вход по аккаунту



код для вставкиСкачать
15 International dimensions of internal
Nils W. Metternich, Kristian Skrede Gleditsch,
Han Dorussen and Andrea Ruggeri
Civil wars are by definition violent conflicts between a state and some
form of non-state actors (Sambanis 2004b). Perhaps not surprisingly, most
scholars have looked for features within countries to account for why such
conflicts break out and how they evolve (Blattman and Miguel 2010).
However, it is misleading to simply equate civil wars with domestic conflict
and look exclusively within countries to understand civil war.
Transnational factors can play an important role for the initial outbreak
of civil wars and the escalation and dynamics of conflict once underway. But transnational factors also can contribute to the resolution and
mitigation of such conflicts. Indeed, since civil conflicts are not necessarily limited to an individual state and often involve participation by other
states in various forms, a strict dichotomy between civil and inter-state
wars is often untenable and inappropriate. Failing to theoretically and
empirically account for the transnational dimension of civil wars is likely
to lead to biased conclusions about the onset, duration and intensity of
such conflicts. Moreover, many policy prescriptions based exclusively
on a domestic understanding of civil war can be highly misleading and
potentially counterproductive.
Most research implicitly or explicitly adopts a ‘closed polity’ approach
to civil war, where civil war is driven exclusively by features and events
within individual countries. The conventional wisdom on civil war tends
to portray civil wars as problems of poor and weak states (Fearon 2005),
largely driven by the greater opportunities for conflict when states are
unable to enforce control or deter protest (Herbst 2000a). In addition,
poor and weak states are thought to provide greater incentives for
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 214
31/10/2012 09:21
International dimensions of internal conflict
Note: Location of intra-state conflicts (circles) and internationalized intra-state conflicts
Based on the Conflict Site Dataset (Hallberg 2012).
Figure 15.1
Location of armed conflicts, 1946–2005
participation in conflict when the formal economy affords few rewards
for normal economic activities (Collier and Hoeffler 2004b). It is often
inferred that civil wars are best prevented through shoring up the capacity of the central government, achieving sustainable economic growth,
and more controversially, establishing more representative political
institutions or power-sharing institutions (Hartzell and Hoddie 2003b).
The transnational dimension can make many of these inferences
problematic. Political processes do not stop at the water’s edge, and
countries are not isolated units, impervious to events and actors outside
their boundaries (Franzese and Hays 2008; Gleditsch and Ward 2006).
This insight is neatly summarized in Tobler’s (1970, p. 236) First Law of
Geography: ‘Everything is related to everything else, but near things are
more related than distant things’. Indeed, when looking at Figure 15.1,
displaying a map of the geographical centre points of intra-state conflicts
between 1946 and 2005, we observe a strong tendency for civil conflicts
to cluster geographically. Certain regions, such as the Caucasus and
Western Africa, have an especially large number of conflicts over this
period, while Northern Europe and Eastern Asia experience only a few
armed conflicts. This in turn suggests that individual conflicts may not
be independent of one another, and that the risk of conflict is dependent
on actors and events outside its borders (Buhaug and Gleditsch 2008).
While political interdependencies are likely to extend beyond the geographical dimension (Beck et al. 2006), there is much reason to believe
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 215
31/10/2012 09:21
216 Elgar handbook of civil war and fragile states
that in the context of civil wars the geographical dimension is especially
An Example of the Transnational Dimension
The Albanian insurgency in Macedonia in 2001 provides a helpful
example highlighting many transnational dimensions of civil war. Any
new state will be less well consolidated than older states, and Albanians
clearly were marginalized in the new and explicitly ethnic Macedonian
nation-state. However, violent conflict erupted 10 years after independence in 1991, and at a point where the state was much better established.
Moreover, the insurgency actually emerged in the aftermath of notable
concessions to Albanian parties by the Macedonian government. Hence,
the outbreak of the conflict clearly contradicts what one would expect
from purely domestic conceptions of civil war.
