institutions are identified to be present. However, as becomes apparent from the findings, the institutions of the region yield limited decision-making power. Therefore, one needs to question whether an analysis of the institutional structure is sufficient in comprehending the civil-military relations of the KR-I (see Pion-Berlin in Revising Theory). The central finding and argument of this thesis is that an assessment of the institutional framework alone holds no explanatory value. Instead, the focus shifts from the perceived institutional gap between the "28 spheres towards the analysis of the differences between structure (institutions) and system. Thereby, identifying the nature of the system becomes a central part of CMR analysis. Two approaches are taken to do so: First, by asking for the future “Goals” and current “Problems”, first conclusions onto the current system can be taken. And second, questions of ‘how’ and ‘why’ allow access into the constituting basics of the system: the historic divisions (“History”), and a culture of centralized localism and non-existent gaps.! ! Goals and Problems! The central goal the parties agree upon for the Peshmerga is unification. This merging is seen to be vital for reducing the threat of another civil war, for supporting the state- and nation-creation process (i8/ i6), for satisfying outside pressure, and for the protection of the region39 (i18, i26). In this context, “institutionalization” and “professionalization” are terms that are regularly thrown into the equation. By asking the question of “how” to unify, to institutionalize, and to professionalize, first conclusions towards the wished for structure as well as the current system can be taken away.! • Control the money.! • “The key to changing the bipartisan nature of the Peshmerga is the wage. If you are a Peshmerga force and you get your wages directly from me, the person who you protect or the person who is your leader, your life is in my hands. If I can cut your wages, I can control your life… When your wages come directly from the ministry, the person who you are assigned to protect can no longer give you orders. You get your orders from the centralized commission.” (i17)! • “The first step is that the KRG has to cut the budget off 70 and 80. If they have no budget, they can not be active. After that both 70 and 80 will be reorganized inside the MoP.” (i3)! • “We want to make sure that every Peshmerga receives his salary through a payroll in the bank. Because the generals misuse the money transfer and the privilege it gives them. Not surprisingly, the generals are against this.” (i23)! • “[The generals] go to senior KDP and PUK members and they say, listen, they [Gorran] want to take the money away […] Tomorrow when I need my unit, they won’t follow me. And if they don’t follow me, they won’t follow you.” (OTR)! • “We can not get the money [back], the one that was already [privately] accumulated, but we can stop them. We need a framework of creating money for the KRG instead of creating money for the parties. Unfortunately, the parties are not very interested in that […] because they see the power though the money that they get for themselves and their party.” (i23)! 39 “When ISIS fires a bullet, they do not distinguish between KDP and PUK.” (i13) "29  • The issue of “ghosts”40: Following the money in Kurdistan is a tricky business. As one source tells me “we have a bagging system, not a banking system” (OTR), meaning that cash in paper bags does not leave traces like a digital transfer. What is officially known is that the system used to look like this: Money from Baghdad41 went to the Kurdish government and was distributed evenly to the parties, who then handed the money to their clients. Thereby loyalty was rewarded, criticism punished, and since they support their military units directly they also maintained a party based monopoly of force. This system has changed, however. Now the money comes to the KRG and goes to the ministries42. The MoP then distributes the money to the units 70, 80 and the UFP. However, the MoP does not know how many Peshmerga there actually are. Each commander gives the number of soldiers in his command to the next higher level and collects the according salaries. Without any means of control for these numbers, however, “a general can say he has 700 but he actually has only 200 men” (OTR). These effectively non-existent men are called “ghosts”43. Additionally, the distribution of money through these channels leads to dependencies of the soldier towards their generals, which is sometimes exploited44. The goal of wiring the money directly through a banking system to the individual soldier is defined to address these problems: One, by individualizing the payment (issuing bank cards), the “ghosts” disappear. Two, through the banking system, the money flow becomes visible which makes the corruption easier to be limited and monitored. And third, the dependencies to the superior dissolve as the money is not dependent on loyalty anymore.! • Law and Institutions instead of Party Politics! • “Politics is damaging the region. We need to depoliticizes the system. - How? - You depoliticize by allowing young generations on their own merits and capabilities and education and training 40This section taps into corruption, which is a very sensitive issue. For the safety of my research subjects, the information given is paraphrased, connected, and not quoted. 41 Iraqi Constitution 2005/06) Art 121,3. 42 With the ongoing budget discussions with Baghdad, however, the source of this money derives not from the federal government but from the official oil pumped towards Ceyhan and the unofficial oil smuggling parallel to it. “The Ministry of Finance belongs to Gorran, but the MoF does not produce money. It gets the money from somewhere: from selling oil. Officially and unofficially. … At the moment we turn a blind eye to the smuggling of oil because the oil money we pay people is theft money. … We can not do anything against it because we do not have enough information and the times are very critical.” (OTR) This unofficial oil business is controlled by the parties. The parties contribute a voluntary share, which then also contributes to the money given to the MoP. News reporting in December 2014 additionally uncovers the involvement of military generals in the oil smuggling. In a 2014 study, Zukerman-Daly points towards the necessity of including these “middlemen”, their interests and influences, into the military equation: Since they have a stake in the system, they are interested in maintaining it. 43The existence of these ghosts becomes apparent when comparing the different accounts of troop strength. During my research I heard the numbers 160.000, 170.000 and 206.000. Other example: How many people do you lead? “I don’t know. I direct four generals directly.” (OTR) 44Stories where the general required soldiers to help his wife clean the house before receiving their salary abound. "30  to rise and take control of security institutions. Instead of needing a letter of recommendation from a politburo or a central polity member.” (i21) ! • “First, pass a law to unite the forces. MoP needs to control the salaries. Second, reorganize them from scratch. Disregard the political affiliation.” (i19)! • “Military people should be separated from politics. I mean they should not belong to KDP or PUK. It should depend on my personal merit and training and education and not on political affiliation.” (i21)! • “Build a new force beside the force that already exists right now, but based on modern criteria and no longer based on you know some guy or are supposedly a revolutionary leader. It won’t matter which political party you come from, they will require proper training…” (i17)! • “Build a proper chain of command that is no longer about the politburo member and not because of the political character and not because I am the second cousin of so and so.” (i17)! In order to reach these goals, several problems were mentioned:! • Mistrust: ! • “Building trust is not easy - so this lack of trust, this bloody background has affected the process of unification.” (i11)! • “Mistake one: not demobilize forces after 1991, two: foreign country involvement, three: not demobilize forces after 2003/05. - Why not? - They don’t trust each other. - Why? - Because of history.” (i1)! • “The problem is a highly politicized society as a result of the internal fighting. We have a lot of mistrust and a lot of rumors and perception. We are dealing with perception more than with reality.” (i8)! • “PUK and KDP are afraid of each other. There is no trust between them. The only thing to make it better, even if slowly, is to increase trust in the elections.” (i16)! • “The biggest problem of the military, everything else aside, is that is is bipartisan, it is politicized. This is because they both originated from revolutionary means. Then there is the civil war. There is complete distrust and fear that if I decrease my military force, the other side would take me.” (i17) ! • “They don’t want to lose their soldiers. They don’t want to be unified. Because they want to protect their family.” (i22/i23)! • “Dismantling the Peshmerga now would create a problem. They don’t trust each other. Let them keep the Peshmerga they have now, create a new force independent from them, and then you can start taking their Peshmerga away slowly. Who is capable can come to the new system, the others