State legitimacy and Third World conflicts

Dealing with the post-World War II security issues, in this article I outline an approach that takes into account the context of the weak states in Third World environments. In this respect, first of all I make a case that domestic politics should be factored in the security concerns of states. Especially in the Third World context domestic political and social dissentions become key factors in defining foreign policy behavior of such states. Secondly, I highlight regional context, especially unstable regional conditions, regional hierarchies, and other factors that contribute to conflict propensities of states situated in the Third World environments. Finally, I make a case that, compared to democracy, legitimacy is a better variable to explain the link between domestic politics and foreign policy behavior of the Third World states. Legitimacy is not only value-neutral variable, but can also be applied across time and place.


Introduction
In the post-World War II scenario most conflicts and wars take place in the Third World, and so factors unique to Third World must be taken into account. In this context, domestic politics is identified as the most important factor in the security conceptions of the Third World states. Domestic politics is important in advanced-industrial countries as well, but the difference is that in Third World countries the domestic arena is also conflictual and internal security problems are prevalent due to ethnic, religious, or regional differences. In other words, their own national integrity is at stake and becomes the dominant theme in their security conceptions. Another unique factor of the Third World states is that due to contested territories and porous borders, regional considerations figure prominently. For this reason, regional hierarchies, regional power struggles, and regional insecurities are also important in the case of Third World states. Overall, in this paper it is argued that domestic politics of the states and regional environments must be dealt with when theorizing about international conflicts. As states in Third World suffer from weaker social and political fabric and they reside in neighborhoods of other states that suffer from similar problems, such states are more prone to interstate conflicts.
After reviewing the literature that highlights the above-mentioned factors unique to the Third World, namely domestic politics and regional context, I see if the assumptions and conclusions drawn from the democratic peace theory could be applicable to the Third World. It is important to deal with this literature because democratic peace theory is the dominant approach that connects domestic political concerns to the foreign policy behavior of the states. In looking at the application of the democratic peace theory to the Third World situations, a lot of substantive and theoretical problems were identified. Most important of them was that a lot of Third World states (especially those involved in interstate conflicts) have hard time meeting preconditions of democracy because of their precarious political conditions. In order to come up with a concept that could be applied across time try to change the constellation of societal interests that control state power rather than from other states" (Wendt and Barnett, 1993: 329). In this context, Holsti (1996) reports that since 1945 "classical interstate wars" have declined to "about 18% of all wars," while most of the wars have been "within and about states, not between" them (Holsti, 1996: 25). Moreover, "almost 77% of the 164 wars were internal, where armed combat was not against another state but against the authorities within the state or between armed communities" (Holsti, 1996: 21). Kelly (2007) explains this phenomenon by pointing out that in the Third World "the traditional mechanics of the security dilemma are there but are placed on hold, especially when proximate states face similar internal challenges" (Kelly, 2007: 217). Interstate wars in the case of Third World states are "doubly dangerous" as "wars make domestic turbulence dangerous to frail regimes. Even victory is threatening. Conquered populations worsen the legitimacy crisis, and flimsy institutions lack the infrastructure and coercive capacity to dominate further, potentially unruly territory" (Kelly, 2007: 217). This is an important point that not only are the weak states less likely to initiate international conflicts due to their domestic problems, but they are less likely to reap the post-war benefits even if they achieve victory. Internal threats to state leadership are more important because in most of the Third World countries state leadership represents one or few groups, while the excluded groups always pose a threat to the state leadership in the capital. This situation gets worse when most of these "subnational groups owe allegiance to and act on behalf of interests other than the national interest." In fact, "instead of identifying with the state, individuals identify with ethnic, religious, or regional groupings" (David, 1991: 239). Resultantly, even when excluded groups manage to come to power, they are less inclined to build national consensus and more likely to promote their subnational agendas. This situation gives other excluded groups an incentive to challenge such regimes.
