POWER PLAYS: EXPLORING POWER AND INTERVENTION IN PROXY WARS. Ryan Justin Youra. Under the guidance of. Professor Benjamin Jensen,

Transcription

1 POWER PLAYS: EXPLORING POWER AND INTERVENTION IN PROXY WARS By Ryan Justin Youra Under the guidance of Professor Benjamin Jensen, School of International Service Submitted to the University Honors Program of American University in Partial Fulfillment of the Requirements for the Degree of Bachelor of Arts University Honors In International Studies Spring 2014 American University Washington, D.C

2 Power Plays: Exploring Power and Intervention in Proxy Wars Table of Contents Abstract... i Introduction... 1 Literature Review... 2 Proxy Warfare... 9 Dataset Methodology Data Analysis Case Study of the 1973 Arab-Israeli War Conclusion Works Cited... 39

3 Youra i Abstract Surprisingly, there is little academic study devoted to proxy wars, though proxy wars are a common occurrence throughout history. Focusing on the proxy supporter, this study examines the involvement of different types of actors in proxy wars in the Middle East to build the proxy war literature. What role does power play in proxy supporter-proxy fighter relationships? How do states different levels of power influence their interventions in local conflicts? This research explores these questions by building a dataset on proxy conflicts in the Middle East from 1945 to 2009 from the Uppsala Armed Conflict Dataset, the Non-State Actor Dataset, and open-source newspapers: The New York Times and The Washington Post. After examining several correlational relationships between the conflict type and the power, location, intensity, and variety of support of proxy supporters, this study finds that weak powers heavily participate and engage in proxy wars, particularly in intrastate and extrastate wars. Such findings stand contrary to the great power emphasis of the study the proxy war. The study then examines the 1973 Arab- Israeli War to explore possible mechanisms for supporter involvement in proxy wars and discovers that weaker powers can even play a role during the initial phases of a proxy relationship and thereby influence proxy wars much more than previously considered. Minor power states augment their power through the use of proxies. Though the Cold War dominates thought concerning proxy relationships, it is clear that minor powers largely dominate this type of war.

4 Youra 1 Introduction Surprisingly, there is little academic study devoted to proxy war. As Andrew Mumford states, proxies are historically ubiquitous and yet chronically under-analyzed. 1 This gap in the international relations literature misses a rather common feature of statecraft in the Westphalian system. 2 Without an understanding of proxy wars, an understanding of intervention is incomplete. The general perception of proxy wars is one of great power interference, as perpetuated by the cold war. Dominant thinking in international relations also confirms this power bias; if international politics, like all politics, is a struggle for power, 3 then we should expect to see the struggle in all forms of international relations both above and below the surface of overt state action. While proxies are often dismissed as the work of major powers, they reveal a whole range of actions that occur under the surface level of international relations. This study attempts to add to the body of existing proxy war literature with a focus on the actors that support proxies. What explains actors active involvement as third parties to a conflict? What role does power play in proxy supporter-proxy fighter relationships? How do states different levels of power influence their interventions in local conflicts? This study explores the correlations between power, location, intensity, and variety of proxy supporters with their involvement in different types of wars. This study specifically examines the involvement of different actors in different types of proxy wars in the Middle East. This research explores these questions by building a dataset on proxy conflicts in the Middle East from 1945 to 2009 from the Uppsala Armed Conflict Dataset, the Non-State Actor Dataset, and open-source newspapers: The New York Times and The Washington Post. The study then 1 Andrew Mumford, "Proxy Warfare and the Future of Conflict," The RUSI Journal 158, no. 2 (2013): See Geraint Hughes, My Enemy's Enemy: Proxy Warfare in International Politics (Portland, Or.: Sussex Academic Press, 2012). 3 Hans J. Morgenthau, Politics among Nations; the Struggle for Power and Peace (New York: Knopf, 1967).

5 Youra 2 examines the 1973 Arab-Israeli War to explore possible mechanisms for supporter involvement in proxy wars. The study demonstrates that weak powers heavily participate and engage in proxy wars, contrary to the great power emphasis previously designated to proxy wars. Weaker powers can even play a role during the initial phases of a proxy relationship and thereby influence proxy wars to a higher degree than previously considered. Re-conceptualizing proxy wars as tools available for any state has implications for the continued study of proxy wars and future security threats. Literature Review In order to understand the role of proxies and their supporter states, it is necessary to first examine the intervention literature. Proxy warfare is inherently a form of intervention in that it involves the actions of a third-party state in an external conflict. While a later section will expand on this rather basic definition of proxy warfare, this focus on intervention serves as a starting point. External intervention presents a challenge to the Westphalian concept of sovereignty. Under the Treaty of Westphalia in 1648, the European powers affirmed the separate jurisdiction of each state over its own territory. This understanding can even be traced back to the Treaty of Augsburg in 1555, when the absolute right of sovereigns in religious matters was confirmed. According to David Armstrong, interest, rules, and institutions maintain order, and therefore the international system is maintained by a general pattern of interaction among states and international actors. 4 Under the practice of sovereignty, states have exclusive control over their territory. This has led to the converse norm of non-intervention in state affairs. Essentially, rights 4 J. D. Armstrong, Farrell Theo, and Lambert Helene, International Law and International Relations (Cambridge [U.K.]; New York: Cambridge University Press, 2007), 5-6.

