Territory and war

Strategic and intrinsic worth

The simplest reason for war over territory might be the strategic worth of the territory itself, a value that does not vary among actors, but extends directly from its geostrategic position and material value. According to this theory, which might be thought of as the ‘geostrategy school’ (Mackinder, 1904; Spykman, 1944; Sprout & Sprout, 1968), strategic worth describes the security value of a given piece of territory. Is the territory astride major routes of communication? Does it share an interstate border? Does it contain natural barriers to invasion from other states, especially historical enemies?

While Mackinder and Spykman spent much of their energy on identifying key locations, Harold and Margaret Sprout explored how geography informs actors’ interests and how it conditions their behavior. Indeed this is what traditional and contemporary international relations theory leads us to expect. And it has found some empirical support. Paul Huth, for example, found that a powerful predictor of territorial disputes among states was the desire to control strategically located territory (Huth, 1996), though he also found that disputes over strategic territory make up only a small percentage of cases. Furthermore, Huth found that the presence of a bordering minority with linguistic or cultural ties to a challenger state was not a primary cause of territorial disputes between states in the post-World War II period, a finding that was later substantiated (Huth, 1996; Toft & Saideman, 2010). More recently, Carter (2010) argues and finds that ‘when territory is strategically located, target states are more likely to consolidate their position, while challenger states are less likely to escalate militarily’. According to Carter, the control over a piece of territory with strategic value has an impact on the balance of military capabilities between disputants. Since the target of territorial claims usually controls the disputed territory, the target state can improve its military strength just by consolidating this strategically located territory. This, in turn, negatively affects the prospects for the challenger state to gain a victory, thus making escalation less likely.

Given that location is often key, scholars have spent a lot of energy examining borders. The most extensive testing of borders and location came in the 1970s with the publication of Harvey Starr and Benjamin Most’s articles testing whether the amount of shared borders and proximity of states to one another contribute to a state’s proclivity for war (Starr & Most, 1976, 1978; Siverson & Starr, 1990; Starr, 2002; Starr & Thomas, 2005). These articles conceptualize as constraints on the interaction opportunities of states. It follows from this conceptualization that bordering states are more likely to engage in war simply because of the ease of interaction. Wars occur only when states can reach each other.

Instead of emphasizing opportunity, Bremer (1992) highlighted various causal mechanisms that can explain his finding that contiguity is a strong predictor of the war onset. Vasquez (1995) also added to the findings on contiguity and war onset by explaining that wars are not only concentrated among neighbors because this provides an opportunity for war, but also because territoriality, under the right circumstances, provides a motivation for war. More specifically, he argued that although non-neighbors may have ideological, economic, or political disagreements, neighboring countries are much more likely to experience armed conflict because territorial issues have greater probability of escalating than these other sources of conflict (Vasquez, 1995). Recent empirical research has established the effects of territory and contiguity on war onset independently. These studies suggest that territory is a more consistent determinant of conflict than contiguity (Vasquez, 2001; Senese, 2005). Senese (2005) points out that since nation-states comprise well-defined areas of land, territorial disputes may become a foundational threat to a state’s security. Consequently, state leaders are more willing to resort to the use of military force when the issue at stake concerns territory. Linking territory to the willingness of states to go to war is a major theoretical advancement, since the link makes it possible to identify specific factors that motivate states to enter into conflict, rather than emphasizing characteristics that are relatively constant such as ample opportunity to start a war because of proximity (Toft, 2003; Senese, 2005).

In distinction to strategic worth, intrinsic worth arguments focus on the wealth or resources that inhere in a territory. Does it contain infrastructure or industry of value? Does it have space for expanded population or arable land that could support an expanded population? Does it contain a concentration of mineral or natural resources? If the loss of contested territory threatens to undermine the security or economic survival of an actor, then that actor may resort to force. The theoretical relevance of the intrinsic value argument is confirmed in a study by Hensel (2001), in which he reports that action over territorial claims is most likely when more valuable territory is at stake.

With regard to conflicts within states, Sorens (2011) argues that, on one hand, local mineral abundance is likely to lead to secessionist objectives, since total or greater control over the contested territory would allow the secessionist movement to enjoy a larger share of the benefits flowing from mineral revenues. On the other hand, mineral abundance provides the government with more incentives to exercise greater control over these areas, which has a negative impact on the rebel parties’ opportunity structure to engage in armed violence. These two contradicting mechanisms may explain why Lujala et al. (2007) find no significant relationship between oil and gas located in conflict areas and the length of territorial conflicts. Similarly, based on evidence from 13 civil wars, Ross (2004) finds that lootable resources generally do not increase the likelihood of conflict outbreak, but rather prolong conflicts once underway.