As a result of internal state weaknesses, such states find themselves in a vicious cycle of unpredictable regime changes. In this respect, David (1991) highlights that "hundreds of Third World leaders . . . have been overthrown by their internal enemies . . . Coup d'état alone have accounted for nearly two hundred regime changes in the Third World" (David, 1991: 240-241). Belkin (2005) provides more detailed evidence of this phenomenon by reporting that, "during the 1950s and 1960s a coup or attempted coup occurred every 4 months in Latin America, every 7 months in the Middle East, and every 55 days in Africa." More recently, "there were approximately 357 attempted coups in developing world from 1986 to 2000, and about half of all third world states experienced a coup during this period. Of these attempts, 183 coups (or 51 percent) were successful" (Belkin, 2005: 20). To explain these frequent coups, Belkin (2005), following Huntington's (1968) theory of praetorian politics, 2 argues that military coups reflect the inability of "political institutions" to channel "participation" into "nonviolent patterns" (Belkin, 2005: 20). This observation implies absence of the fundamental condition of state formation in the Third World, namely absence of violence from political life. Krause (1998) also supports this argument by stressing that in order to talk about state security it must be acknowledged that "the claim of the state to loyalty and legitimacy that is at the heart of the modern project is rooted in its promise to evacuate organized (and sometimes random) violence from political life, to provide at least the minimal conditions of order, and perhaps (although not always) to support some measure of representativeness in its political institutions" (Krause, 1998: 132). Frazer and Hutchings (2011), on the other hand, give a different perspective on link between politics and violence. While acknowledging that "the overall aim of liberal politics is to remove violence from the political agenda," they argue that "liberal political theorists persistently ignore the ongoing reliance of the liberal state on specific kinds of "virtuous violence," 3 both inwardly and outwardly directed. From the liberal point of view, violence is an instrument that may be used to attain political ends to the extent that those ends are legitimate and other means are less appropriate. (Frazer and Hutchings, 2011: 68-69) Keeping in view this argument, we can presume that

State legitimacy and Third World conflicts Waheed A. Khan
what Belkin (2005) and Krause (1998) are referring to are private acts of violence by the groups that challenge state legitimacy, not legitimate use of force by state. Secondly, even if state violence is used it must be directed to legitimate ends. In short, all this argument boils down to Weber's justification for the "legitimate" use of force by state (see Weber, 1922). Legitimate use of force became problematic in the Third World. As Krause (1998) argues, modern means of coercion and accompanying ideas and institutions, transmitted from the North to the South, failed to bring stability to those regions. The armed forces in the Third World countries did not play a positive role in the development of their countries. Rather, they became tools of political power in the hands of specific groups that used them "to impose a particular (often violent) order on civil and political life," resulting in "the transplantation of unprecedented means of institutionalized violence and surveillance into political arenas that were empty of the countervailing checks and balances" (Krause, 1998: 135-136). In some cases, institutions of public violence developed their own narrow interests at the expense of the social and political interests. Atzili (2006/07) explains this dilemma faced by the state leaders by arguing that "well-trained armies and efficient bureaucracies -two central institutions of state building -are not only expensive to create and maintain, but given their potential to compete for power, they may be perceived as potential threats to the ruler" (Atzili, 2006/07: 150).
Like Krause (1998) and Atzili (2006/07), Bobrow and Chan (1988: 54) also come to the conclusion that major problem in the Third World countries is "the lack of role specialization by coercive institutions and by the military in particular. In gross terms, Third World military establishments are less focused on external threat, less reliable instruments of state policy, and more prone to internal fragmentation." To sum up the argument, David (1991) argues that the traditional conception of neorealists that a "legitimate government" protects its citizens from "private acts of violence" is not applicable to a large number of people in the Third World. In such situations, "the government is neither legitimate nor protector . . . the government is less of guardian against threats than it is the chief threat" (David, 1991: 251).
The argument presented above not only points to the complexity of Third World security problems, but also highlights naivety of an approach that seeks to solve their security problems through mere injection of military hardware. To an outside observer, increased military presence appears to be providing a condition of order, but in reality it becomes a violent instrument in the domestic politics of these countries. Azar and Moon (1988) nicely sum up the above arguments and explicitly link them to the security of such states. In their view: Defining the concept of national security in terms of physical protection of nationstates from external military threats is not only narrow, but also misleading. The threats facing the Third World are diverse and complex, and so are the dimensions and content of national security . . . The complex and multiple vulnerabilities of Third World states compel us . . . to look at a deeper structure and a broader spectrum of issue-nexuses . . . Domestic factors such as legitimacy, integration, ideology, and policy capacity play equally important roles in shaping the national security posture. (Azar and Moon, 1988: 12) Keeping in view Azar and Moon's (1988) above analysis, I focus on the domestic factors of states to explain conflicts among them. However, unlike Azar and Moon (1988), I take "domestic" legitimacy as an umbrella term that encompasses other domestic factors, like ideology, integration, and policy capacity of the central government. In other words, the goal of all these factors is to gain and enhance domestic legitimacy of states.