6 Youra 3 to sovereign domain belong to the state as a participant in the international system. 5 This principle has been reaffirmed in the League of Nations and later the United Nations Charter, and thus, there appears a general preference against intervention within the international system. 6 While the English school sees this imposition of order as the foundation of an international society, the Realist tradition views the continual violation of the rule as evidence of an international system less grounded by rules than a society. While this preference for sovereignty exists, it is impossible to dismiss interventions throughout the international system s history. Hans Morgenthau dates intervention back to the Greeks. 7 Hedley Bull states that coercive intervention is endemic to the international system. 8 Several of the studies reviewed here make similar claims about the pervasiveness of intervention in international affairs. There are several slightly differing definitions of intervention, though they generally agree on the violation of another state s sovereignty and the intertwined aspect of power. If sovereignty is the control of one s own domestic affairs and exclusion of others from this jurisdiction, intervention is defined as action outside of a state s jurisdiction or territory and within the sphere of another state s jurisdiction or territory. 9 As the root word implies, intervention is the act of one or more countries figuratively coming between another state and its existing territory/jurisdiction. Intervention implies that the ideal of sovereign, autonomous actors is not completely reachable Friedrich Kratochwil, "Soveriegnty as Dominium: Is There a Right of Humanitarian Intervention?," in Beyond Westphalia?: National Sovereignty and International Intervention, ed. Gene M. Lyons and Mastanduno Michael (Baltimore: Johns Hopkins Univ. Press, 1995). 6 Hedley Bull, Intervention in World Politics (New York: Clarendon Press; Oxford University Press, 1984), 3. 7 Hans J. Morgenthau, "To Intervene or Not to Intervene," Foreign Affairs 45, no. 3 (1967). 8 Bull, Intervention in World Politics. 9 Ibid., Stephen D. Krasner, Sovereignty: Organized Hypocrisy (Princeton, N.J.: Princeton University Press, 1999).

7 Youra 4 Intervention can take several different forms. It can be forcible or non-forcible, direct or indirect, and overt or clandestine. 11 All inter-state wars can be considered interventions, but states can also intervene in other wars. Bertil Dunér and Richard Little discuss third-party intervention in civil wars. 12 In this case, a state that is not a warring faction in the civil war may join in the fight for one side, either directly or indirectly. John Stuart Mill even legitimizes this form of intervention in A Few Words on Non-Intervention, stating that third parties (either multiple neighbors or a powerful neighboring country) can act as arbiters and mediators to demand an end to conflict. 13 Additionally, when there is a movement against a foreign yoke or against a native tyranny upheld by foreign arms, a counter-intervention is justified. 14 Though Mill references international law, legitimacy, and morality aspects of an international society and rules Mill also discusses the imbalance that such a foreign yoke represents. For Mill, intervention to enforce non-intervention is always rightful, always moral, if not always prudent. 15 Intervention may also focus on changing domestic affairs or changing foreign policies, though the foreign policies approach is more nuanced. In Intervention in World Politics, Bull limits intervention solely to domestic affairs, but this limits the study of intervention. 16 Outside entities, including states, international organizations, and non-state actors, may act outside of their jurisdiction in order to assist or manipulate either the formation or outcome of foreign 11 Bull, Intervention in World Politics, Bertil Dunér, Military Intervention in Civil Wars: The 1970s (New York: St. Martin's Press, 1985); Richard Little, Intervention: External Involvement in Civil Wars (Totowa, N.J.: Rowman and Littlefield, 1975). 13 John Stuart Mill, "A Few Words on Non-Intervention," in International Relations in Political Thought, ed. Chris Brown, Terry Nardin, and N. J. Rennger (Cambridge: Cambridge University Press, 2002 (1859)). 14 Ibid., Ibid. 16 Bull, Intervention in World Politics.

8 Youra 5 policy decisions. 17 While any attempt to affect the domestic affairs of a country is considered intervention, attempts at foreign policy changes can only be considered interventions if actions change the domestic condition. Changing the international conditions without an intention to change the domestic condition of a state is not intervention. This may have the perceived effect of an intervention, but the action did not specifically target another state or its foreign policy outcomes. For example, if a country manipulates its currency exchange rate on the international market in order to increase exports, its action may decrease sales of a similar commodity produced by a different country. While this has the effect of impacting a country s economy, the action was not intended or directed as such and therefore cannot be considered interventionary. In the increasingly globalized world with increasingly intermestic issues, policy decisions by one state may appear as interventions in another state s affairs. At the same time, these actions are not interventions since they are not intended as such. Intention and targeting plays a role in intervention. Intervention exists in dyadic relationships, with an intervener party and an intervened party. Though a state s intention may never be fully discovered, actions directed specifically at one country without reversal may be considered intentional and therefore interventionary. For Bull, it was simpler to declare that all interventions must remain limited to domestic affairs in order to capture intervention through the war, but this misses the possibility for states to aid, assist, and affect the foreign relations between other states. The act of intervening breaks down into many types of classification. While Bull categorizes interventions in a general sense based on role of power, 18 Mumford classifies each specific form of intervention, ranging from the addition of manpower to financial assistance It is important to note that non-state actors do not have defined jurisdiction in the international system, although some may have granted jurisdiction on the domestic level, such as Hezbollah s control of South Lebanon. 18 Bull, Intervention in World Politics. 19 Andrew Mumford, Proxy Warfare (Cambridge, UK: Polity, 2013).