Because of the tangible nature of territory and its divisible properties, conflict is not seen as inevitable, and in fact, might be avoided if an equitable division could be arranged (Rosenau, 1966, 1967). This corresponds to the rationalist logic set out by Fearon (1995) that conflict parties usually can agree on a division of the issues over which they compete. The argument that territory can be divided, thus making territorial wars avoidable, contains a powerful logic and has been found to explain some variation in outcomes. Consider the work of Gibler on borders. According to Gibler, a key reason democracies do not fight each other is because they have stable borders between them (Gibler, 2007; Miller & Gibler, 2011). Gibler argues that states commonly resolve their most dangerous disputes prior to becoming democracies, including territorial issues. As a result, conflicts involving democracies are usually of less salience and therefore easier to resolve (Miller & Gibler, 2011). This is in line with the finding that in territorial disputes issue-related variables matter more than regime type variables in explaining whether these disputes will turn violent (James, Park & Choi, 2006). Moreover, it was the settlement of borders that enabled states to become democracies not vice versa (Gibler, 2007). Although other studies support Gibler’s conjectures (Kacowicz, 1995; Mitchell & Prins, 1999; Miller & Gibler, 2011; Owsiak, 2012), not all evidence supports Gibler’s argument that resolved territorial issues rather than democracy by itself explains the absence of armed conflict between democracies. For instance, Huth & Allee (2002) find strong support for democratic dyads being less conflictual, controlling for many other factors.

That territory can be divided, thus making territorial wars avoidable, is also reflected in the numerous studies on peaceful resolution of territorial disputes. Several studies have found that negotiations are more likely when disputes concern territory; yet, the findings in the same studies suggest that military strategies are also more likely to be employed in territorial disputes (Hensel, 2001; Hensel et al., 2008; Mitchell & Thies, 2011). Nonetheless, it seems that states have strong incentives to peacefully resolve territorial disputes, since allowing these disputes to remain unresolved can carry high economic and military costs (Simmons, 2002, 2005). This also explains why states are willing to commit to arbitration in order to resolve territorial disputes. Arbitration is an option often preferred by political leaders from democratic states, since, compared with bilateral negotiations, arbitration better shields political leaders from domestic criticism (Huth, Croco & Appel, 2011).

A number of factors with a positive impact on the peaceful resolution of territorial disputes have been identified. Huth, Croco & Appel (2011, 2012) find that international law increases the prospects for the peaceful resolution of territorial disputes when relevant legal principles justify the resolution of the dispute in favor of one state.

Symbolic worth

However, territory may not be only a material, tangible good, but also of intangible and symbolic value. Although noted by Rosenau, broader theorizing and testing of this argument came only in the early 1980s when John Vasquez engaged a ‘hunch’ that territory’s value might be intangible, and therefore more prone to conflict escalation and war (Vasquez, 1983, 1995).

This research was followed by other compelling takes on the potential unique role of territory (Diehl & Goertz, 1988, 1991; Diehl, 1992; Forsberg, 1992). The authors find that territorial changes that are regarded as more ‘important’ are most likely to lead to violence.

What is important to point out is that this research considers the ‘relational’ value of the territory in which the actors attach some sort of historical or identity-value to it, and that this value varies across actors. This is a powerful insight, but when it came to operationalization, the research faltered. The indicators the scholars used in their model were all ‘materially based’. Nevertheless, the contribution was significant for calling attention to different meanings of territory and how one might think about testing those different meanings statistically.

This notion of the perceived meaning or value of territory has emerged as a dominant theme in the literature that addresses bargaining failures and war, with dispute over territory often featuring centrally.

Bargaining and indivisibility

James Fearon (1995) argues that commitment and information problems, although acknowledged as theoretically possible, account for much violence in the real world, including that over territory, and dismisses the notion that indivisible issues account for much. His model assumes that the issue at stake can always be made perfectly divisible (by, for example, allowing two princes to alternate stints on an otherwise single and indivisible throne). Although numerous territorial disputes have been resolved through peaceful means, the numerous territorial disputes that have turned violent contradict the logic that division of territory is always possible. In general, whether negotiations can resolve territorial disputes seems to depend on the value of the territory. Another reason why Fearon does not regard the issue of indivisibility as an analytically distinct path to war is that some possible solutions to issue indivisibility are infeasible due to commitment problems (for instance, two princes alternating stints on a throne might be infeasible, since the prince that sits on the throne first will have incentives to hold on to the throne, because of a concern the other prince will do the same). For example, conflict parties both claiming a piece of territory with strategic value may resort to violence because of a commitment problem rather than the indivisible nature of this territory (Carter, 2010).