retire.” (i17)! • Budget:! • “We tried to fasten the process but the problem was the budget and the ability. As I mentioned, we did not receive anything from the Iraqi side for eight years.” (i12)! "31  • “In the unification, the budget for equipment was the problem, not the party division. Since it was the party leaders who decided on the unification, there was not much resistance. There is just not enough budget.” (i18)! • “Right now we pay the Peshmerga by cutting off budget every month from all other ministries, because the Peshmerga do not have a budget. They are supposed to receive this money from the Iraqi defence system.” (i11)! • Foreign powers:! • “The regional powers do their best to keep the Kurds divided along the parties and families.” (i21)! • “Nobody wanted this experience to succeed, because there are Kurds in Iran, Turkey, Syria. They [the states] are concerned that this would effect the Kurds of these countries.” (i7) ! • “The problem is we are a new democracy, we still need help and we are caught in a net of power rivalries.” (i7)! • “It is Iran who wants PUK to have their own forces to use it in the way Iran wants. Because they see that Barzani moves towards Turkey and he is more orientated towards Sunnis.” (i23)! • “We do not have a national strategy, that is a problem. We need to have a neutral stance because it is a nasty neighborhood. We need a balanced relation between the two neighbors.” (i19) ! • “We are surrounded by countries that do not want Kurdistan to be strong. And their tool is Baghdad.” (i11) ! • “Iraq deals with us like we are the enemy sometimes. Did you know that right now they are still paying salaries to the territories under ISIS control? They still don’t pay Kurdistan.” (i25) ! • Mentality:! • “The biggest problem we encounter is that people think that this is the only way of administering a country. People who lived here their entire life are after loyalty.” (i23)! • “People are afraid of their own government. They whisper and turn around before they say something critical about a party or the system.” (OTR)! • “When we [Gorran] established a media in 2005, we were just spreading ideas. We spoke about the main problems: Corruption, inequality in the government, favoritism, the bad in the ministries, the relationship to Baghdad, tribalism of KDP, family structure of PUK, Peshmerga wages, transparency … We also explained to them what [terms] mean, for example what is a budget. People did not know.” (i9)! • “Another problem is that in all of Iraq people never understood the military as a means for defence outside but always as a means towards the own people, to calm the population, to break protests, to kill the opposition”. (i17)! • “Those people who prefer to make the parties priorities, they have their own interest in the situation. - What kind of interest? - Those kind of people sometimes they have aim but "32  sometimes they are without bad aim but they are used to work like that because the parties are just closer to them than the nation.” (i3)! • Institutions and corruption:! • “In my opinion having a unified force in Kurdistan is somehow difficult. Right now we do not have a constitution. If you do not have a constitution how you could establish an institution, let alone a unified military institution?” (i25) ! • “The problem is in the system. It is not just about changing an individual minister, like in Kafka, there is a system that is a nightmare and forces you to resort to ‘wasta' or favoritism.” (i19) ! • “One obstacle for institutionalization is oil, there is too much money outside the government. Also the judiciary system is biased towards the parties and corruption is a big test.” (i2)! ! The topics “Goals” and “Problems” are based on the questions during the interviews that inquired this sort of information. It is regarded that through the assessment of the future and the potential problems in reaching this goal, it is possible to draw conclusions onto the current state of affairs. Topic “Goals” reveals the importance of money, as well as it gives an additional confirmation of the institutions being weak (as does the identification of weak institutions being a problem). What is best captured in the concept of clientalism and corruption, refers to a system in which money is the glue between the party (the patron) and the people (the clients) that accounts for the prevalence of unquestionable loyalty, “wasta” (favoritism), and obedience. It also serves as a support to the priorly declared “hierarchical nature” of the Kurdish society and the reason of why this research focuses on the elite. Until today the power still resides outside the institutions45, particularly because there is still money outside these institutions.! Topic “Problems” allows to conclude that there is a Prisoner’s Dilemma situation between the parties (mistrust) and their populations (mentality), which is influenced by the greater regional dynamics (foreign powers). In addition, the category also supports the prior conclusion of money being a central issue that is still not regulated (budget), as well as the general lack of institutionalism (institutions and corruption). Having already discussed the weakness of institutions and the relevance of money, the continuing mistrust of the parties with its influences and consequences presents the focus of analysis. Robert Griffith (2015) identifies trust to be “the glue that holds societies together” (Griffith 2015). Thereby, deficient trust in the society is often seen in the form of clientalism structures (meaning the favoring of regular business contacts over competence), patronage (the prioritization of family and proteges), as well as cronyism (the favoring of friends) and corruption (ibid.). More than that, by referring to Robert Putnam, he states that interpersonal trust is directly linked to trust in institutions (also called “generalized trust”) (ibid.). Having already established the prevalent system of low institutionalization, clientalism, favoritism, 45The weakness of the current institutions is also seen in the explicit goal of “institutionalization” referring to the increased allocation of money (and influence) to them. In this context, the goal of “professionalization” is seen as an antidote to the allocation of jobs according to family- and party- relations. "33 and corruption, the identification of mistrust as one of the core problems in the way of unification is confirmed by Griffith’s assessment. And by taking Griffith’s reference to Putnam one can also conclude that the mistrust between the parties and their populations is part of the reason for the low level of institutionalization in the region, as there is a lack of trust in these institutions. Combining this mistrust and low level of institutionalization with continuing maintenance of the parties’ military forces as an expression of this existing mistrust further confirms the fit of the Prisoner’s Dilemma model. And as it is concluded from the section “mentality”, it is unlikely that the population will have an impact on this division for now, as large parts are still used to the way of doing things through the party and not the institutions.! This problem is, just as the influence of foreign powers in the region, also connected to the fact that the KR-I is only de-facto independent, which means that it is all the more effected by Baghdad (particularly regarding the budget) and the surrounding neighbors. The Kurds are stuck in a regional power struggle, between Turkey and Iran, between Sunnis and Shias. There is no agreement on whether it is foreign interests exploiting the division or whether it is the parties exploiting the regional divisions in order to maintain influence or even increase strength against the other party. However, since the focus of the study relies on a theoretical vacuum perspective, these aspects are no further considered than to be influential on the continuing mistrust between the parties. Therefore, to sum up, the most important findings of this section is the relevance of money, the reconfirmed weakness of the institutions, and the continuing mistrust between the dominant players of the region, the parties. In a following step, these first insights into the system will be extended by the subsequent findings on the nature of the division in topic “History” and the relevant cultural aspects in topic “System”.! ! System! Examples for Division46:! While never actually asking for history, history is unavoidably part of the answer in Kurdistan47. Since 194648 and even more so since the first signs of civil war in 1965, every step along the way proves relevant to understand the current divisions in the region. History is used to explain, to defend and argue, and, as these findings will show, also to interpret. Thereby the differing interpretation of history itself already highlights a relevant divide between the parties. While some aspects of history are widely agreed upon, others are not. This differentiation leads to conclude towards the political nature of the division and is therefore, for the ease of reading, presented in a 46 Category Division is derived from combining the findings of topic “History” with the finding of division in topic “System”. 47 There is a joke circulating saying that if you ask a Kurd for the weather, he will likely start with “In 1946…”. 