9 Youra 6 Mumford bases his classification on the actions taken by the intervener. Using a similar approach, Dunér creates a hierarchy of interventionary techniques to capture the instruments and levels of involvement in interventions. 20 The lowest level is direct supporting activities such as threating of military intervention, transporting external forces, or providing safe haven. The next level is indirect para-combat involvement, including military training, armed blockades, and financial support. The level above that is direct para-combat involvement, including advising and supply arms. Above that is indirect combat involvement such as irregular invasion and shelling, and finally there is direct combat involvement, which captures regular invasion and special operations. Using this hierarchy, Dunér is able to classify intervention acts based on high levels (direct and indirect combat involvement), middle levels (direct and indirect para-combat involvement) and low levels (direct supporting activities). This paper intends to use a modified version of Dunér s intervention classification as well as his hierarchy to examine the proxy warfare subset of intervention. Beyond the general range of actions, definitions of intervention focus on the role of power. There is an asymmetrical distribution of power between the intervener and the intervened. 21 The intervener is able to subvert another s sovereignty with its overwhelming resources. This power dynamic is clear when the intervener initiates a conflict, such as Nazi Germany s invasion of Poland in This is a paradox in the international system: the practice of intervention in the face of a system organized by sovereignty. While this paper does not delve into the tension from these two concepts, it is important to recognize the existence of these competing concepts. 20 Dunér, Military Intervention in Civil Wars: The 1970s, Bull, Intervention in World Politics; Dunér, Military Intervention in Civil Wars: The 1970s; Krasner, Sovereignty: Organized Hypocrisy; Morgenthau, "To Intervene or Not to Intervene."

10 Youra 7 At the same time, the extent to which the intervener affects sovereignty is less clear when intervening in an existing conflict. There are two theories to explain such interventions. The first theory focuses on the intervener and states that outside powers have a propensity to intervene. 22 Interveners are pushed to intervene by their general power politics. 23 This is consistent with the Realist tradition and realpolitik. Intervention is another form of coercive politics in order to exercise power and make other states change behavior. As formulated by Morgenthau, these interventions are strictly in the interests of the intervener and against the will of the intervened. 24 Power politics are constantly at play in this formulation of intervention since the strong exercise power over the weak. Under this push-theory, 25 strong states intervene to acquire territory; protect social groups; protect economic, diplomatic, or military interests; act in accordance with a specific ideology; and create or maintain regional power balances. 26 Strong states are continually working to gain additional power vis-à-vis one another. 27 This suggests that small states are always at risk of intervention by larger states. Kenneth Waltz states, This imbalance of power is a danger to weak states. 28 Stanley Hoffmann reinforces the claim: For a small power the balance of power could be a mixed blessing because instead of being eaten by one great power, one could be gnawed by all five or six. 29 The push-theory was not originally formulated to apply beyond states, but the focus on power relations can be expanded to discuss intervening in affairs of non-state actors. For example, Israeli intervened in southern Lebanon in 2006 with the intention to forcefully change 22 Dunér, Military Intervention in Civil Wars: The 1970s, Little, Intervention: External Involvement in Civil Wars, Morgenthau, "To Intervene or Not to Intervene." 25 Little, Intervention: External Involvement in Civil Wars. 26 Mumford, Proxy Warfare, Morgenthau, Politics among Nations; the Struggle for Power and Peace. 28 Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley Pub. Co., 1979), Stanley Hoffmann, "The Problem of Intervention," in Intervention in World Politics, ed. Hedley Bull (New York: Clarendon Press; Oxford University Press, 1984).