In this example, the transnational context provides a helpful frame to
understand the timing of the civil war as well as the specific conduct. While
the Albanian population in Macedonia by itself is small and lacks significant resources, it is clearly part of a larger Albanian community in neighbouring Albania and Kosovo. Moreover, the Albanian community in
Kosovo had a long history of mobilization, and had recently seen violent
conflict following the emergence of the Kosovo Liberation Army (KLA),
which challenged the Yugoslav government by violent means (as opposed
to older organizations such as the Democratic League of Kosovo party,
which favoured non-violent strategies). In the aftermath of the Kosovo
war, an Albanian organization similar to the KLA called the National
Liberation Army emerged in Macedonia, drawing on individuals that
had previously participated in the Kosovo conflict, as well as arms and
resources from the previous war in neighbouring Kosovo. The fact that
the fighting overwhelmingly took place in border areas attests to the close
links between the insurgency and the Kosovo war.
The 2001 Albanian insurgency in Macedonia provides a compelling
example where transnational features were important for the outbreak,
and it seems unlikely that conflict would have come about without the
linkages to the external prior conflict in Kosovo. However, sceptics
may wonder whether this is an isolated case, and how representative the
example may be. Although there can be many important transnational
linkages, we shall here focus on a set of common transnational mechanisms that often affect the risk of conflict, which we shall discuss under the
labels ‘transnational actors’, ‘transborder conflicts’, and ‘transnational
conflict externalities and spillovers’.
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 216
31/10/2012 09:21
International dimensions of internal conflict
The actors in civil wars are not necessarily confined within individual
countries. For example, ethnic groups often span international boundaries, and transnational kin frequently participate in or provide important
support for insurgencies in other states (Cederman et al. 2009). Resources
mobilized transnationally can often be more difficult for governments to
control or target. As such, the domestic political status and resources of
an ethnic group can be a poor guide to the motivation for violence and
the events that influence mobilization and changes in political strategies.
Cederman et al. (2009) provide systematic evidence suggesting that transnational groups are much more likely to engage in conflict than would
be expected on purely domestic characteristics. They extend a purely
domestic model of conflict among excluded ethnic groups and a central
government, where the risk of conflict is proportional to the relative size
of the ethnic group, to consider the difference between groups with and
without transnational segments. Figure 15.2 illustrates the core empirical
findings, with the solid line representing the predicted risk of conflict over
the period from 1946 to 2003 for a group with transnational segments
and the dashed line the predictions for a purely domestically confined
group. Figure 15.2 demonstrates that conflict is dramatically more likely
for groups with transnational kin. As an illustrative example, in the case
of the Kurds in Turkey we have a relative group size of 0.14, or about
seven Turks for every Kurd. If this were an isolated ethnic group, the
With transnational group
Without transnational group
Relative group size
Figure 15.2
Risk of ethnic conflict by relative ethnic group size and
transnational ethnic kin
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 217
31/10/2012 09:21
Elgar handbook of civil war and fragile states
model would imply a predicted probability of conflict of less than 20
per cent. However, given the presence of Kurds in neighbouring states
(that is, Iran, Iraq and Syria), there is a much higher predicted risk of
conflict, that is, 43 per cent. The opportunities for the Kurdish Workers
Party (PKK) to operate out of Syria and more recently Iraq have been
important for the organizations ability to engage the Turkish government
in conflict over the repeated outbreaks of the violent conflict. From the
perspective of accommodation, Dorussen (2005, 2009) finds that groups
with transnational kin also are more likely to receive concessions from
governments, reflecting their stronger bargaining power. More generally,
Salehyan (2009) demonstrates that rebel organizations frequently find
sanctuaries in bordering countries, which makes it difficult for the government to successfully defeat insurgents. In addition, Salehyan (2007)
argues that transnational rebels complicate the bargaining environment.
First, the government has less verifiable information about transnational rebels as they operate and hide beyond its borders. This increases
informational problems and leads to a higher probability of bargaining
failure. Second, any agreement between the government and the rebels
will involve the demobilization of military forces. In addition, there are
incentives for the rebel organization to keep forces beyond the borders
of a state, which gives rise to commitment problems that can prevent
peaceful settlements.