Regional insecurities, constraints, and opportunities
In addition to the innumerable domestic vulnerabilities, states in Third World regions face highly precarious external environments. Most scholars dealing with Third World states agree that the arbitrary colonial boundaries and inadequate conception of stateness in the Third World led to the contested territories and frequent interstate conflicts. Among such scholars, Ayoob's (1995) point is worth noting, who explains that "state building in the Third World is not conducted in a regional vacuum. Third World states impinge upon the statemaking process of other states in their neighborhood, especially those contiguous to them. This is a mutual and reciprocal process" (Ayoob, 1995: 47). Lake (2009) highlights the importance of regions and sets out the concept of regional security complex (RSC), which he describes as "a set of states continuously affected by one or more security externalities that emanate from a distinct geographic area" (Lake 2009, 35). Such RSCs "range from balances of power, to regional power concerts, collective security organizations, pluralistic security communities, and integration" (Lake, 2009: 36). Overall, presence of such RSCs also highlights some hierarchy at the regional level. However, level of such hierarchy varies from region to region. As he explains, "security hierarchy varies from diplomacy, at the anarchic end of a continuum, to protectorates, at the hierarchical end" (Lake, 2009: 45). In justifying his argument and countering realist assumption about anarchy at international level, Lake (2009) argues that if the system is anarchic it does not mean that all relationships within that system are anarchic. Rather, "relations between states can be and often are characterized by varying degrees of authority and, in turn, hierarchy" (Lake, 2009: 37). Sometimes regional security arrangements emerge on their own, but other times regional or extra-regional powers can promote them. As Lake (2009) explains, "regional orders emerge because of the strong positive externalities of social order and economies of scale in its production, and the mutually reinforcing legitimacy accorded the dominant state by local subordinates" (Lake, 2009: 36). In sum, adding region as a level of analysis adds new dimensions to the security dynamics of the states. Especially in terms of international relations theory, the argument about varying degrees of anarchy and hierarchy characterizing relationships among states is important to enrich international relations scholarship. Kelly (2007) deals more comprehensively with this literature. In terms of implications of regionspecific argument for international relations theory, he explains, "Buzan and Waever have provided four "key elements of essential structure" in regions -boundary, anarchy, polarity, and social construction. Waltz had only listed two -anarchy and the distribution of power" (Kelly, 2007: 199). Kelly (2007), however, cautions against analyzing regions from the perspective of system-specific theories. He points out that most of the scholars who deal with the regional security have deductively "'downscaled' Extant IR theory to the regional level." This is the approach that has been rejected by scholars such as Mohammad Ayoob, who propose a more inductive approach "'up' from the reality of regional security" (Kelly, 2007: 198). This new approach provides a different perspective to major international relations issues and, resultantly, different conclusions. As he explains, "Against the critical and neofunctional expectations of regional integration, weak-state regions actually generate regional IOs that reinforce, not erode, sovereignty . . . The joint strategy regionalizes not sovereignty but domestic conflict and elite pushback" (Kelly, 2007: 218).