11 Youra 8 the behavior of Hezbollah. While Israeli action had the effect of intervening in Lebanese territory, the direct intervention was a result of a stronger entity (Israel) intervening in the affairs of a week entity (Hezbollah). This theory appears to work well when the intervener and the intervened are opposing forces, but this is not always the case. When the intervener does not violate the will of the intervened, what explains the intervention? Can it be considered an intervention? The opposite side of the intervention coin is the pull-theory of intervention, and it attempts to address those questions. 30 Richard Little traces this back to Thucydides, who detailed the ways that local rivaling parties called in either the Spartans or the Athenians to strengthen their own position. 31 Instead of great powers intervening, this theory focuses on the ways in which small states and/or non-state actors request outside intervention, thereby pulling in interveners to the conflict. Interveners respond to a stimulus, not simply a general propensity towards intervention. 32 In the pull-theory formulation, the situation must meet a pre-existing condition, namely conflict, while the pushtheory only requires a power imbalance, which inherently exists in the international system. Realism strongly emphasizes the role of strong states over weak ones, but there both Bull and Dunér discuss the role of weaker, developing states in intervention. In contrast to Hoffmann s belief that weak states become even more open to intervention with a balance of power, Bull suggests that balances of power provide weaker states with an opportunity to play the great powers off one another. 33 In this conception, the weaker state forges the path of intervention and uses superpowers for their own strategic reasons. Weak states show an ability to exert agency in the international system. 30 Little, Intervention: External Involvement in Civil Wars, Ibid. 32 Ibid.; Dunér, Military Intervention in Civil Wars: The 1970s, Bull, Intervention in World Politics.

12 Youra 9 Additionally, great powers are not the only interveners. According to Dunér s study on intervention in civil wars in the 1970s, developing states greatly involved themselves in interventions. 34 He finds, A low intervention capacity does not hold the LCDs [least developed countries] as might be easily thought on theoretical grounds. 35 As Dunér states, this is surprising given the large resource commitment required of interventionary forces. From Dunér s study, developing states are willing to commit resources to intervention based on their own strategic determinations. While individual resource commitments may not be large, the weaker states appear to have an impact as a group. In fact, the study finds that developing countries comprised 68% of all the intervening actors in the 1970s. 36 Within the scope of a decade, this is only a snapshot of interventions, but the large share of interventions by developing countries even in the midst of the Cold War potentially points to an outsized and unjustified focus on only the strong powers in interventions. As will be discussed below, the push and pull theories regarding intervention will inform the hypotheses regarding proxy wars. Before reaching that, a greater discussion of the unique features of proxy wars is required. Proxy Warfare The study of proxy war suffers from definitional ambiguity. Geraint Hughes defines proxy wars in a narrow sense. Hughes asserts, A proxy is defined as a non-state paramilitary group receiving direct assistance from an external power. 37 This means that states cannot serve as proxies. Hughes appears to grant states the power of autonomy and self-interest but not to non-state actors. While his study is incredibly thorough on the characteristics of proxies, Hughes 34 Dunér, Military Intervention in Civil Wars: The 1970s. Dunér uses the term least developed countries to connote all developing and weaker powers. 35 Ibid., Ibid. 37 Hughes, My Enemy's Enemy: Proxy Warfare in International Politics, 11.

13 Youra 10 ignores a large occurrence of proxy war namely states serving as proxies. Therefore, his conclusions are incomplete. At the same time, Hughes provides an excellent list of criteria for proxy wars. He discusses the need for a direct relationship of assistance, a common enemy for both the sponsor and the proxy, and a sustained relationship over time between the supporter and the proxy force. Of these three criteria, the most interesting is the common enemy. Absolute control would provide the supporter with the ability to choose the enemy and disregard the interests of the proxy. Instead, this refutes the notion that a proxy supporter is in full control of the proxy force. What happens in the case of divergent interests? Can proxy war exist in the presence of uncommon enemies? It seems imprudent to discount this possibility, though the assumption that both supporters and proxies have autonomous interests is certainly valid. Like Hughes, Chris Loveman defines proxy wars and interventions through a set of criteria instead of providing a definition in one sentence. This includes the provision of material aid, coordination of activities, basis in a local conflict, and a resulting conflict escalation. 38 The definition of proxy wars based on their end result, however, may be conceptually problematic. By these criteria, if indirect third party intervention does not result in conflict escalation, then it is not considered a proxy war. Another scholar, Jeffrey Bale, argues that a proxy relationship must change a proxy s objectives or target in order to be considered a proxy. 39 Essentially, proxies do not exist only with a coincidence of interests. This is problematic because such a conception of proxy intervention relies solely on push theory suggestions. Greater powers are viewed as sole parties 38 Chris Loveman, "Assessing the Phenomenon of Proxy Intervention," Conflict, Security & Development 2, no. 03 (2002): Jeffrey M. Bale, "Terrorists as State "Proxies": Separating Fact from Fiction," in Making Sense of Proxy Wars States, Surrogates & the Use of Force, ed. Michael A. Innes (Washington, D.C.: Potomac Books, 2012), 28.