48 When the declaration of the Mahabad Republic in Iran was the first nine-month experience of Kurdish statehood. "34 separate paragraph. Other divisions, such as the territorial and the military division, are subsequently addressed.! Historic-Political Division:! • What is agreed upon:! • There used to be an ideological difference, but not anymore.49! • “Until the uprising, there was an ideological difference between KDP and PUK. Any clash between these two was a clash between two visions. After the uprising it converted into a personal clash between Barzani and Talabani about who controls more in Kurdistan.” (i23)! • “There used to be a difference, but now both parties have a similar structure - both focus on personalities.” (i1)! • “These differences between the two parties are not principals. At least not very big ones. They are two families. Barzani and Talabani. Everybody knows that. And they established for themselves so many economic sources and fought with each other in the civil war. Now we are living in a period, the civil war is practically finished, but politically is still continuing.” (i9)! • Divisions are deep. While there is no agreement on how far back one goes to explain the roots of the division, everybody agrees on almost unreconcilable differences.! • “The division goes back all the way to a river that separates the principality of Badinan and Soran. The differences we see today, in the language, the culture, and the politics all date back to this river.” (i1)! • “The division already started in 1514 when the Ottomans and Persians fought and decided never to let this region be one.” (i19)! • “Since 1994, when the internal fighting started between the PUK and KDP, we have two administrations, two Peshmerga forces, two command and control centers, two everything…” (i21)! • “The civil war deepened the differences.” (i16)! • “Today, Mister Barzani [KDP] is still not ready to come to Sulaymaniyah [PUK] and give a lecture at the university and students come and criticizes him.” (i23)! • The region has come far.! • “If you look at the government, there used to be 50-50 share between the parties, now it is 38-24-18-10-etc. This is a sign of political maturity and of how far we have come.” (i7)! • “In order to avoid that story which happened in the civil war, we will focus on some points: there is a KRG which consists of 5 main parties, there used to be only two. This point is one of the evidence to tell us that this story will never happen again. Second, ISIS makes them more unified. […] And third, there is a new generation coming.” (i3)! 49 The difference referred to is between a leftist-marxist-revolutionary idea represented by the PUK vs. the traditional, tribal ideology of the KDP. "35  • “We have seen a transformation from a revolutionary movement to administration. To fight an enemy was easy. To be in charge of the civil affairs of the people is not easy.” (i6/i7)! • Who are the Peshmerga?50! • “They are the ones who voluntarily went to the mountain to fight the Saddam regime, to fight for Kurdish freedom.” (i8)! • “Peshmerga is contained of two words. Pesh and merg. Pesh means front and merg means death. It means those […] who chose death in front of themselves to protect the people and the country.” (i12)! • “The Peshmerga is the political guy. They have a target: to protect the lands called Kurdistan. Not everyone who holds a weapon is a Peshmerga.” (i18)! • “The name of the Peshmerga is holy for us. All of us we were Peshmerga in the past.” (i6)! • What is disagreed about51:! To avoid consequences for my research subjects, the information for this section is largely paraphrased (some quotes could be traceable to the speaker), connected, and not disclosed. To fully understand the “connotation” in the telling of history mentioned, one analyzed example is provided:! • KDP: PUK was created against Barzani52. ! • PUK: PUK was created because of the ideology of egalitarianism.! Technically both statements share the same information. While the KDP is based on traditional values and modes of operation, such as maintaining the hierarchical structure of tribal societies, by mobilizing tribal sheikhs for support and votes rather than individuals, the PUK is originally founded on a marxist-leninist idea based on egalitarianism. With egalitarian ideas being contra towards feudalistic structures, as well as succession based on blood and family relations, the PUK is naturally created in opposition to the KDP. But while the first statement hints towards a personal animosity, the second hides this connotation and instead highlights the inherent values of the party.! • KDP: focus on the fact that PUK had more strives with more parties aside from them (with socialist party, communist party, with islamists party etc.)! • PUK: focus on the part where KDP aligned with Saddam Hussein to take over Erbil 1996! ! 50 “Understanding the historical impact of the Peshmerga is [..] valuable in analyzing the cultural affinity the Kurds have towards their soldiers and why they are so hesitant to disband them in the new Iraqi power structure.” (Lortz 2005: 2) 51 The diverging interpretations presented are not exclusively divided between KDP and PUK. Also other parties and personalities have the tendency to highlight certain aspects of history. The divisions in the presentation, interpretation, and connotation of events coincide with the information received in my interviews. 52It has to be noted that in all parties (KDP, PUK and Gorran) the founder and leader stands not just as a representative but most often also as a synonym to the party. "36  • KDP: “We said it is enough with the fighting. We established the Kurdistani Front. We had 51%, they had 49% in the election. In order to cool down PUK and in order to establish institutions, we agreed okay let’s have 50-50. When the KDP leadership agreed on the 50-50 it was for keeping conditional security.” (OTR)! • PUK: In the first elections, the result was 50 - 50.! ! • KDP: PUK betrayed the revolution and now they are all about the family as well.! • PUK: KDP wants all of Kurdistan (Greater Kurdistan) to be ruled by their family.! ! • KDP: KDP once was the only party.! • PUK: “The actual party of KDP was created in Sulaymaniyah [PUK territory], but Mustafa Barzani, because he was a tribal sheikh and he had so many armed tribal men following him, he managed to take control of the party.” (OTR)! ! • KDP: 1992 we dissolved all our party Peshmerga, had them join the ministry. PUK did not dissolve their forces entirely because they were afraid they would lose the 1994 elections.! • PUK: KDP paid off tribal leaders and spoiler parties (“political shops”) to boost their votes (referring to the unification of Kurdistan United Party with KDP before elections in 1994).! ! Examples for other Divisions:! • Territorial division:! • “Kurdistan is two, up to now, two governments. One country controls half of it, the next regional neighbor, the other half is controlled by the other neighbor.” (i19)! • “So there is two zones in Iraqi Kurdistan, one turkish and one iranian. This fact gives the two parties an excuse to make themselves more powerful than before.” (i9)! • “The Directorate of Border control is divided along the parties according to their territory.” (OTR)! • “There are two military colleges. Two parties. Two territories. Even the way the residency gives visas for foreigners is different.” (OTR) ! • Would it be possible for KDP to defend PUK territory? “Yes, we are doing it now. In the war in Jalawla [PUK territory]. 15 Peshmerga were killed and 7 of them were from the KDP.” (i11)! • Peshmerga division:! • “At the moment there are three kinds of forces. You have a group of forces and brigades that are under the control of the MoP but within those forces, if the commander is from PUK, the deputy commander is from KDP […]; it is 500 from each party and the General from PUK will have the control over the 500 from the PUK and the Deputy General will have control over the "37  rest. Outside that we have two forces, unit 70 and 80. […] They are absolutely under the control of both parties.” (i19)! • “Even in unified brigades we still feel KDP and PUK division.” (i24)! • “There is one overarching commander per area. [MoP: 8 areas (i3, i12, i26, i27)] He coordinates with the commanders of the units. Each unit has one commander, and one deputy. They are in charge of their units, their brigades. They can decide about administrative issues and are ‘legally’ linked to their units. But when there is a military operation, everybody has to listen to the one highest commander.” (i5)! • “The commanders currently heading the eight operation rooms were chosen by the parties. They were already in place by the time I [the Minster of Peshmerga] took office. But right now there is no alternative. We are at war.” (i27)! • 70 unit in KDP land: “Most of the operation rooms here are headed by KDP. In the green land they are headed by PUK. We follow the operation room. They coordinate with the ministry and the president.” (i13)! ! The category Division confirms the asserted depth of the divide between the parties. On the basis of historic events, different layers of division still exist today. These are territorial, political, and military. The most relevant conclusion drawn from their conceptual separation is their interdependent relationship. And the different connotations shown in “Historic-Political Division” allow to conclude for a degree of tensions to exist in this divide.! The territories of the Kurdish Region of Iraq were historically split by a river between Badinan