The examples reviewed illustrate how fighting and mobilization in civil
wars often takes place in border areas. Although borders in a technical
sense are just lines in the sand and often not difficult to cross from a purely
military perspective, the status of international borders generates different constraints and opportunities for governments and rebels. The fact
that borders formally delineate state sovereignty makes it more difficult
for governments to violate the sovereignty of other countries, while such
constraints are less relevant for rebels. Governments risk retaliation from
neighbouring countries from territorial incursions, and face difficulties in
targeting transnational support. This in turn means that rebels can have a
logistical advantage in operating out of extra-territorial bases, and transnational rebel movements can be more difficult for governments to deter
or defeat (Salehyan 2009).
Buhaug and Gates (2002) demonstrate more systematically that a large
share of civil wars tend to be fought in border areas. This prevalence of
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 218
31/10/2012 09:21
International dimensions of internal conflict
fighting in border locations in civil wars in turn raises the question about
whether rebels operate in neighbouring countries with the explicit consent
of other states. Although there are situations where inter-state conflicts
have emerged out of direct support for insurgent groups (such as the
1978–79 Tanzanian Ugandan war that toppled Idi Amin), direct intervention on the side of insurgents is a very costly action for states. Support
is more likely to come in the form of indirect military support, and such
support is often officially denied by the supporting state. Salehyan et al.
(2011) estimate that about 45 per cent of all the civil wars in the post1945 era saw rebel organizations receiving support from a foreign state.
External support requires both some degree of demand by insurgents
and supply by a potential supporting state. It can be shown that support
seems more likely in instances when groups have transnational segments,
states have a history of antagonisms, and rebel groups are moderately
strong relative to the government. One interpretation of the last aspect is
that rebels must be sufficiently weak for foreign support to be attractive,
given the problems that this may entail with respect to potentially competing goals of the patron state and the potential loss of legitimacy arising
from external backing. At the same time, a rebel organization must also
be sufficiently strong to be considered a plausible candidate by potential
The general notion in the literature is that conflicts that include transnational actors and are closely located to international borders are more
difficult to resolve and more likely to escalate. In the last section we
highlight another important aspect of the transnational dimension of
civil wars: the risk of contagion through conflict externalities and spillovers. Recalling Tobler’s (1970) First Law of Geography: ‘Everything is
related to everything else, but near things are more related than distant
things’ (p. 236) alludes to the risk of conflict diffusion and clustering.
Indeed, Gleditsch (2007) finds that countries with neighbours experiencing civil war are at a greater risk of conflict than would be expected by
their purely domestic characteristics. Although this in part may be due
to transnational actors, fighting can also create externalities and spillover
mechanisms that by themselves increase the risk of conflict. The presence
of conflict in another state can help facilitate violent mobilization even in
the absence of direct linkages between actors, through either emulation
of successful rebellions, or the direct import of arms and combatants.
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 219
31/10/2012 09:21
Elgar handbook of civil war and fragile states
Put differently, spatial dependency goes beyond the assumption that near
things simply experience similar structural conditions and are therefore
related to each other. From our perspective, transnational dependencies
arise from interactions between the relevant conflict actors and their
strategic dependencies.
In the context of economic policies Simmons et al. (2006) discuss four
mechanisms that can easily be related to the transnational dimension of
civil wars. The first mechanism is coercion, which may be direct or indirect
and hard or soft. Coercion is frequently exercised by external actors that
try to influence the dynamics of civil wars. Interventions by international
organizations or neighbouring countries are commonly observed in civil
conflicts (Gleditsch and Beardsley 2004). These direct interventions may
include military, economic and diplomatic aspects to change the course of
an armed confrontation. More indirect measures might be the toleration
of transnational rebels on its territory (Salehyan 2009).
The second mechanism highlights that civil war actors can learn across
borders (Horowitz 1985, p. 6). This extends the idea that learning and
informational updating only occurs between the belligerents (Slantchev
2003; Smith and Stam 2004). For example, rebel organizations might learn
from insurgents in neighbouring countries which strategies and logistical
networks are successful in an armed struggle. Similarly, governments are
able to observe successful policies to prevent and mitigate armed conflicts.
Governments can also observe the behaviour of the international community and form expectations about potential international interventions
on their or the rebels’ behalf. Therefore, learning is a powerful mechanism
that leads to transnational dependencies.