14 Youra 11 directing proxies in each position. The pull theory necessitates considering the interests of the proxies. With a relationship of assistance, corresponding interests fostered organically and not through forms of power pressure still create proxy relationships. It is conceivably possible for one state to still receive support from another state while having a coincidence of interests. Dunér also attempts to define proxy wars with parameters: a compatibility of interests, support, and the exercise of asymmetrical power. He finds that only support and power are necessary components. 40 He argues that the exercise of power is central to the conception of proxy wars. A proxy must be pressured to intervene. 41 This becomes the defining characteristic of proxy relationships since without pressure, the proxy force is not a proxy and instead serves as a partner. The transfer of resources does not necessarily need to occur because a proxy may not be dependent on material support to carry out the interests of the supporter. 42 It is impossible to separate a state from the support it receives. While it is true that there are other forms of assistance that a proxy supporter can provide for proxy forces such as economic aid, it is not clear that any resources provided outside of a power relationship can be eliminated from inspection. In fact, with an international system organized around power, it is difficult if not impossible to ever have a relationship of assistance without some component of power. If the supporter chooses not to directly engage, it still has a presence on a battlefield through the use of its material by another state. A hypothetical example can help here. Imagine there are three states: State A, State B, and State C. State B and State C are two powers at war with one another. Though State B is losing, with heavy manpower and territorial losses, State A decides to send State B material support in the form of arms. Is State B considered a partner with State A or a proxy of State A? 40 Dunér, Military Intervention in Civil Wars: The 1970s, Ibid., Ibid.

15 Youra 12 State A decided to indirectly intervene in the conflict and have a presence through State B. As the word proxy implies, State B is fighting in place of State A. Adding interests into this does not change the outcome of State B as a proxy. For example, imagine that both State A and State B are opposed to the policies of State C and therefore State A decides to support State B. In this case, the interests are not only compatible but also congruent. An alternative formulation is also possible: imagine that State B is generally more favorable to the policies of State A and therefore, State A decides to support State B. While the interests are different (State B s interests concern State C while State A s interests concern State B), they are certainly compatible and can exist together. Divergent interests would be problematic because such an instance would not provide an incentive for State A to provide assistance for State B. Even if the pull theory was realized and State B requested aid from State A, there would still be no incentive if State A could not attempt to attain its own interests. Presumably, if State B requested aid from State A, there is some compatibility of interests on which State B can draw for persuasion. Power does not seem to change the outcome of State B as a proxy either. Should State B exist in a true power equity with State A, then there would be no need for material support. Alternatively, differences in power only strengthen the proxy dynamic. For example, imagine that State B and State C are two weaker powers at war with one another, while State A is a developed, military superpower. Power plays a role in that State A has an excess of power in the form of resources that it can provide to State B. Additionally, state A could pressure State B to continue fighting and make assistance contingent on a refusal to ceasefire. This only strengthens the assertion that State B serves as a proxy for State A.

16 Youra 13 In each scenario, material support appears essential to the proxy relationship. This support can only be material or immaterial. Immaterial support such as raising political favor or creating a coalition of politically supporting states is extremely hard to measure. Dunér s lowest rung on the instruments of intervention ladder includes military warning by an outside force, but this cannot be considered tangible, material support. Immaterial support contains the rhetoric of action but no identifiable action, while material support signifies the action of support. Power and interests both play an important role but once a state makes the decision to intervene indirectly and thereby provide a form of support, it has decided to play a role in a conflict as a proxy supporter. Power plays a necessary but not sufficient role in the proxy relationship by serving as the foundation for resource assistance. This stands in stark contrast to Dunér. The implications of this conception are great. If the transfer of support plays an integral role in a proxy-supporter relationship, is all support considered a form of proxy support? When is support not working to build a proxy relationship? Morgenthau may argue that any support is directed at achieving one s own interests and therefore all support is aimed at developing proxies for one state s bidding. For him, controlling others actions (through either force or support) provides the avenue for increased power, which is the ultimate and central aim of international politics. 43 At the same time, labeling all support as proxy support is not useful as a concept for examination. The study of proxy wars requires a clear delineation between proxy support and other forms of support. To start, it may be useful to outline the difference between proxy support and support in direct war. Returning to the hypothetical example above provides some insight. If State A were to intervene directly and send troops to fight alongside State B, then it is no longer using State B s forces in place of its own. Even if State A sends arms in addition to troops to State B, the 43 See Morgenthau, Politics among Nations; the Struggle for Power and Peace.

17 Youra 14 arms are no longer serving as State A s presence on the battlefield. Instead, the arms are merely supplemental. Therefore, the context in which assistance is given appears to have an impact on the classification of a proxy relationship. The above analysis, however, only focuses on support during conflict. Can proxy support consist of assistance provided outside of the bounds of a conflict? Modifying the hypothetical examples above, if State B is not in conflict with State C and still receives material support from State A, does it serve as a proxy for State A? It is not clear that this is the case. Without conflict, State B may serve as a trading partner of State A, but there is no current conflict in which to fight. Context is important. Without State B fighting, State A cannot have presence on a battlefield through its material support. Andrew Mumford, another proxy war researcher, captures this issue of context. He defines proxy wars as the indirect engagement in a conflict by third parties wishing to influence its strategic outcome. 44 This definition highlights several important factors: intervention must be indirect (with no men representing the third party in an on-the-ground, fighting capacity); third parties can be either non-state actors or states; the conflict must have started before outside intervention; and the conflict can be between states, between states and non-state actors, or between non-state actors. Although Mumford asserts that the intervention must be indirect for it to be considered a proxy situation, he examines proxies that use both direct and indirect forms of intervention. This becomes complicated because a warring party will likely use all tools at its disposal, including proxy fighters, once it has committed to directly engage. Once foreign troops are present, proxy warfare becomes a tool for a larger fighting strategy. It is a type of warfare instead of war itself. This stands in contrast to proxies as the sole means of 44 Mumford, Proxy Warfare, 11.