and Soran, by a different dialect, and most recently by two parties. These lines of territory have moved during the civil war but have since been frozen. The division is polarized between the parties, who are each headed and represented by one family (Barzani in the KDP and Talabani in the PUK). While these differences initially arose out of ideological disagreements over the traditional, hierarchical, and tribal based or egalitarian administration of society, today the dispute relates to the party and family wanting to increase their personal influence. Until today they maintain clear zones of influence53. What these findings show is that despite the first steps of unifying forces, and the rhetoric of KRG rule, there is still a strong interrelation between territory, party, and power. Given this association of the party with the state, government, and administration was the predominant way of organization for a long time (Stansfield 2003), it is not surprising that it still partially exists today. As Hemin Hawrami (2014) explains: “For 80 years, Kurdistan was in a revolutionary part. The [party] was almost everything: It was a state that was delivering justice, it was an institution that was protecting the liberated areas, it was delivering security, education, and even giving salaries.” What I argue, therefore, is that, despite the presentation of problems in the region as “state-internal” party-accommodation issues, it is vital to recognize and include the 53  See Appendix for election results. "38 continuing division, for example by applying the Prisoner’s Dilemma model. Particularly also since the division of territory and according power of the party still relates strongly to the set-up of the Peshmerga today: The front line with the Islamic State is divided into eight administrative sections, each headed by one central command, whose head has not been chosen by the MoP. Instead, they were selected by the party in the territory. Communication and coordination happens along the eight administrative sections and the following official structure: The KRG president (Masoud Barzani) heads the General Command Staff, which is made up of the commanders of unit 70 and 80 (Sheikh Jafar Mustafa and Faruq Zirwan respectively), the Minister of Peshmerga (Mustafa Said Qadir), the Minister of Interior (Karim Sinjari), and the Leader of the National Security Council (Masrur Barzani). The president is in charge of all military decisions and is advised by his General Command Staff. Decisions of war lie with him, and the management of day-to-day affairs with the Ministry54. While this set-up looks “proper”, it is relevant to note that the unit on the ground does not know whether the command or money was given from the president to the MoP, to the unit commander, to their head of the operation room, and then to them, or from the president directly to the unit commander, or from the politburo to the unit commander to them, or whichever other way possible (i14). It is this example that manages to highlight both the difference between structure and system as well as the inherent connection between territory and party-power.! ! Examples of (Political) Culture:! • Loyalty:! • “People who have lived here all their life, they are after loyalty. They want to prove that they are loyal to this leader or that leader so they get a post, they get a salary, they get married, they have a flat and this and that and that is it. That is the only way they see how country is run.” (i23) ! • “Ask the people: What are the good characteristics of your leader? They will say because he is the son of Malamstafa Barzani or because he belongs to the people of Mam Jalal Talabani. Because Mam Jalal struggled since 1966, we have to follow him. [The parties] have money and privileges. When you take these two things from them, they are nothing.” (OTR)! • “Loyalty is still not with Kurdistan, but with KDP and PUK. On the front lines there is a negative effect because of that. If that didn’t exist, we would defend better than before”. (i25)! • “The simple people, for him, he has been in this party and he is proud of his party. People think about the interests of the Kurds and Kurdistan before anything else. And they think that through the [party] they can achieve that.” (i8)! 54“The president is the commander of the Peshmerga. The fighting is not under the MoP.” (i18) “Right now we are not under attack. When there is no attack, we get support from the MoP. Otherwise we get from the President.” (i25) "39  • Is everyone part of a party? “Not everyone but most. We do not find a lot of people independent because the persons do not find work. Everything is behind the parties. If you look for work they demand from you to show the paper from a party.” (OTR) ! • Personalized System:! • “The problem is people are very passionate about politics and that passion is usually about a political party and about an individual rather than a political programme.” (i19)! • “The entire culture is based on the person; the feeling of entitlement, loyalty, the cult and prestige that comes from followers. Even democracy here works on a tribal basis.” (OTR)! • ”The power is in the hands, in one persons hands.” (i23)! • “There is not one KDP and one PUK either. They have internal disputes and personalities.” (OTR) ! • “In a region where it is not institutionalized, people rely on tribes and personalities, particularly in a region with so much instability. I have no problem to have a George Washington of Kurdistan, as long as […] it is not some guy who yields absolute power.” (i17)! • “For example Nechervan Barzani. He owns a TV station, a university, oil companies, hotels, and many other businesses. Even when Nechervan Barzani was not the Prime Minister it was still him who went to Turkey to conclude big business deals with the officials.” (OTR)! • Civil-Military-Gap:! • “All military guys are political figures as well. According to the constitution you have to be retired, but that does not mean you lose your influence or authority. All you need to do is pick up your phone and say ‘can you do this?’” (OTR)! • “If commanders on the frontline are also members of the politburo that is okay because they are part of making the government. All of us is the party, and we are inside the government. Our party is in power.” (i6)! • “Just because you are an old Peshmerga commander, it should not mean and entitle you to be commander on every front line.” (i21)! • Example Barham Salah: “He can not control one single thing in the PUK, because he does not have a Peshmerga background. Others they can talk, function, with all the respect, no matter whether they are qualified or not. But they used to be a great leader and therefore they must lead until the day they die. While Barham, if he were anywhere else, he would be in the top position right now.” (i23)! • “The role of the Assayish is to keep the internal security. But they are very well trained. And during the war now they became like a rapid reaction military force.” (i11)! ! As asserted above, the choice of focusing on the elite does not just derive from the internal hierarchical structure of the party and the military institution, but also from the fact that hierarchy and loyalty are part of the wider culture and are hence found throughout all aspects of society. "40 Particularly the concept of loyalty helps to highlight a relevant aspect: “Just because [the forces are] structurally unified, [it] does not mean that loyalty runs with [the] ministry.” (i5) Loyalty influences the decision, when a and b disagree, which one will be followed.55 The institution or the party. What can be read from the findings is that loyalty largely depends on benefits, particularly social (prestige) and economical (income, advancement etc.) ones. The findings show there is a continuing strong reliance on parties, tribes, and personalities. Particularly the last aspect plays a vital role in understanding the informal system of the civil-military relation in the region. The findings in section “Personalized System” highlight this relevance. ! The person matters (more than the office he/she holds)56. Thereby it is neither about personality or charisma, nor about the government position. It is about the social position and the access to power in their immediate area. The system is comparable to a localized structure, where towns are small, people know each other, and everyone knows who is in charge - it does not matter whether that person heads an office or not. Personalities are based on a network and a sense of localism. At the same time these local personalities operate under the wider arch of centralized party-loyalty and -command57. This system is best seen in the example of the PUK, where, ever since Jalal Talabani proved physically incapable of running the party, the structure started to fall back to its local components while still remaining under the party umbrella58. Less obvious though still present, a similar set-up is generally visible in all parties and even more so in the Peshmerga. With military leaders, their power derives largely from the access to and the distribution of money59. However, the personal contact and relationship to the general is equally important to the question of obedience and loyalty. The individual commander influences the structure on the basis of salary, friendship ties, and loyalties. A conclusion can be drawn that these networks of personal contact are relevant to consider. This system of networks, I argue, is best described in the terminology of centralized localism60. ! Having identified loyalty and personality to be relevant, one starts to identify a certain hierarchy in the relations and the intersecting layers of loyalty. As discussed with friends as well as with research subjects, different loyalties are defined to be central to the Kurdish society: the family, the town, the tribe, the party, the nation. Depending on the person and the situation, different 55 Following Desch’s (1999) credo of CMR to find who people follow when the leaders disagree, I found a level of pragmatism between the generals, saying that “this is a war, we have no time for party politics” (i24/ i22/ i13/ i25), but nevertheless the individual commander or politician influences the structure on the basis of likes and dislikes, friendship ties, and (party) loyalties. 