Region-specific arguments are further highlighted by the scholars who have pointed at the "Zones of Peace" and "Zones of War." It is clear from these arguments that if neighborhoods are peaceful they can have positive effect on the domestic politics of states involved. Gibler (2007), in this context, singles out settlement of borders as the major issue for the presence of peaceful regions. According to him, "'Zones of peace' can thus be understood as the contagion effect of stabilized border," which, in turn, also leads to more economic development for the states. Due to the lack of militarization, defense spending can be reduced and the spared money can be used for economic development. Furthermore, cross-border trade also increases along the settled borders. This clustering of peaceful states should also affect the economic development of the states involved. Even if such states get involved in wars, those wars are fought at far away places, as local borders have already been settled (Gibler, 2007: 529). Gibler (2007) credits even democratic peace to settled borders when he

State legitimacy and Third World conflicts Waheed A. Khan
states that "the development path necessary for democratization selects democracies into a group of states that have settled borders, few territorial issues, and thus, little reason for war against neighbors" (Gibler, 2007: 529). Significance of the regions and neighborhoods is further highlighted by Bobrow and Chan (1988), who explain that Taiwan and South Korea do not differ much from Norway and Belgium in terms of "their size, military vulnerability, level of external threat, or even in their regimes' persuasive ability," but rather "being securely integrated in a superpower's core domain." States like Taiwan, Israel, and South Korea, despite being situated in the conflictual zones, however, were successful in carving out "a special niche in the strategic conceptions, political doctrines, and domestic opinions of their chief ally" (Bobrow and Chan, 1988: 57). The security paradigm outlined above might work for a handful of weak states, but for a vast majority this paradigm is inapplicable. There are many reasons for this inapplicability. First, as Bobrow and Chan (1988) themselves mention, most of the weak states hold only "marginal or derivative interests" to those in the powerful states (Bobrow and Chan, 1988: 57). Second, most of the Third World countries lack internal unity. As a result, any attempts by the major powers to solve security problems of such states end up further exacerbating domestic political rivalries. As Buzan (1988) argues, "a unified state cannot by definition engage in relations with an anarchic state without intervening directly in its domestic politics . . . Contact with such a state almost automatically involves taking sides in its internal rivalries, because the state machinery itself is an expression of those rivalries" (Buzan, 1988: 29). Furthermore, dependence on foreign economic and military resources impedes bargaining between elites and masses and contributes to the establishment of "illegitimate regimes" that follow the "colonial path" of "capital-intensive coercion rather than consent" (Wendt and Barnett, 1993: 332). Third, arms dependence of Third World countries on the major powers, instead of solving their external security problems, creates a classic security dilemma as neighboring countries become more suspicious and seek their own armaments. Resultant arms races are extremely dangerous and unpredictable in Third World environments. For example, Buzan (1988) explains, "in a race between producers the pace of competition is measured by the relatively slow moving and predictable factors of production," while in most of the small and weak states "the supply of even relatively small number of sophisticated weapons can cause large and abrupt changes in the local balance of powers" (Buzan, 1988: 38). In other words, if external powers are not sensitive to domestic political dynamics and regional balance of power, even the most sincere efforts by them might end up damaging the security situation, rather than improving it. All this discussion does not imply that international-level factors do not have any effect on domestic or regional level, rather that international factors are mediated through the domestic and region-specific dynamics.
In highlighting the more intimate link between internal and external dynamics of states, Jacobsen (2008) provides an interesting theoretical perspective that challenges the notion that international-level forces "impose" one-way changes at the domestic level. According to him, "behavior in one realm infects the other. Citizens -even those who cheer ruthfulness abroad -grow nervous when it spills over at home . . . The relevant analysis here is not one separate domestic and international dimension but rather one asking which dimension has primacy at a given time and under what conditions" (Jacobsen, 2008: 345). Jacobsen (2008) goes further in claiming more dominant role for domestic factors in international scenario. He argues, "international forces acquire social meaning and political muscle only as they are factored into national politics in ways that accommodate the interests, strategies and ideologies of dominant local players" (Jacobsen, 2008: 359). Like Jacobsen (2008), Westa (2010) also finds a more intimate relationship between domestic and international factors. In emphasizing a harmony of rules or norms between domestic and international dynamics, he explains that during the Cold War some states deeply internalized UN rules and developed corresponding domestic constraints. However, this internalization reveals that "states value a particular international order insofar as it corresponds to the configuration of their domestic social identities" (Westa, 2010: 520). Hence, in most of the like-minded states liberal democratic values were internationalized and became dominant norms in the international system. Acharya (2011) makes an interesting observation that problem with the Third World states is that they are uncertain about the adoption of international norms and rules. By referring to Ayoob, he argues that "Third World countries suffer from an "acute schizophrenia;" they have simultaneously rebelled against and adapted to the norms of the international system inherited from the colonial powers" (Acharya, 2011: 99-100).