In most cases of disputes over territory, however, especially where an identical location is crucial to both opponents, such as Jerusalem, the prize may simply not be realistically divisible. With regard to the Palestinian–Israeli dispute, Kaufman (2009) notes that emotive symbols and narratives concerning territorial indivisibility enabled hardline leaders on both sides to block compromises and escalate conflict. This is in line with an argument made by Hassner (2003) that conflicts over sacred spaces are indivisible. Besides emphasizing the symbolic value of territory, evolutionary theory has been used as an explanation for why humans are so concerned with preserving territorial integrity (Johnson & Toft, forthcoming 2013/14). Statistical evidence seems to confirm the argument that there is something special about territory. Hensel & Mitchell (2005) find that disputes over a piece of territory with high intangible salience are significantly more likely to turn into wars. This suggests that intangible issues are often indivisible. Indeed, even allowing one side access to some revered resource may represent a defeat for the other (Toft, 2006).

While theorizing on territorial indivisibility was first conducted by scholars studying intrastate conflicts (Toft, 2003), several recent works also address this issue in relation to interstate wars. Hassner (2006) observes that many territorial conflicts between states last longer than the material benefits linked to the piece of land would justify. He explains this puzzle by the fact that the symbolic value of disputed territory grows over time, eventually leading to indivisibility. Territory invested with nationalist, religious, ethnic, and other emotional value makes partition and redefinition of the disputed territory, as well as compensation for it, more difficult to implement.

Furthermore, without actually addressing the issue of indivisible territory, several authors have made the case that since territory is loaded with a number of emotional and normative elements, territorial disputes between states should be explained by adopting social and psychological perspectives, and by referring to subjective conceptions of justice and international norms (Barzilai & Peleg, 1994; Forsberg, 1996).

This indivisibility of territory along with the interplay of domestic and international politics might help to explain why the number of states in the interstate system has quadrupled in the last 100 years. After World War I, and especially after World War II, the rights of national and ethnic minorities became more privileged in international law, opposing self-determination of ethnic groups against the territorial integrity of states. Such conflicts have been pushed to the center of global politics, leading to the demise of once formidable empires. Although self-determination movements tend to be defensive, territorially confined, and limited in scope, the dynamics of bargaining and the nature of stakes compel patron-states and outside actors to get involved. Consequently, national self-determination and the question of secession are often an international problem as domestic actors vie for control over territory (Saideman, 1997).

Just consider the end of the Cold War and the disintegration of the Soviet Union and Yugoslavia, which led to the creation of almost two dozen new states. This transformational shift in the global balance of power resulted from groups that advanced claims based on the legitimacy of national self-determination over pieces of territory (Ayres, 2000). These claims ranged from modestly increased autonomy (economic or political) to secession and independent statehood.

Since 1945, the Minorities at Risk data project, for instance, has identified at least 283 groups at political risk for discrimination and repression, many of which have engaged in militant activism in order to advance claims of self-determination. Historically, most groups that have engaged in secessionist war have failed to achieve their objectives. Since 1990 about half of all civil wars resulted from ethnic groups seeking greater autonomy or statehood, and a number of conflicts remain violent or unresolved.

The problem is perhaps more common than uncommon the world over and spans all kinds of conflict. Robert Pape argues that the principal impetus behind all major suicide campaigns is not religion per se, but rather territorial defense against foreign invaders and occupying powers (Pape, 2005). Osama bin Laden himself referenced the foreign occupation of Muslim lands as a primary basis for his holy war against the West, calling in his February 1998 fatwa for the killing of US and allied military personnel and civilians ‘in order to liberate the al-Aqsa Mosque and the holy mosque from their grip, and in order for their armies to move out of all the lands of Islam’ (Simon & Benjamin, 2000: 68).

States and reputation

The reason this is so is because territory is vital to all sides, though in different and not necessarily mutually exclusive ways. For ethnic and religious groups, territory is invariably tied to the group’s identity. Control over homeland means a secure collective identity. For states, control over territory is directly linked to physical survival: all else equal, more territory means more physical security from conquest or from coercion. States are likely to view an ethnic group’s bid to control territory – even objectively worthless or costly territory – as a threat whenever they fear precedent-setting (Toft, 2002, 2003; Walter, 2003, 2006, 2009).