56An example is the fact that not just every party maintains its own media channel, but almost every important “fraction” (with one leader each) has their own media outlet. 57 KDP is more centralized than PUK. 58As I was told: “Today you don’t know who to talk to when you want to contact the PUK. Most of the time you just end up with the PUK official who has most influence in the area you are at.” (OTR) 59 See money distribution discussion above. 60 With localism referring both to the local territory and the personalized focus of power in the location. "41 identifications, loyalties, and roles are dynamically adapted. This primary impression is further enforced when addressing the different forms of civil-military gaps identified in Rahbek- Clemmensen et.al. (2012). Having already asserted the existence of an institutional gap, the focus here lies on the cultural-, policy-preference-, demographic-, and legitimacy gap.! As already shown in the prime-example of the political and military figure Sheikh Jafar Mustafa, the impression develops that there is no such thing as a distinguishable “gap” between a civilian and a military sphere. Due to the strong interconnection and almost impossible separation between military and civilian influence in the party-structured system, and combining this idea with the concept of centralized localism, power can be asserted to lie with the person and not necessarily the uniform or the office61. Therefore any cultural, demographic, and policy-preference gap between the conceptual spheres is dismissed. Also, the gap of legitimacy is not present in the Kurdish case since every influential politician has a dominant Peshmerga background and the absence of the same can even be harmful to a career (example Barham Salah). ! Altering this insight by combining the low level of institutionalization and the strong personalization of the system, there seems to by a dynamic interrelation between the different public and private roles a person incorporates. For example, just as many Peshmerga also work as taxi-drivers or small business owners parallel to their military duties, the institution of internal security, Assayish, usually belongs to the Ministry of Interior, except when operating on the front lines as they then follow the commander of the region62. Roles and identifications are therefore not rigid and marked by clear distinctions or “corporate”-identities. They are instead multi-layered and highlighted according to the social situation. It is in the moments when people need to chose sides that these layers become apparent.! These assertions are also confirmed by the definition of goals presented above: In the unification of the Peshmerga, the terms “institutionalization” and “professionalism” are used regularly. Thereby, institutionalism refers to people following the office and not the person, therefore hinting towards the currently persisting personalized system. And professionalism implies that people take over a public role that is different from the personal self. Given that professionalism is also identified to be a goal, and the absence of it is named to be a problem, the conclusion towards a very dynamic and not clear cut differentiation between the private and public and also between the different public roles currently exists. Both of these conclusions are vital to the understanding of the civil-military relations of the Kurdish region: While there might be an institutional form of a gap (structure), due to the weakness of these institutions and the parallel strength of a personalized system in addition to the almost non existent differentiation of roles, no clear gap between the civilian and the military can be assumed. ! 61 It depends on the individual whether a gap is perceived, professional standards are valued, and party or institutions are prioritized. 62 As Sheikh Jafar (i18) asserts: “If there is a Assayish, Police, Anti terror, anybody on my [front] they will follow what I say”. "42 Concluding Statements! Combining all the findings, there are several points that need to be concluded for the Kurdish case: There exists an official structure and parallel to it the system of party-domination is still the influencing factor in the region. Each of the parties maintains control over territory, resources, and to large extent their people, due to the system of clientalism, patronage, and loyalty.  Money is identified to be the key variable in this relationship. Mistrust of having the other party taking away this power for the benefit of a hegemonic domination is identified to be the variable keeping their followers apart and the military forces loyal. Therefore, despite an official institutional “structure” existing, in order to understand the actual civil-military relations one needs to look outside the “office”, rank, and uniform. The findings suggest there to be a continuing reliance on parties and personalities. Thereby it is not necessarily about charisma, rank, or position. It is about the social position the person holds, which derives from party-rank or family-membership, as this in turn grants access to resources that can be distributed to clients. Since this money is mostly located in the parties and only slowly transferred through the institutions, local influencers relate to a party patron. The system, therefore, is one of centralized localism and needs to be understood in its territorial set-up. For the analysis of civil-military relations these networks of personal contact are relevant to consider as they trump the official institutional distinction of civilian and military spheres. ! Combining this knowledge with the assessment of there being very dynamic and not clear cut differentiations between the private and public, as well as between the different public roles, the civil-military relations of the Kurdish region factually show no clear gap between the spheres. While the gap exists in institutional terms, the institutions do not yet yield actual decision-making influence in the region, which therefore renders the division marginal. Culturally a distinction of the spheres is absent, as is the general differentiation between public office and private person. The private person, their family membership, and relations to others builds the core of the system of power in the region. These networks of relations strongly influence all civilian and military aspects of society. Current CMR theory is not capable of conceptualizing a system without a gap, that is instead based on personal networks which run parallel to the official structure theory tends to analyze. It is the intent of this thesis to find alternative concepts that can substitute this explanatory shortcoming.! ! Revising Theory! ! Having discussed the aspects of civil-military relations in the Kurdish case that CMR theory is not able to account for, this section aims to identify alternative concepts that help substitute for these deficiencies. Three shortcomings stand out: One, there is no tangible gap between civilian and military spheres, but instead there is a gap between an institutional structure and an unofficial system, with the later providing the explanatory variable to the relationship. By focusing on the "43 relations between two institutions, CMR theory does not account for decision-making power residing outside the structure. Two, this system is characterized by party-based division and constituted by personal networks based on a set-up of centralized localism. Assuming a state (monopoly of violence) and defined institutions, current theory fails to comprehend pluralistic and intersecting relationships between several actors. And three, the networks’ territorial division aids the conceptualization of the intended military unification in a Prisoner’s Dilemma perspective, which CMR theory is also not able to incorporate.! In this chapter, sticking to the order of the issues presented, I introduce concepts that, with some alterations, can help explain the KRG case and revise prevalent theory.! Starting with the first, it has to be noted that it is not so much the not-tangible-existence of the gap that is the source of revision, but it is the way in which the non-existence of the gap is contrasted by instead identifying a gap between structure and system. Therefore, the first aspect provides arguments to the nature of the structure-system-division and its consequences for CMR theory. ! The idea of formal and informal aspects in civil-military relations has been introduced in Latin American cases by Pion-Berlin (2010)63. He finds that in some cases informal venues