Master’s Thesis

Spring 2009
Master in Peace and Conflict Studies (two-year master’s degree)
Department of Peace and Conflict Research

Master Programme in Politics and International Studies

UPPSALA UNIVERSITY

Living in the Wrong Neighbourhood

State Failure and its Implications for Neighbouring

Countries

Author: ALBERTO FERNÁNDEZ

Supervisor: KRISTINE HÖGLUND

Abstract *

Why and how do failed states affect neighbouring countries? The attention of the
international community towards state failure has grown significantly in recent years,
improving the understanding of this phenomenon; nevertheless, the knowledge about
the influence of state failure on neighbouring countries remain scarce. This research
aims at contributing to filling up the existing gap by analyzing two different cases of
state failure –Liberia and Afghanistan– and its consequences on four of their
neighbours –Sierra Leone, Guinea, Pakistan and Tajikistan. More concretely, this
research investigates the importance of three elements in the relationship between these
countries; Insurgency, Identity and Illicit trade. The research argues that failed states
generate conflict-enhancing mechanisms –which might lead to conflict outbreak– in
their neighbours through the creation of informal networks. The empiric evidence
shows how insurgency-based informal networks have a decisive role in the outbreak of
conflict. At the same time, illicit trade-based networks wield an insignificant influence
on the outbreak of war. Identity-based informal networks are only significant in the
Afghan case, and a deeper understanding of their operation is still needed.

* Author’s note: The author wishes to thank Kristine Höglund for her invaluable assistance and
supervision. The author is also grateful to Annina Murray for her comments and corrections in the latest
version of the manuscript.

-1-
Contents

1. Introduction 4

2. Theoretical Framework 6
a. Situating Theory 6
b. Theoretical Construct 12

3. Methodology and Case Selection 17

4. Liberia 22
a. Liberian Failure: The three I’s origin 22
b. Liberia – Sierra Leone 24
i. Insurgency 25
ii. Identity 27
iii. Illicit Trade 28
c. Liberia – Guinea 31
i. Insurgency 32
ii. Identity 33
iii. Illicit Trade 34
d. Within Country Analysis 35

5. Afghanistan 38
a. Afghan Failure: The three I’s origin 38
b. Afghanistan – Pakistan 39
i. Insurgency 40
ii. Identity 44
iii. Illicit Trade 45
c. Afghanistan – Tajikistan 47
i. Insurgency 48
ii. Identity 50
iii. Illicit Trade 50
d. Within Country Analysis 52

6. Analysis 55
a. Insurgency 57
b. Identity 60
c. Illicit Trade 62
d. Critical Evaluation of the Findings 63

7. Conclusions 65
a. Policy Implications 66

Bibliography 67

-2-
Abbreviations

AFL Armed Forces of Liberia

AFRC Armed Forces Revolutionary Council
APC All People’s Congress
ECOMOG ECOWAS Monitoring Group (see ECOWAS)
ECOWAS Economic Community of West African States
FATA Federally Administered Tribal Area
HT Hizb ut-Tahrir al-Islami
IMU Islamic Movement of Uzbekistan
INPFL Independent Patriotic Front of Liberia
IRP Islamic Renaissance Party
ISAF International Security Assistance Force
ISI Inter-Services Intelligence
LURD Liberians United for Reconciliation and Democracy
MODEL Movement for Democracy in Liberia
NATO North-Atlantic Treaty Organization
NPFL National Patriotic Front of Liberia
NPRC National Provisional Ruling Council
NWFP North West Frontier Province
PANAFU Pan-African Union (Sierra Leone)
RFDC Rassemblement des Forces Democratique de Guinée
RUF Revolutionary United Front
SLA Sierra Leone Army
SLPP Sierra Leone People’s Party
UCDP Uppsala Conflict Data Programme
UIFSA United Islamic Front for the Salvation of Afghanistan
UNAMSIL United Nations Mission in Sierra Leone

-3-
 “Rain does not fall on one roof alone”
African Proverb

1. Introduction

Why and how do failed states influence their neighbouring countries? The failure of
Somalia as a state was direly felt in its neighbours, especially Ethiopia and Yemen.
Thousand of immigrants have crossed into Yemen seeking a better life in the oil-
wealthy Arabic peninsula. At the same time, Ethiopia considered necessary to
overthrown the Supreme Islamic Court from Somalia’s government. To carry out such
task, Ethiopian troops entered Somali territory in 2006 with a high number of casualties.
Interestingly enough, countries such as Kenya have not suffered the effects of Somalia’s
failure to the same degree. The aim of this study will be to explore potential
explanations of the questions posed and establish the ground for the development of a
theoretical understanding of the problem.

Within the community of academics and policy-makers alike, state failure has
been identified as one of the major threats to global security. The events of September
11, 2001 in the United States and the recent outbreak of piracy in the Gulf of Aden,
which threatens one of the main maritime lines in the world, have increased the
attention and resources devoted to state failure. Nowadays, both communities have
gathered a considerable amount of knowledge on state failure as an intrastate
phenomenon, but the effects of state failure on its neighbouring countries remain poorly
researched. Several hypothetical effects of state failure have been underlined by the
academic research community (Bøås and Jennings 2005, Carment 2003, Kasfir 2004,
Klare 2004, Rotberg 2004). Most of this research have been done within an institutional
approach, taking into consideration only the pre- and post-failure stages, but no the
actual scenario. By approaching the issue from this perspective, the importance of
agency has been dismissed, leading to a lack of understanding of the phenomenon
beyond the institutional realm. Therefore, the gap that needs to be filled is twofold.
First, the understanding of state failure has to be enhanced. Second, the influence of
failed states on neighbouring countries has to be uncovered.