That territory is indivisible due to states’ concerns for their reputation has also been put forward as an explanation of interstate territorial wars. According to Wiegand (2011), states not only use force to demonstrate their resolve to an opposing state, but to credibly signal their resolve to other states. The willingness to use force sends a costly signal to all the adversaries of a state, mitigating the incomplete information these adversaries have about a state’s resolve. In essence, the outbreak of a militarized territorial dispute transfers reputation for resolve from one dispute to other disputes, making the outbreak of other militarized territorial disputes less likely. Furthermore, Atzili & Pearlman (2012) explain how states have incentives to use threats and/or punishments against another state to coerce it to prevent non-state actors from conducting attacks from its territory, a dynamic Atzili describes as triadic deterrence.

Territory as opportunity

This brings us to the conditions under which conflict is likely to happen. Again, we can turn to territory for an answer. The symbolic and identity attachment aspects of territory have already been addressed. But there is another crucial element: the territorial disposition of the challenging group. Research clearly shows that where groups live will influence whether they come to blows with the state. The location of rebels with the objective to secede is likely to be at a distance from the capital, whereas rebel groups taking up arms over state governance are more likely to be near to the capital (Buhaug & Gates, 2002; Buhaug & Rød, 2006; Cederman, Buhaug & Rød, 2009). Since rebel groups with secessionist demands aim to create a new state, these rebel groups are not likely to take up arms in the territory that is proximate to a capital holding the seat of state power. Most secessionist movements emerge in peripheral regions and a separatist war is more likely to take place in regions of states’ borders than nearer to states’ capitals. Location thus geographically defines the characteristics of a war.

Moreover, since the projection of state power tends to decline with increasing distance from the capital, the location of armed conflicts has an impact on the capabilities, limitations, and vulnerabilities of armed forces. One of the major advantages rebel parties have compared to governments is that they can choose the area of operation. Rebel parties aim to minimize the possible damage inflicted by government forces, which is why they are more likely to take up arms in remote areas where the balance of power is less in favor of the government. Fuhrmann & Tir (2009) identify distance as one of the tactical advantages of rebel groups involved in territorial disputes, making territorial conflicts more likely to evolve into enduring internal rivalries than other types of domestic conflict.

Whether rebel groups are concentrated within or dispersed across a territory will also influence the likelihood of war. Groups that are regionally concentrated are much more likely to mobilize for self-determination and engage in violence than those that are dispersed throughout the territory of the state or live largely in urbanized areas (Toft, 2002, 2003; Jenne, Saideman & Lowe, 2007; Weidmann, 2009; Weidmann, Rød & Cederman, 2010). In fact, it turns out that regional concentration of a group within a circumscribed territory serves as practically a necessary condition for a self-determination movement and secessionist war to emerge (Toft, 2003).

Why is this? It appears to be the case that group concentration (1) makes political organization easier over a compact territory, thus overcoming the collective action problem; (2) facilitates military operations; and (3) defines the territory over which claims can be made (Toft, 2003; Weidmann, 2009).

We know from the insurgency literature that foreign borders provide needed sanctuary for rebels challenging states by allowing access to goods and services that might otherwise be inaccessible (Arreguín-Toft, 2005). Indeed, Salehyan highlights that rebel parties can evade the power of the state in several ways, but one particularly effective way is to set up safe havens in a neighboring country because state strength is limited by international boundaries. This makes it possible for rebel parties to mobilize and sustain their forces, thus making rebellions feasible (Salehyan, 2007, 2008, 2009). In addition, neighboring states and ethnic kin living across the border might support groups seeking self-determination (Brubaker, 1996; Jenne, 2006; Cederman et al., 2013). To date, most of the evidence suggests that decisions to support self-determination movements are driven by politics at home rather than the plight of the group, especially when ethnic ties are involved (Saideman, 1997; Saideman & Ayres, 2008).

While there is a growing body of research on the impact of external support on the onset and resolution of armed conflicts, further research is needed with regard to what causes some external states to support self-determination beyond their borders. There is an assumption of a status quo bias among states when it comes to borders, and given this bias, they are expected to show restraint and not support secessionists. The reason is that they too are vulnerable to secessionists (most states in the world are multi-ethnic) and therefore would not want to weaken the norm of inviolate territorial borders. Although this logic is sound, the behavior of states consistently undermines it. Time and again, states with their own secessionists have supported secessionists elsewhere. The most obvious case is Russia, which has been fighting secessionist Chechens for two decades

OPEN IN VIEWER

Figure 2.

Number of journal articles that addressed territory and the causes, conduct, or resolution of war, 1964–2012

In general terms, however, the dynamics during wars are less studied than dynamics pre and post conflict (see Figure 2), making war itself the least understood phase. It is only recently that the role of territory in relation to conduct of war has been firmly established on the peace research agenda (Kalyvas, 2006; Kalyvas & Kocher, 2009; Balcells, 2010, 2011; Bhavnani, Miodownik & Choi, 2011).