18 Youra 15 engagement. In these such instances, the proxy fighters become the proxy supporters sole means of fighting. Only this fully indirect approach isolates proxy war as a strategy within itself. Even with some flaws, Mumford s definition is certainly the most simply stated. Indirect engagement in a conflict captures the importance of material support while also highlighting the central role of context. 45 Three things, however, are unclear. First, Mumford does not clearly state that the indirect engagement is the sole form of engagement for a third party in proxy wars. Second, power is taken as given in the indirect engagement, though this may not be a large issue since power may be captured by the fact that there is an engagement in the first place and therefore the third parties must be able to support such an intervention. Third, there is no mention of a compatibility of interests, which is necessary for two parties to work together. Moreover, there is no mention of a warring party receiving the support. Hughes captures it well when discussing a direct relationship of assistance, but this is missing from Mumford s definition. Indirect engagement could be non-targeted and therefore falls outside of the bounds of proxy wars. For example, imagine that instead of working with State B, State A indirectly engages in the conflict between State B and State C by imposing embargoes on all warring countries. If State B is economically stronger than State C, then State A has helped State B weaken State C without engaging with State B. In such an instance, State B cannot be labeled as a proxy of State A. This highlights the importance of direct assistance to a warring party. Proxy wars involve more than simply third parties indirectly engaging since the third parties must be engaging with directly warring parties. These actions must be positive in nature. Essentially, the indirect engagements of third parties must be directed through a warring party. To summarize, the above definitions each appear inadequate to capture the concept of proxy wars. From this analysis, the present study defines a proxy relationship as a dyad of 45 Ibid.

19 Youra 16 material support between an outside party and a warring party based on a compatibility of interests and an asymmetry of power in a context of conflict. The third party entity, the warring entity, and the conflict is left ambiguous to apply to any issue; the third party or warring party can be a state or non-state actor in an interstate, intrastate, or extrastate 46 conflict. This definition captures Dunér s three components of interests, support, and power, while including Mumford s focus on the context. The proxy relationship of direct support is necessary to examine the ways that all of the components come together. Regarding the compatibility of interests, while it is difficult to pinpoint one guiding interest in a country s foreign policy, a proxy war requires a relationship in which an outside supporter and a warring party s interests work in harmony. This is essential to defining proxy wars in relation to other forms of intervention because unlike other forms of intervention, proxy intervention requires the establishment of this relationship. When the relationship is applied surrounding war, the result is a proxy war. Therefore, a proxy war is a conflict through the use of an entity receiving material support based on a compatibility of interests and an asymmetry of power. This is an indirect intervention in a conflict through another entity. To use Dunér s hierarchy of intervention instruments, any material support below indirect-combat involvement may be part of a proxy relationship. It is important to state here the difference between proxies and forces of international organizations. Some of the definitions can theoretically capture international organizations. Conceptually, states provide resources to forces headed by international organizations in order achieve larger strategic objectives. These are NATO forces, AU forces, or UN forces operating under Chapter VII of the UN Charter. While the member states donate troops, the forces operate as independent actors. While Mumford explicitly excludes international organizations from his 46 The term extrastate conflict is taken from the Correlates of War Project, which defines an extrastate war as a war involving a state actor and non-state actors located outside of the warring state s territory.

20 Youra 17 proxy war definition, he only considers balanced, peacekeeping forces, not peace-making forces that inherently favor one warring side over the other. 47 The central reason for removing international organizations from this discussion is their structures. The international organizations mentioned above do not pose as fully independent actors. In the case of NATO, the consensus-based North Atlantic Council ensures that each country has a crucial vote in the engagement process and therefore should be viewed as merely an extension of the state. There are no independent interests/actions afforded to these organizations without state consent. Even in the Security Council, a permanent member must at least abstain and not oppose an action for it to come to fruition. Dataset Methodology This study sets out to examine proxy wars in greater detail, with a specific focus on the proxy relationships themselves. Before examining a specific case study, it is important to garner a sense of the larger macro proxy war context. Hughes produced the most comprehensive study regarding state supporters and their non-state actor proxies, but the study uses each instance as a case study in narrative form. This does not suit macro-level analysis. Understanding the larger picture of proxy relationships and wars requires an examination of the component parts of the definition. First, a proxy dataset should be organized in proxy dyads based on conflicts since the dyad is the organizational unit through which support flows in a proxy relationship. As discussed above, any evidence of material military support between a warring party and an outside third party marks a proxy dyad. These dyads can be state-state, state-non-state, non-state-state, or even organization-state or organization-non-state. In order to gain a general understanding of the relationship between proxies and their supporters, this study builds a subset of a larger proxy war dataset, focusing on the Middle East 47 Mumford, Proxy Warfare.