So far I have made a case that the core security problem for most of the weak states lies at the domestic level, with a spillover effect primarily at the regional level. Internationallevel factors do matter, but they are mediated by domestic and regional-level factors. In the next section, I turn to the question why legitimacy is a better variable, compared to democracy, to link internal and external dynamics of state security?

Why legitimacy, not democracy?
In order to substantiate a case for "legitimacy" as a variable in explaining security issues of states, it is important to clarify the distinction between "democracy" and "legitimacy," as for some observers these two variables might be overlapping. It is the argument of this study that, even though democracy is the most popular and dominant legitimizing force, legitimacy can be achieved through non-democratic means as well. These means can provide at least short-term stability and security that are themselves needed to pave the way for democracy to hold if it is attempted later on. In other words, a stable democracy itself cannot be achieved without legitimacy. In this sense, this study does not in itself challenge democratic peace theory literature, but rather attempts to put it in proper perspective. Here, it is argued that a well-established democracy represents a state that enjoys very high level of legitimacy.
Keeping in view the above argument, we can identify two types of literature that attempt to explain the link between legitimacy and democracy. First, some authors find legitimacy and democracy as separate and mutually exclusive variables. They deal with both democratic and non-democratic states, but their main focus is to enhance legitimacy of a regime or a state. In other words, their main goal is to achieve legitimacy, which is vital for security. Second, there are scholars who find a more intimate relationship between legitimacy and democracy and take legitimacy as a precondition for democracy. For the most part, these authors either deal with the democratic countries or seek to democratize nondemocratic states. They are mainly concerned with the legitimacy of democratic practices, procedures, and policy outcomes. The main goal (or mission) of such authors is to promote and strengthen democracy around the world, which is essential for security, while legitimacy is taken as a means to achieve that goal.
Representing the authors who take legitimacy and democracy as separate variables, Azar and Moon (1988) argue that "national security performance may be enhanced more by 'benevolent' dictatorship with high legitimacy than by the fragile and incompetent pluralist regime with low legitimacy" (Azar and Moon, 1988: 16). Legitimacy as a variable is more desirable than democracy because of its applicability to diverse regimes across time and place. Furthermore, legitimacy is preferable to democracy because of its dynamic nature and its tendency to be value neutral. Holsti (1996) raises this point when he explains the reasons for preferring the variable "legitimacy" (termed by him as "state strength") 4 over democracy precisely because of the dynamic nature of the legitimacy as a variable, in addition to its tendency to be value neutral. He states: Democracies and strong states . . . are indeed overlapping categories, but to give . . . a generic quality not so bound by time and location, state strength [or legitimacy] is a more encompassing term. Moreover, the concept of strength is a more dynamic,

State legitimacy and Third World conflicts Waheed A. Khan
changing variable rather than a static democracy-non-democracy dichotomy. That dichotomy is also subject to criticism on the grounds that there is no consensus on the essential components of democracy. (Holsti, 1996: 145) White (1986, in this regard, pronounces that "it is generally agreed that all regimes, from naked tyrannies to pluralistic democracies, seek to legitimate themselves" (White, 1986: 462). Bukovansky (2002) affirms White's (1986) argument regarding absolutist monarchies when she declares, "a social bargain between the rulers and ruled existed, and legitimacy was a central issue, even for absolutist states" (Bukovansky, 2002: 70). Finally, Tarifa (1997) applies the same argument to Communist regimes and sees the attempts at indoctrination by different regimes, especially the authoritarian ones, as attempts on the part of the leadership to achieve legitimacy. In explaining this phenomenon, he states, "as far as ideological persuasion or indoctrination are concerned, it is fairly common knowledge that their aims are making the citizens identify with the goals of the ruling elite and legitimizing its rule. However, this applies to all types of systems, not just the Communist one[s]" (Tarifa, 1997: 448). The arguments presented above clearly specify that, compared to democracy, legitimacy is a better variable to explain security dynamics of a state.