are chosen to get around formal civil-military relations. While not challenging the premise of the institutional structure of civil-military spheres, he concludes his analysis by asking if “informal routines consistently substitute for the formal ones, [whether] they should constitute the dominant focal points for the analysis of civil– military interactions” (Pion-Berlin 2010: 540). For the Kurdish case, this question can be confirmed. Not only are informal routines substituting for the formal ones, I argue that they constitute the center of the relationship (see “Structure”).! Casting a closer look at Pion-Berlin’s definition of formal relations, it becomes obvious that the latter are irrelevant in the study of the civil-military relations of the Kurdish region. As he characteristically says:! ! Formally “contact between government officials and military commanders [..] takes place ! within the defense organizations in and around the chain of command. It is along this ! ! organizational ladder of influence that political overseers and soldiers interface on a daily ! basis, within well-established agencies using official lines of communication, conforming to ! official procedures. […] Formal behavior is officially authorized and legally sanctioned. […] ! [In] [i]nstitutionalized patterns of interaction […] actors expect their behaviors to be guided ! by the same rules [in] the future.” (Pion-Berlin 2010: 528 ff.) ! Taking the current conflict as an example, the front line with ISIS is divided into eight sections, four of which are dominated by the leading party according to the territorial divide. All eight military figures are appointed by the parties, not the MoP. Therefore, the findings suggest that institutions are weak and their formal acceptance of this by the MoP thereby resembles a “technicality of 63While there is others that discuss informal institutions, there is, to the author’s knowledge no one else who applies informality to civil-military relations specifically. "44 stamping party-interests”. Several other instances of party-appropriation of the institutional infrastructure indicate similar distanciation between structure and system or form and content to use another metaphor.! Another aspect that Pion-Berlin’s definition highlights is the ability of institutions to regulate behavior in a manner that makes actions predictable. This predictability is considered a core stabilizing factor that decreases mistrust and at the same time aligns with the expressed goals of institutionalization and professionalization of the region.! Informal behaviors, in contrast, are ! ! “those that do not follow the official script (are not mandated by law), that are not always ! ! situated in official venues (along the chain of command), and that depart from statutory ! ! rules of conduct (official procedures). […Thereby] [s]ome informalities are highly patterned, ! even rule bound, but not enforced [while other] non-institutionalized informalities are not ! ! regularized; they are capricious, ad hoc, and contingent in nature.” (ibid.: 529) ! In his analysis Pion-Berlin concludes that ad hoc informal behaviors are more commonly related to the unofficial relations observed in civil–military relations (ibid). The Kurdish case, however, shows almost institutionalized patterns of informal behaviors and networks, as the parties still yield the ultimate decision-making power in the region. Influence does not derive from the “office”, but from who you are, which family you belong to, or who you are personally close to. Therefore, concluding the informal paths to be stronger and more common than ad-hoc occurrences, the system of the region is comparable to the notion of institutionalized informalities introduced by Helmke and Levitsky (2006) as “socially shared rules, usually unwritten, that are created, communicated, and enforced outside officially sanctioned channels.” (Helmke, Levitsky 2006: 5). Both Helmke, Levitsky and Pion-Berlin, thereby, understand informalities to be either complementary, substitutive, or competitive to the formal arrangements. For the Kurdish case, I argue that one needs to invert the understanding; it is “the system” of informalities that is currently complemented by formal institutions (the structure).! Summing up, the integration of the system into CMR theory constitutes an essential corrective to its prevalent and misleading structure emphasis.! This informal system observed in the Kurdish case is characterized by several variables. One of the most pertinent, but more importantly largely overlooked, aspect is the influence of the persisting territorial division of the parties. Support towards the inclination of ascribing a relevance to territory can be found in the theory of Zukerman-Daly (2012), who, referring to the case of Colombia, bridges the geographical distribution of social networks with the idea of demilitarizing militias. The theory of of demilitarization notwithstanding, her argument is still valid in the Kurdish case. Having two geographically split administrative systems in the KRG that each correlate with one dominating party and their social network structure (Stansfield 2003: 5), Zukerman-Daly’s theory needs some alterations to be applied to the KRG but is still a necessary inclusion to the "45 thesis. The value of her perspective derives from her focus on organizations (meso-level) more than on the state (macro-level) or individuals (micro-level). When analyzing the relationships of the KDP and the PUK to their respective military force and to each other, an organizational perspective is invaluable. Within these organizations, Zukerman-Daly places an importance on informal networks and their geographic distributions (both being factors that have been identified as crucial for the KR-I) demonstrating how the one influences the other by showing that “organizations whose combatants enjoy multifaceted bonds and remain geographically concentrated after disarming [..] retain a structure and disciplined collective action” (Zukerman-Daly 2012: 5). Despite Zukerman- Daly’s application of the network-geography nexus on the issue of demobilization, her findings lead to support for the conclusion of this research that the geographic mirror of the political polarization influences the social network structure by making in more resistant to change. Concluding from this perspective, the Kurdish region is likely to maintain divided even within the context of increased unification, as the cluster of personal networks is concentrated around the geographic distribution of the parties. It is this spatialization of political power in a polarized system that leaves its imprint on the civil-military relations.! This brings us back to the Prisoner’s Dilemma model, which is reinforced by the grafting of the political division onto territory: conceptualizing the unification of the two administrations and their respective security forces not just as a “(state-)internal” problem of accommodating two elites but as a merging of two units under the auspices of an almost anarchic system, theories of international relations can be applied. Therein lies the heuristic value of the usage of the Prisoner’s Dilemma model, in a “de-facto state” context. There are two territorially split units, which both mistrust the other due to prior events, in a stage of quasi-anarchy, as the state Iraq has limited authority over the region and the shared institutions are still weak, that intend on increasing their cooperation by unifying their armed forces. Caught in an infinite game scenario, common agreement and according action leads to an increased stabilization and trust as the behavior of the mistrusting partners becomes more predictable. Disregard, circumvention, or shirking from one side, may potentially increase the payoff to the shirking party, but is likely to hurt the rule-abiding party and thereby reinforces the already existing mistrust.! Currently, the common agreements are found in an informal setting, mostly located outside the formal structure of institutions and offices. Thereby, any agreement relocating the military forces towards the Ministry of Peshmerga not only puts each party in a vulnerable position by decreasing the own military force, it also increases the power of the institution MoP which is currently held by Gorran. While for the Prisoner’s Dilemma, the assumption of a neutral arbiter to bridge the dilemma is a valid solution, the arbiter in this example is equally political. In a situation where only one out of four alternate options leads to a decrease in mistrust, it is most likely that the parties are going to maintain at least part of their forces outside of the official institutions, as well as maintain an informal relation to some loyal forces inside the Unified Peshmerga Forces (UPF). And it is this "46 conclusion which in turn effects the analysis of the civil-military relations of the region, as it allows to predict for a continuation of the relevance of informal relations.! With all different aspects hinting towards a likely continuation of the informal system, and CMR theory being unable to account for the same, I argue that an increased attention needs to be paid towards developing toolkits and methodologies to help theory account for these complex cases. Abandoning preliminary assumptions and normative postulations of prevalent theory has to be a first step in this process.! ! Conclusion! ! This research has to be understood