21 Youra 18 region from 1945 through The Middle East was chosen because it has a high prevalence of local conflict and contains no great power. The presence of a local great power has the potential to change the dynamic of proxy wars by acting as a passive deterrent for outside intervention. If an actor planned to use an actor located close to a great power as a proxy, it may be deterred from doing so simply because the reaction of the great power would enter into its cost-benefit calculation. As illustrated by the Monroe Doctrine, stronger states first view their power projection on the local level and therefore indirect intervention may be seen as undesired intervention nonetheless. Without a great power, the Middle East provides a clear, isolatable opportunity for proxy support. To build the dataset and refine variables, the study started with Uppsala s Conflict Data Program s External Support Data Disaggregated/Support Dataset from The disaggregation ensured different entries for relationships between one warring party and multiple supporters, and therefore it suited the study of the exact relationships between each warring party and third parties. The coding scheme for the type of external support was especially useful. The dataset includes a dummy variable for whether external support existed, thus even taking account of conflicts not involving external support. UCDP coded each type of external support in a categorical scheme similar to Dunér: troops, joint operations, access to territory, weapons, materiel/logistics support, training/expertise, funding/economics support, intelligence materiel, other forms of support, and unknown support. Each of these variables served as variables denoted with a 0 if such support did not occur and a 1 if such support did occur. To these variables, as well as general information on conflict year, conflict name, name of warring party, and the name of the external supporter, the present study added dummy is the latest available data from the Uppsala Conflict Data Program. Stina Högbladh, Therése Pettersson, and Lotta Themnér, "External Support in Armed Conflict ," ed. Uppsala Conflict Data Program (UCDP) (Department of Peace and Conflict Research, Uppsala University, 2011).

22 Youra 19 variables for the location of the external supporter (whether in the region or outside of the region), the type of conflict (intrastate, intrastate, non-state, or extrastate), and the power level of each external supporter (great power, major power, minor power, international organization, the level of involvement according to Dunér s hierarchy, and the number of forms of support. In order to gain a macro-level picture from 1945 to today, this study combined UCDP s External Support Data with the Non-State Actor Data developed by David Cunningham, Kristian Gleditsch, and Idean Salehyan. 49 The UCDP dataset begins in 1975 so the Non-State Actor Data, which uses the UCDP dataset, expanded the survey to include conflicts involving non-state actors from 1945 to Neither the UCDP External Support Data nor the Non-State Actor Data captured the possibility of external or proxy support in all of the interstate wars in the Middle East. To compile this information, the study used the historical archives of The New York Times and The Washington Post. These two papers were chosen based on their relatively large focus on politics and international affairs, increasing the likelihood that outside third party state involvement in a conflict would be covered. Since most external support activities are covert, the two newspapers acted as fact checkers for one another by confirming, clarifying, or denying an actor s external involvement while also ensuring as wide a coverage as possible. The often-covert nature of these interventions means the use of open source investigation is one of the only feasible means to capture data, though it is admittedly imperfect at best. All war coverage for the eight Middle Eastern interstate wars was examined in an attempt to gather references and specifics on external involvement in these wars. These wars include: the 1948 Arab-Israeli War, 1956 Sinai War, the 1967 Six Day War, the War of Attrition, the 1973 Yom Kippur War, the 1974 Turco- 49 David E. Cunningham, Kristian Skrede Gleditsch, and Idean Salehyan, "Non-State Actor Data," ed. Kristian Skrede Gleditsch (2012).

23 Youra 20 Cypriot War, the Iran-Iraq War, the 1982 War over Lebanon, the 1991 Gulf War, and the 2003 Invasion of Iraq. Any states that participated directly by sending troops were not considered part of a proxy dyad, even if they used proxies as a supplementary tool in their intervention. Such an action moves beyond the use of a proxy relationship as the sole means of intervening in a conflict. This fits with Dunér s instruments of intervention ladder. Sending troops is certainly above indirect combat involvement.. Another aspect of the definition is power. All literature reviewed above discusses the necessary role of power in the proxy interventions. Supporters of the pull theory of intervention may even assert power as a sufficient cause of interventions. Either way, power as a variable appears important. In an international system of anarchy, it is assumed that greater powers have more mobility and flexibility in their actions than weaker states and thus leads to the following hypothesis: H1a: If a proxy war occurs, it is more likely that the actors are strong actor in the international system. Should the data should show that a larger percentage of weaker states support proxies than powerful states, this hypothesis would be provided wrong. At the same time, this may be supported by the idea that proxy wars lessen the risks associated with breaking state sovereignty since these supporting actions are less visible and direct than complete intervention, 50 and these actions are less financially costly. While powerful states have the ability to directly intervene, weaker states may be reliant on indirect intervention to achieve their interests. Therefore, another hypothesis is the exact opposite: 50 See Bale, "Terrorists as State "Proxies": Separating Fact from Fiction."; Hughes, My Enemy's Enemy: Proxy Warfare in International Politics; Mumford, Proxy Warfare.