With regard to scholars who see legitimacy as a precondition for democracy, it must be highlighted that most of their work is in the context of the democratic peace theory. The main argument of such scholars is that, in order to have a stable democracy, issues of legitimacy must be resolved. This might be the situation because legitimacy draws attention to more fundamental and deep-rooted questions faced by states, including acceptance of authority structure, values of the citizens, concerns of national unity, etc. In other words, it can be argued that democracy is no doubt the preferred form of government around the world, but its successful achievement and sustenance depends on paying attention to the historically and culturally contingent contexts of individual states, which can be better observed through the variable of legitimacy. This argument points to the fact that, whenever democracy departed from legitimacy-related issues, there were dire consequences for democracy itself. In other words, if a stable democracy is sought, attention must be given to the concept of legitimacy. Steen (2001), for example, argues that most scholars will agree that to achieve a stable democracy both leaders and masses have to have some confidence in the institutions, and elites must have "a basic level of trust" in public support as well as in each other. Especially, "a change of regime away from authoritarian rule makes the question of legitimacy particularly crucial since 'democracy' presupposes backing for its main institutions" (Steen, 2001: 697). Similarly, Chapman and Roeder (2007) imply that states that were created by partition (with resultantly improved legitimacy) experienced more democracy. Through their analysis they prove that "states created by partition were approximately four times as likely to democratize in any time period" (Chapman and Roeder, 2007: 687). Legitimacy remains an important consideration even after democracy has been established. For example, Jacobsen (2008) argues, "insofar as policymakers in democracies begin to encounter more attentive publics, the touchy issue of legitimacy also arises" (Jacobsen, 2008: 342). Anderson and Singer (2008) also find evidence of a link between legitimacy and democracy. They add the variable of distributive justice in this equation and find support for distributive justice affecting legitimacy, and legitimacy in turn affecting stability and longevity of a democracy. Dealing with the first part of this equation, they argue that "a long and growing line of research (mostly in psychology) has shown that both process and outcome (distributive) fairness powerfully affect the legitimacy of institutions" (Anderson and Singer, 2008: 572-573). They move on to find a link between the above-mentioned variables. They elaborate, "in a perhaps not-so-obvious way, the legitimacy-inequality link is also related to Lipset's (1959) argument about the importance of the size of the middle class for democratic longevity, presumably because a large middle class is synonymous with greater income equality and thus a better chance at achieving stable democracy" (Anderson and Singer, 2008: 573).
Legitimacy can become an issue for well-established democracies at any time. However, for new democracies it is even more potent. Moehler and Lindberg (2009), in this regard, argue that legitimacy is a very important consideration to consolidate democracy. They distinguish between the legitimacy of winners and losers after an election in the context of African countries and find evidence of different levels of legitimacy of political institutions among winners and losers after an election. In this respect, they find that electoral turnovers are vital in enhancing legitimacy of the political institutions. They further elaborate the significance of legitimacy even after democracy is established in a country. According to them, "electoral turnovers have a significant moderating effect on the citizens as winners and losers converge in their attitudes about the legitimacy of their state institutions thus creating incentives for elites on both sides to comply with the rules of the democratic game." They further elaborate that "not only are turnovers indicators that elites have accepted democracy, as Huntington (1991) argued, but power alterations also appear to guarantee shared levels of legitimacy between winners and losers in general population, thus furthering democratic consolidation" (Moehler andLindberg, 2009: 1449).