at the unique point in time it is undertaken. Having been autonomous within the northern parts of Iraq ever since the establishment of the UN-enforced no- fly-zone, and having an official territory of government based on the constitution of Iraq 2005/06 in addition to maintained spheres of influence outside these territories due to the lack of capacity of the Iraqi state, the Kurdish region is currently faced with a new threat: ISIS captured the city of Mosul on June 9th, 2014. On August 17th, 2014 ISIS advanced towards the Kurdish territories, penetrating an area close to the Kurdish capital Erbil. And it was on August 25th that my field work commenced. The situation of war influenced the research in several ways. One, the war made Kurdish officials realize the need of unification. This intensified attention to the subject triggered changes that might likely have taken more political bargain at another time. In addition, the same realization opened doors as people realized the relevance of my research. And two, the war allowed for conducting observations on the front lines which provided insight into how forces actually work on the ground.! To sum up, the presented thesis understands itself as a preparatory project for the analysis of the military transformation process in the Kurdish region of Iraq (KR-I). An assessment of the existing civil-military relations is considered (CMR) an invaluable starting point for any observation of transformation. By relating the political and military division of the region, however, limits of the explanatory power of prevalent theory are reached. Identifying these theoretical shortcomings along the case study is consequently the primary ambition. Thereby, the aim is to accomplish an added value for both a further understanding of the Kurdish region as well as a criticism and attempted alteration of CMR theory.! Starting from the analyses by Stansfield (2003) and Chapman (2011), a lack of connecting analysis between the political and military sphere as well as a disregard for the quasi-anarchy in which the military unification of the party-based Peshmerga occurs is observed. Arguing for the invaluable comprehension gained from including these aspects, the hypotheses of the research state the existence of quasi-anarchy and the need for applying the model of Prisoner’s Dilemma to alter the "47 understanding of the civil-military relations of the region, which can further be used to support the revision of CMR theory.! To address these hypotheses and the resulting research question, I first provide two preparatory chapters, and then show how the civil-military relations in the region were accessed by employing the method of qualitative interviews. Thereby, I understand my work not to reflect a single picture of reality, but I identify patterns out of the various pieces of information like a puzzle. In Findings and Analysis I reveal the puzzle built during my field work: There is an institutional gap between a civilian and a military sphere in the region. However, due to the weakness of the institutions, a differentiation of this institutional structure with the actual working mechanisms of the system needs to be undertaken. The core aspects of the system is a centralized (party) localism (person and territory) that is structured by hierarchy, financial dependency, and dynamic, intersecting loyalties. Two of these systems exist parallel to the structure and to each other, due to the ongoing division of the parties. Historic reasons are the basis for deep mistrust between the two sides and foreign interference, corruption, and nepotism further enhance this tension. Due to these factors and an inherently insecure environment of quasi-anarchy, the maintenance of influence through the informal system is just as vital as the upkeep of the party-loyal Peshmerga forces for the dominant parties. ! In a last step, these findings are altered by the Prisoner’s Dilemma model and concepts relating the distribution of social, informal networks onto the geographic division to assert the likelihood of their continued maintenance. Thereby, I assert that CMR theory is currently not able to explain these complex cases due to its persistent subsistence of preliminary assumptions and normative postulations: It is not able to provide explanatory power to a system in which personal networks and relations explain the political and military set-up, because it remains its focus of analysis on the institutions even when they do not have actual decision-making powers. It is not capable of conceptualizing a system in transition because it assumes a powerful state, distinct institutions, and disregards sub-state actors. And it is not able to conceptualize multiple, interrelating relationships as the state-centric perspective assumes a monopoly of power, force, and legitimacy and thereby fails to account for the possibility of several actors being involved.! With that big a gulf of explanatory shortcomings, future research is challenged to turn its analysis away from fixed templates of models and assumptions towards an opening up for blurred lines, multiple intersecting relations, and dynamic role-identities. After all, in an increasingly unstable world for both geopolitical zones and theoretical dichotomies, the Kurdish region is likely to not be the only case that requires a more refined toolkit and methodology for understanding complex relations between institutions and actual power yielding forces.! ! ! ! "48 Bibliography! ! Ahmed, Mohammed M.A. Iraqi Kurds and Nation-Building. London: Palgrave Macmillan, 2012.! Anderson, Liam, and Gareth Stansfield. The Future of Iraq - Dictatorship, Democracy, or Division?. New York: Palgrave Macmillan, 2004.! Andreoni, James, and John H. Miller. “Rational Cooperation in the finitely repeated Prisoner’s Dilemma: Experimental Evidence.” The Economic Journal 108, (May 1993): 570-585. ! Ångström, Jan. “The changing norms of civil and military and civil-military relations theory.” Small Wars & Insurgencies, 24:2 (2013): 224-236. ! Albright, David E. “Comparative Conceptualization of Civil-Military Relations.” World Politics, vol. 32. no. 4 (Jul 1980): 553-576. ! Axelrod, Robert. “Effective Choice in the Prisoner’s Dilemma.” The Journal of Conflict Resolution, vol. 24, no. 1 (1980): 3-25. ! Bengio, Ofra. “Nation Building in Multiethnic Societies: The Case of Iraq,” in Minorities and the State in the Arab World, edited by Ofra Bengio and Gabriel Ben-Dor Lynne, Part 3.8, Rienner Publishers Inc (e-res), 1999.! Bengio, Ofra. “Building Regional Peace: Iraq and the Kurds.” Middle East Review of International Affairs, vol. 13, no.3 (2009): 50-57. ! Burk, James. “Theories of Democratic Civil-Military Relations.” Armed Forces & Society, vol. 29., no. 1 (Fall 2002): 7-29.! Chapman, COL Dennis P. Security Forces of the Kurdistan Regional Government. Costa Mesa, California: Mazda Publishers, 2011.! Chuter, David. “Civil-Military Relations: Is there really a problem?” Journal of Security Sector Management, vol. 7, no. 2 (Nov 2009): 1-17.! Cohen, Eliot A. “Are U.S. Forces Overstreched? Civil-Military Relations.” Orbis (Spring 1997): 177-186.! Cohn, Lindsay. “The Evolution of the Civil-Military ‘Gap’ Debate”. Duke University. TISS project on the Gap Between the Military and Civilian Society, 1999.! Collins, Joseph J. and Ole R. Holsti. “Civil-Military Relations: How Wide Is the Gap?” International Security, vol. 24, no. 2 (Fall 1999): 199-207.! Davis, Diane E. Irregular Armed Forces and their Role in Politics and State Formation. Cambridge: Cambridge University Press, 2003.! Desch, Michael C. Civilian Control of the Military: The Changing Security Environment. Baltimore: Johns Hopkins University Press, 1999.! Ellison, Glenn. “Cooperation in the Prisoner’s Dilemma with Anonymous Random matching.” Harvard University Review of Economic Studies 61, (1994): 567-588.! Feaver, Peter D. “Civil-Military Relations.” Annual Reviews Political Science 2, (1999): 211-241.! Finer, Samuel E. The man on Horseback - the Role of the Military in Politics. London: Transaction Publishers, 2006.! "49 Gruber, Verena. “Between Influence and Independence - conceptualizing the Kurdish Regional Government (KRG) of Iraq”. Center of Middle Eastern Studies, MA Course paper: Introduction into Middle Eastern Debates, Course leader: Umut Özkirimli, Lund University, 2013.! Gruber, Verena. “Accessing Kurdish Geo-Strategy in Iraq - An analysis of the consequences of the post- Mosul military movements by the Peshmerga on the status of the Kurdish Regional Government in Iraq.” King’s College Summer School, Course paper: Making Military Strategy and Defence Policy, Summer 2014, Course leader: Michael Codner, London, 2014.! Helmke, Gretchen and Steven Levitsky, Eds. Informal Institutions and Democracy: Lessons from Latin America. Baltimore: Johns Hopkins University Press, 2006.! Holland, Ben. “Are Kurdistan’s oil contracts constitutional?”