24 Youra 21 H1b: If a proxy war occurs, it is more likely that the actors are weak actors in the international system. This study operationalizes power as the capabilities of a state, as measured by the Correlates of War Project s national capabilities index. The index creates Composite Index of National Capability (CINC) rankings for each country by examining a country s energy consumption (pec), iron and steel production (irst), military expenditures (milex) and personnel (milper), total population (tpop), and the number of inhabitants in urban cities compared against all other countries (upop). The CINC ranking is a percentage of one country s capabilities compared to that of the entire international system. While the CINC rankings place a large emphasis on demographic size and traditional measures of industrialization, 51 it provides a farreaching overview of a country s overall capabilities: military, economic, and demographic. This is more useful than examining GDP or another single-measure of power. As Waltz states, States are not placed in the top rank because they excel in one way or another. Their rank depends on how they score on all items. 52 While non-material forms of power, such as Joseph Nye s soft power may be useful in other areas of study, proxy support concerns the ability of a state to give material support. The CINC rankings capture this emphasis on material support. While military advising and therefore brainpower is part of proxy support, it would be problematic to capture such military brainpower as a percentage of the soft power overall. For the purposes of this study, it can be assumed that great and major material powers are correlated with high degrees of military training because these powers have more resources to devote to training and war games. 51 Steve Chan, "Is There a Power Transition between the U.S. And China? The Different Faces of National Power," Asian Survey 45, no. 5 (2005). 52 Waltz, Theory of International Politics, 131.

25 Youra 22 To apply power positions to the proxy dyads, each recipient of support and each proxy supporter is organized by type: great power, major power, and minor power states; international organizations; and non-state actors. This study considers great powers as comprising at least 12% of the world s capabilities. Both the Soviet Union from 1946 until 1990 and the United States from 1945 through today fit this definition as great powers. Major powers are those countries individually capturing at least 2% of the world s capabilities. These include the United Kingdom, France, China, and Germany. Minor powers comprise any country with a CINC ranking of less than 2%. While it is assumed that great powers are able to intervene more due to a greater access to resources, it is interesting to note the ability of non-state actors to serve as proxy supporters specifically because they do not have the resources of a state. Instead of operating on the state and systemic level, which is constrained by international rules of sovereignty and nonintervention, non-state actors operate on the sub-state level and often outside the rules of the state system. In this way, they may have more flexibility regarding proxy support. At the same time, these non-state actors are often weaker than states and may be unable to gather a sufficient amount of resources to substantially assist a warring party. Regarding data collection, it is harder to map the resources and operations of non-state actors, especially since their activity may occur on the illegal, covert black market. This has the possibility of displaying a smaller role for nonstate actors as proxy supporters than is actually the case. The ability to provide material support may not be dependent on the capabilities of a country alone; it may also be linked to the location of a state. Sending capabilities (arms, materiel, advisors) across far distances is a financially costly endeavor. Shorter distances provide decreased transportation and shipping costs. Therefore,

26 Youra 23 H2a: If a conflict occurs in a given region, then it is likely that actors in the region will serve as proxy supporters to a greater degree than actors located outside of the region. Moreover, neighboring countries are more concerned with sub-state or localized conflicts. The multiple hierarchy model proves informative here since it conceptualizes different power projection cones in the international system. 53 In each cone, there are great powers, major powers, and minor powers. This can be extended to include international organizations and nonstate actors. According to the model, the largest cone is the global hierarchy itself, followed by regional cones, though one could also add a state-level cone. Countries have a direct interest in the affairs of their neighboring state since an intrastate conflict has the ability to impact the regional cone and potential local power projection. Major and great powers not located in the region may only be concerned with a conflict once it has reached the point of the regional or international cone. Previous studies on intervention support this claim. Examination by Ronald Tammen et al. shows, The great powers have intervened in minor power interstate conflicts in just over 1 percent of all opportunities to do so. 54 If that is the case for interstate conflict, then the percentage is likely less for intrastate and extrastate conflicts. Local conflicts involving nonstate actors are not likely to draw foreign great or major powers. Therefore, there is a potential refinement to hypothesis 2a: H2b: If an intrastate or extrastate conflict occurs in a given region, then it is likely that weaker actors in the region will serve as proxy supporters to a greater degree than actors located outside of the region. It is not assumed that weak countries located outside of a given region are likely to act as proxy supporters simply because the cost of involvement is too high. For Dunér, the 53 Ronald L. Tammen et al., Power Transitions Strategies for the 21st Century (New York, N.Y.: Chatham House Publishers, 2000); Douglas Lemke, Regions of War and Peace (New York: Cambridge University Press, 2002). 54 Tammen et al., Power Transitions Strategies for the 21st Century, 70.