After clarifying the distinction between legitimacy and democracy, for the sake of this study it is important that we make a case that link between legitimacy and conflict is more significant than democracy and conflict. This is possible only if we tackle the basic arguments of the democratic peace theory. In addition to non-universal applicability of democracy as a variable, it can be argued that scholars dealing with the democratic peace theory face other daunting conceptual and methodological challenges when they apply the dichotomous variable of democracy-authoritarianism to a variety of states around the world, which are neither fully democratic nor fully authoritarian. As Kang (2010) mentions, "today . . . few authoritarian states trumpet their authoritarianism with pride; almost all claim to be some form of democracy and justify their rule based on some special need or circumstance" (Kang, 2010: 620). To generalize this tendency, Ottaway (2003) points out that "there is a vast grey zone that occupies the space between authoritarianism at one end and consolidated democracy at the other. The existence of this gray zone is well recognized by analysts, but there is little consensus on the types of regimes residing in it" (Ottaway, 2003: 6). Different authors have used different labels to describe this "grey zone," including political hybrids, mixed regimes, semi-authoritarian regimes, semi-democratic or illiberal democracies, etc. (See, e.g., Geomans (2000); O'Donnel (1994);Ottaway (2003); Reiter and Stam (1998);and Zakaria (1997).) The need to look beyond such fuzzy terms, which imply that these regimes are moving on a continuum from authoritarianism to democracy, becomes imperative especially if we consider the diversity, number, and permanence of such regimes. Fareed Zakaria (1997), for example, calculates that out of 117 democracies around the world, 35% are illiberal. For the newly established democracies this percentage goes up to 50% (Zakaria, 1997: 22). In such states, according to Zakaria (1997) andO'Donnel (1994), duly elected regimes end up adopting authoritarian practices. Ottaway (2003) would label such regimes as semi-authoritarian that "already do much of what the most democratization projects encourage: They hold regular elections, allow parliament to function, and recognize, within limits, the rights of citizens to form associations and of an independent press to operate" (Ottaway 2003, 6). Finally, Skene (2003), affirming Zakaria's (1997), O'Donnel's (1994), and Ottaway's (2003) position, concludes that "in many instances in the developing world, democratization has brought elections, but little else. Worst of all . . . these unconsolidated and illiberal democracies are not getting better, and in some cases are actually deteriorating" (Skene 2003, 189-190). Even democratic dividends are not achieved in most of the cases. Gibler (2007), for example, explains, "democracies, especially young ones, are no less likely than non-democracies to experience insurgencies and civil wars, despite expectations that the protection of civil rights and broad participation should pacify rival ethnic groups" (Gibler, 2007: 513). To explain this situation, defenders of democracy claim

State legitimacy and Third World conflicts
Waheed A. Khan that we find ourselves in such limbo situation because democracy was attempted without its preconditions. Ottaway (2003), for example, explains that in those states "where formal democracy is accompanied by high levels of poverty, or where ethnic or religious conflict divides and mobilizes the population, for example, semi-authoritarian governments play on the grievances and fears and get support by promising solutions" (Ottaway, 2003: 17). This situation, in turn, boosts the legitimacy of authoritarian practices at the expense of democratic procedures. Especially in those cases where security and stability are major preoccupation of the population, not only provisions of security and stability enhance legitimacy of semi-authoritarian regimes, but increased legitimacy, in turn, also improves security situation of such states. In Ottaway's (2003) words, "the semi-authoritarian outcome is not always something imposed by autocratic leaders on a population that wanted something quite different, but it is often something accepted and even desired by population" (Ottaway, 2003: 13). Even regarding the non-security issues, legitimacy of the government can be enhanced by authoritarian practices. In the case of colonial Hong Kong, Ngok (2011), for example, explains, "the colonial regime based its legitimacy on performance. The Hong Kong people were content with being governed by a non-elected bureaucracy because the latter could deliver economic growth, protect freedom, and was corruption free . . . The Hong Kong people were materialistic and pragmatic" (Ngok, 2011: 685).
In highlighting the significance of legitimacy, especially with reference to democracy, Bukovansky (2007) argues that even the goal of viable democracy should be to create legitimate authority. She is of the view that neoliberalism's (and neorealism's) emphasis on scientism has led to the advocacy of political models "based on asocial, egoistic individualism." Such overemphasis on scientism "has led to a neglect of rhetoric, persuasion, and deliberation as political activities designed to construct authority and produce legitimacy by articulating and facilitating debate about the public good" (Bukovansky, 2007: 177). Efthymiou (2008) affirms Bukovansky's (2007) approach that the main goal of the democratic theorists was to establish public legitimacy of the institutions of power. According to him: A legitimate government is one to which people have transferred their right to rule themselves and to which they have consensually agreed to allow act on their behalf. However, Locke leaves open whether majority rule is a necessary element of democratic decision-making. It is only in Rousseau (1762) that the idea of the social contract comes closer to political equality in the idea that laws must be under the frequent review of the whole citizenry as it is in this way only that liberty can be guaranteed. We may call this the classic consensus theory of legitimacy. . . . This conception of legitimacy does not come without problem . . . achieving the general and full consent of all adult citizens is difficult if not impossible . . . The introduction of majority rule may be a pragmatic way of dealing with such disagreements but it comes with a heavy price tag on the natural rights of some citizens by asking them to consent to what they may find disagreeable . . . Majority rule seems in such cases to clash with the classic theory of legitimacy. (Efthymiou, 2008: 412-13) The above statement clearly shows that legitimacy was the main preoccupation of the democratic theorists, and that legitimacy is considered a preferred variable that would fix some of the problems related to democracy, especially with reference to the rights of the minority. In other words, main goal of a democracy (or any other regime) is to establish legitimate authority.