. Energy in the Middle East, 28-29, URL: www.petroleum-economist.com, accessed April 5, 2012.! Huntington, Samuel P. The Soldier and the State - the Theory and Politics of Civil-Military Relations. Cambridge, Massachusetts: The Belknap Press of Harvard University Press, 1985.! Iraqi Constitution 2005/06. URL: http://www.iraqinationality.gov.iq/attach/iraqi_constitution.pdf, accessed January 13, 2013.! Janowitz, Morris. The Professional Soldier: A Social and Political Portrait. Glencoe, Illinois: Free Press, 1960.! Kaldor, Mary. “In Defence of New Wars.” Stability 2(1): 4, (2013): 1-16. ! King, Diane E. Kurdistan on the Global Stage - Kinship, Land, and Community in Iraq. New Jersey: Rutgers University Press, 2013.! KRG Parliament. URL: http://cabinet.gov.krd/p/page.aspx?l=12&s=030000&r=317&p=229&h=1, accessed December 23, 2014..! Licklider, Roy. “Merging Militaries after Civil War: South Africa, Bosnia and a preliminary search for Theory”. Paper delivered at the 2008 Annual Meeting of the American Political Science Association, August 28-31, 2008.! Licklider, Roy Ed. New Armies from Old - Merging competing Military Forces after Civil Wars. Washington: Georgetown University Press, 2014.! Lortz, Michael G. "Willing to Face Death: A History of Kurdish Military Forces - the Peshmerga - From the Ottoman Empire to Present-Day Iraq”. Electronic Theses, Treatises and Dissertations. Paper 1038, 2005.! Luckham, A.R. “A comparative Typology of Civil Military Relations.” Government and Opposition, vol. 6(1), (1971): 5-35! Mahmoud, Nawzad. “Barzani Orders Peshmerga Forces Reformed, United.” Rudaw 25/08/2014, URL: http:// rudaw.net/english/kurdistan/250820142, accessed August 26, 2014.! McDowall, David. A Modern History of the Kurds. London: I.B. Tauris, 2007.! Michael, Kobi (Jacob). “The Dilemma behind the Classical Dilemma of Civil-Military Relations.” Armed Forces & Society, vol. 33, no. 4 (July 2007): 518-546. ! Morgenthau, Hans J. Politics Among Nations - The Struggle for Power and Peace. New York: McGraw-Hill Companies Inc, 2006.! Natali, Denise. “The Spoils of Peace in Iraqi Kurdistan.” Third World Quarterly, vol. 28, no. 6 (2007): 1111-1129.! Natali, Denise. “The Politics of Kurdish Crude.” Middle East Policy, vol. XIX, no. 1 (Spring 2012): 110-118.! "50 Nowak, Martin and Karl Sigmund. “A strategy of win-stay, lose-shift that outperforms tit-for-tat in the Prisoner’s Dilemma Game.” Nature, vol. 364 (1993): 56- 58.! Olson, Robert. The Goat and the Butcher: Nationalism and State Formation in Kurdistan-Iraq since the Iraqi War. Costa Mesa, California: Mazda Publishers, 2005.! Perlmutter, Amos and William M. LeoGrande. “The Party in Uniform: Toward a Theory of Civil-Military Relations in Communist Political Systems.” The American Political Science Review, vol. 76, no. 4 (December 1982): 778-789.! Pion-Berlin, David. “Informal Civil-Military Relations in Latin America. Why Politicians and Soldiers Choose Unofficial Venues.” Armed Forces & Society 36(3), (2010): 536-544.! Rahbek-Clemmensen, Jon and others. “Conceptualizing the Civil-Military Gap: A Research Note.” Armed Forces & Society 38(4), (2012): 669-678.! Schiff, Rebecca L. “Civil-Military Relations Reconsidered: A Theory of Concordance.” Armed Forces & Society, vol. 22, no. 1 (1995): 7-24.! Schiff, Rebecca L. “Concordance Theory: A Response to Recent Criticism.” Armed Forces & Society, vol. 23, no. 2 (Winter 1996): 277-283.! Schiff, Rebecca L. “Concordance Theory in America: The Post-Revolutionary Period (A Study in Civil-Military Relations).” APSA Conference Paper, Chicago, 2004.! Stansfield, Gareth. Iraqi Kurdistan - Political development and emergent democracy. London: Routledge Curzon, 2003.! Stansfield, Gareth. “The unravelling of the post-First World War state system? The Kurdistan Region of Iraq and the transformation of the Middle East.” International Affairs 89: 2, (2013): 259-282. ! Stepan, Alfred. 1971. The Military in Politics: Changing Patterns in Brazil. Princeton, NJ: Princeton Univ. Press.! Stepan Alfred. 1988. Rethinking Military Politics: Brazil and the Southern Cone. Princeton, NJ: Princeton Univ. Press.! van Bruinessen, Martin. “The Kurdish Question: Whose Question, Whose Answers? The Kurdish Movement Seen by the Kurds and by their Neighbors”. Wadie Jwaideh Memorial Lecture in Arabic and Islamic Studies, Department of Near Eastern Languages and Cultures & Middle Eastern and Islamic Studies Program, Indiana University, Bloomington, 19 November 2004.! Wimmer, Hannes. “The State’s Monopoly on Legitimate Violence. Violence in History and in contemporary World Society as Challenges to the State”. Conference ‘Transformations of Statehood from a European Perspective’ organized by the Austrian Academy of Science, Vienna, January 23-25, 2003. ! Zedalis, Rex J. “The role provincial government units can play regarding oil and gas development agreements in the Kurdish North: allocatin of Iraqi constitutional power” Journal of World Energy Law and Business vol. 6, no. 4 (2013): 314-347.! Zukerman-Daly, Sarah. “The Geography of Social Networks, Balance of Power, and the Durability of Post- War Peace.” working paper, Stanford University (e-res), 2012.! Zukerman-Daly, Sarah. 2014. “Dark Side of Power-Sharing - Middle Managers and Civil War Recurrence”. Comparative Politics April (2014): 333-353.! ! "51 Appendix! ! Interviews! i1: Abdulla, Khabat. Interviewed by author. Qualitative Interview. Erbil, 31/08/14.! i2: Ahmed, Muhammad Ali. Interviewed by author. Qualitative Interview. Sulaymaniyah, 10/11/14.! i3: Ali, Helgurd Hikmet Mela. Interviewed by author. Qualitative Interview. Erbil, 04/12/14. ! i4: Asuad, Nahro. Interviewed by author. Qualitative Interview. Mula Abdula, 19/09/14.! i5: Atrushi, Farhad Ameen. Interviewed by author. Qualitative Interview. Duhok, 25/11/14.! i6: Aziz, Qadir Hama Jan. Interviewed by author. Qualitative Interview. Sulaymaniyah, 02/10/14.! i7: Bakir, Falah Mustafa. Interviewed by author. Qualitative Interview. Erbil, 15/11/14.! i8: Bakir, Falah Mustafa. Interviewed by author. Qualitative Interview. Erbil, 01/12/14.! i9: Draey, Babakr. Interviewed by author. Qualitative Interview. Sulaymaniyah, 10/11/14.! i10: Hama, Rizgar Hajy. Interviewed by author. Qualitative Interview. Erbil, 16/09/14.! i11: Hawrami, Hemin. Interviewed by author. Qualitative Interview. Erbil, 01/12/14.! i12: Ismail, Hajar. Interviewed by author. Qualitative Interview. Erbil, 15/09/14.! i13: Jawad, Said. Interviewed by author. Qualitative Interview. Bashiqa Frontline, 17/09/14.! i14: Dulmany, Salah. Interviewed by author. Qualitative Interview. Jalawla Frontline, 26/09/14.! i15: Mirani, Salah. Interviewed by author. Qualitative Interview. Zumar, 29/10/14.! i16: Mohammad, Yusuf. Interviewed by author. Qualitative Interview. Erbil, 04/09/14.! i17: Mustafa, Chia Nashirwan. Interviewed by author. Qualitative Interview. Sulaymaniyah, 10/11/14.! i18: Mustafa, Sheikh Jafar. Interviewed by author. Qualitative Interview. Qaradigh, 26/09/14.! i19: Omar, Hoshyar. Interviewed by author. Qualitative Interview. Sulaymaniyah, 13/11/14.! i20: Radsh, Hama. Interviewed by author. Qualitative Interview. Kirkuk, 19/09/14.! i21: Rahim, Mohamad ‘Hama’ Tofiq. Interviewed by author. Qualitative Interview. Sulaymaniyah, 25/09/14.! i22: Rashad, Mala. Interviewed by author. Qualitative Interview. Bashiqa Frontline, 17/09/14.! i23: Resool, Shorsh (Haji). Interviewed by author. Qualitative Interview. Sulaymaniyah, 10/11/14.! i24: Salah, Said. Interviewed by author. Qualitative Interview. Talaward Frontline, 01/09/14.! i25: Salih, Najat Ali. Interviewed by author. Qualitative Interview. Makhmour, 17/11/14.! i26: Qadir, Mustafa Said. Interviewed by author. Qualitative Interview. Erbil, 12/10/14.! i27: Qadir, Mustafa Said. Interviewed by author. Qualitative Interview. Erbil, 01/12/14.! ! ! ! ! ! ! "52 Maps: the Kurdish influence vs. KRG- territory! Map taken from Olson (2005)! ! ! ! ! ! Election Results per Governorate 2013! ! The territorial division is also reflected at the governorate distribution of the votes during the elections, which portray the existing “zones of influence”: KDP is strongest in the province of Duhok (73.32%; in contrast to PUK gaining only 8.37%), PUK is strongest in Sulaymaniyah (35.94%; in contrast to KDP gaining only 9.77%) and Gorran is strongest in Sulaymaniyah as well (40.55%; thereby even dominating the classical PUK stronghold). The capital city of Erbil is divided between the three dominating parties KDP, PUK and Gorran, with KDP taking the lead of 49.31%, PUK having 22.31% and Gorran 14.31%. (KRG.org)! ! ! "53 Protocols! ! ! Field Work Protocol, August - December 2014! ! ! Military Protocol! Explain situation*! ! division? how? why? when?! ! difference KDP PUK?! This front line! ! who stationed? who command? which forces?! ! support: who refer to when needing help?! ! who immediate superior? who gives command to? where gets command from?! ! recruitment, training, education! Personal history! ! recruitment! ! ! ! ! ! Civilian Protocol! Explain situation*! ! division? how? why? when?! ! difference KDP PUK?! Peshmerga! ! past, current, future - why two, hiring, firing, effects of current conflict! Goals for the future! Problems! ! current / potential for the future! ! *Note 1: This form is just a summarized reflection of the actual question, since the actual formulation differed according to the research subject. “The situation” was always elaborated before asked for. Other, related “general questions” were, e.g.: Can you explain why the political parties are divided? Why are there two units, 70 and 80? How do you feel about unifying the forces? What do you think, how can the forces be unified? What are the potential problems in the unification of the forces? What does a “unified” system look like?! "54 Pictures! Minister of Peshmerga, Mustafa Said Qadir! Former Minister of Peshmerga, Sheikh Jafar Mustafa (and Zhyar Ahmed)! "55 Visiting the Front Lines …! …in Talaward, Kirkuk! … at Bashiqa Mountain! ! "56 … in Jalawla! … in Zumar! "57