The third mechanism, competition, refers to pressures that actions
within one country place on surrounding ones. Armed conflicts within
one country are likely to have externalities that affect proximate countries
(Murdoch and Sandler 2004). For example, negative externalities arise
when countries have important economic ties and dependencies with a
country that experiences civil war. But in some cases armed conflict can
have short-term positive externalities for surrounding countries. The civil
conflict in Iraq, for example, hampers oil production in one of the largest
exporters, thus decreasing the global supply of oil and benefiting other oilexporting states in the Gulf region.
The potential impact of externalities should also have an effect on
whether surrounding countries consider coercive transnational interventions. Competition might also refer to more indirect consequences of, for
example, economic development. If proximate countries experience strong
economic competition, this should lead to regional economic development in the long run. However, we know that economic development is
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 220
31/10/2012 09:21
International dimensions of internal conflict
negatively related to civil war and thus transnational economic activities
can potentially lead to a lower risk of armed conflict.
Civil wars can also decrease the government’s military competitiveness
vis-à-vis neighbouring countries (Gleditsch et al. 2008). While the government has to allocate its forces against an internal threat, this may increase
the risk of inter-state violence by lowering the neighbours’ expected costs
or increasing the expected benefits of using military force. Civil wars
and insurgencies expose and exacerbate weaknesses in the state’s military capabilities and divert resources away from defence against foreign
enemies. This position of weakness may invite opportunistic attacks
against the state, which would not have taken place in the absence of
the internal conflict. In such cases, the attacker is not concerned with the
outcome of the civil war and does not necessarily sympathize with rebel
aims, but is primarily motivated by capturing territory or resources. In
some cases, conflicts arise over the externalities generated by conflicts,
including refugees and collateral damage.
Many disputes arise out of border violations, where government forces
pursue rebels into the territory of neighbouring states. Conflict may also
arise out of alleged support for rebels. In some cases, states may respond
to alleged support for insurgents by neighbouring countries, by either
direct military retaliation or support for insurgent groups in the neighbouring states. Western Africa in the 1990s provides many examples of
governments supporting insurgencies in neighbouring countries. Finally,
many studies document large negative social and economic effects of civil
wars for neighbouring countries, which in turn could undermine political
stability. This attests to how civil war cannot be considered exclusively
domestic problems, but in many cases will constitute transnational security issues and challenges for other states.
The fourth mechanism is emulation, which refers to behaviour that is
not coerced or responsive to competition and learning. In the context
of civil wars, emulation has gained little attention. However, ideological movements frequently seem to follow this mechanism by attaching
themselves to a momentarily in-vogue ideology. Leftist movements in
South America or Islamist movements in Southeast Asia represent spatially clustered organizations which do not obviously result from the
success of similar movements (learning) or identical external support
Finally, Franzese and Hays (2008) add a fifth category to the diffusion
mechanisms proposed by Simmons et al. (2006): migration. The relationship between migration and conflict diffusion has received some notable
attention. Salehyan and Gleditsch (2006) point out that migration is an
important aspect of conflict diffusion. First, refugee flows from war-torn
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 221
31/10/2012 09:21
Elgar handbook of civil war and fragile states
countries can imply the direct influx of combatants, arms and ideologies
from neighbouring states that facilitate the spread of conflict. Second,
rather than fighting openly with the host government, refugee populations can provide resources and support to domestic opposition groups
of a similar ethnicity or political faction. Third, as a negative externality, refugee flows can change the ethnic balance in a country, sparking
discontent among local populations towards the refugees as well as the
government that allows access. Finally, refugees may pose actual or perceived negative economic externalities because they compete with locals
over scarce resources such as employment, housing, land and water,
constituting an economic ‘threat’.
Graham K. Brown and Arnim Langer - 9781848448421
Downloaded from Elgar Online at 10/24/2017 11:59:31AM
via (NUS) National University of Singapore
M2983 - BROWN & LANGER TEXT.indd 222
31/10/2012 09:21
Без категории
Размер файла
204 Кб
00021, 9781781006313
Пожаловаться на содержимое документа