Contrary to the above arguments, the predominant approach in international policymaking circles is to seek democracy without paying proper attention to the dynamics of legitimacy. Negative consequences of such orientation are far-reaching, especially for young democracies where true essence of democracy is not fully realized and a link between objectives and procedures of a democracy is not visibly established. This overemphasis on scientism has resulted in an international discourse by which "sending advisors who show local authorities how to set up elections or courts or banking systems is presented as a technical rather than a political act. Calling such dissemination of ideas 'technical' links them to science and engineering, rather than politics and ethics" (Bukovansky, 2007: 184). Such a situation arises because foreign advisors try to impose a specific state model (mostly liberal, democratic) without taking into account domestic conditions of the state. The significance of this point can be highlighted by acknowledging diversity among established democratic states based on their domestic political conditions as well as their geographical locations and historically contingent dynamics. Validating Bukovansky's (2007) argument, Eriksen (2011) also cautions against judging diversity of states from the point of view of a specific state model (or an ideal). He argues that scholars need to develop new concepts that take into account the diversity of states. According to him, "instead of developing concepts which are better suited to analyze existing states, the gap between ideals and empirical reality is treated as justification for interventions which aim to close this gap, and make empirical reality conform to the model" (Eriksen, 2011: 231). Eriksen (2011) is of the view that if we develop concepts that take into account diversity of states, "the enormous variation between states would then be conceptualized as variation in the form of statehood, and not as degree of statehood or of 'failure.'" (Eriksen, 2011: 237).
To conclude this section, it must be emphasized that link between legitimacy and democracy is very intimate especially because legitimacy focuses on the public preferences and voluntary compliance, both of which are vital for a mature democracy. As long as international peace and security are concerned, legitimacy either contributes directly to the peaceful behavior of the states or it strengthens democracy that in turn leads to international peace (i.e., typical democratic peace argument). Legitimacy is definitely the preferable variable, especially in newly established states (democratic or otherwise) as such states face chronic legitimacy problems that threaten their internal and external security.

Conclusion
With specific focus on the Third World states, this articles highlight omissions in the dominant international relations theories that do not pay attention to domestic political dynamics and regional factors. Those theories that do pay attention to domestic political factors (e.g., democratic peace theory) attempt to impose concepts and assumptions drawn from advanced industrial states to weaker Third World states, without giving proper attention to the dynamics unique to the Third World. In addressing this problem, I propose the concept of state legitimacy, which plays an important role in capturing domestic political dynamics of such states. State legitimacy is the key component of state security as it directly translates into stability at home and recognition abroad, both of which contribute to reduced conflicts with other states.
Some scholars dealing with the connection between domestic and international factors do give passing reference to domestic legitimacy (along with other factors), but they do not deal with it in a more exclusive manner. This study fills that gap by specifically linking state legitimacy to international conflict. The approach adopted in this article takes into account contextual variables like history, geographical location, and international political considerations that help explain dynamics of legitimacy as well as relationship among states. Most of the literature that deals with the relationship between domestic politics and international conflict, on the other hand, consists of dyadic analyses of the states engaged in conflicts, without much regard to such contextual information.
Another major contribution of this study is that it draws attention to the fundamental questions faced by all state, regardless of their regime types. Value neutrality of legitimacy variable makes it well suited to explain behavior of states (and political entities) across time and place. Especially compared to democracy, legitimacy is not only a better predictor