One of the very first attempts to enhance the understanding of the effects of
failed states on their neighbouring countries was done by Iqbal and Starr (2008).

-4-
Through a statistical analysis, Iqbal and Starr (2008) conclude that failed states do not
spread more failure, but rather help to create conflict in their neighbours, and these
results were more robust when it came to contiguous states. Iqbal and Starr (2008)
conclusions are therefore used as a departing point for this research, which will take
advantage of the neatness and generalization power of their study and enhance its
potential through a qualitative study.

Thus, the main aim of this paper will be to expand the knowledge and research
on state failure, answering the research questions exposed at the beginning: Why and
how do failed states affect their neighbouring countries? The paper will be based upon a
hypothesis which states that failed states spread conflict-enhancing mechanisms through
the creation of informal networks between the failed state and the neighbouring state.
Conflict-enhancing mechanisms refer to those circumstances that help to create the
characteristics necessary for the outbreak of the conflict, such as the creation of
grievances, access to weaponry, or access to funding (Ohlson 2008). Informal networks
are those instances of relationship at all levels – economic, political, power, social – that
are carried out outside the formal structures created by the state (Bøås and Jennings
2005). This hypothesis will be tested in two failed states, Liberia and Afghanistan. The
paper will show that informal networks based on insurgency create conflict-enhancing
mechanisms in neighbouring states, contributing decisively to the outbreak of war. At
the same time, economy-based informal networks do not decisively create conditions
for conflict outbreak. Ethnicity-based informal networks are decisive only in some
cases.

The organization of the paper will be as follows: in Section 2 the reader will be
situated within the field, exposed to the history of research regarding state failure and
presented the theoretical framework and the proposed theory. Section 3 will expose the
research design and the case selection. Section 4 is dedicated to Liberia and its influence
on its neighbours, to be concluded with an analysis of this case. Section 5 will be
structured in the same way but dedicated to Afghanistan and its neighbours. Section 6
will analyse the results of the two cases in order to test the strength of the theory.
Section 7 will be used as a summary of the whole research, drawing the main
conclusions and providing avenues for policy-implications and future research.
.

-5-
 2. Theoretical Framework

2.1 Situating theory

The increased interest in state failure during the last years has generated a growing body
of literature on the issue. This literature has been produced through different disciplines,
but mainly International Relations and Peace and Conflict, two closely related
disciplines that will be used to lay the foundation of this paper. State failure within these
two disciplines is not free of controversy, mostly as a consequence of the
conceptualization of the state that the term state failure implies. Zartman (1995) was one
of the first to establish a definition of state failure. He argued that state failure refers to
‘a situation where the structure, authority (legitimate power), law and political order [of
the state] have fallen apart’ (1995:1). Zartman’s definition has been widely used, but the
use of the term state failure that some governments have implemented has created
profound controversy in the realm of policy-making and academia. To avoid a debate
regarding what the state is, and the western-based bias that some authors argue, this
paper will stick to Zartman’s definition, considering its concepts at a minimum level
rather than at the level of western stable democracies.

This paper aims to shed light upon the effects of state failure on neighbouring
countries, but it will be also influenced by diffusion studies1. Therefore, it is important
to understand the core idea behind diffusion and its implications. Stemming from the
seminal work of Most and Starr (1980), diffusion studies have explored the different
causal mechanism leading to the spread of civil unrest and conflict. Most and Starr
argued that a phenomenon taking place outside the borders of a given country may ‘alter
the possibility of subsequent events through diffusion or contagion’ (1980:933). The
effect of events happening in nearby regions has also been noted by Skocpol (1979),
who pointed out the importance of cross-border relations in the rise of revolutionary
events, which can also be applied to the importance of the effect of events taking place
outside a country’s borders on its internal affairs. More recently, Iqbal and Starr (2008)

1
State failure and state collapse are two concepts that have been used interchangeably across the
literature. Zartman (1995) uses state collapse as the definition of crumbling institutions in weak states.
Nevertheless, lately the academic world has tended to understand state collapse as the complete
destruction of any sort of authority in a given country, Somalia being the only clear case of state collapse
(Rotberg 2004). This paper will use the term state failure to avoid misunderstanding and aiming at
theoretical clarification. To clarify to the reader, the term ‘neighbouring country’ will always refer to
countries bordering failed states

-6-
showed that the possibility of subsequent events also takes place in the particular case of
state failure. Nevertheless, Iqbal and Starr (2008) argue that what is diffused is not state
failure, but some of its more negative consequences. To understand its negative
consequences, it is crucial to comprehend the nature of state failure from a theoretical
perspective and apply the theoretical underpinnings to the hypothetical causal
mechanisms behind state failure’s spread of negative consequences. The theoretical
analysis and framework used in this paper will focus on the effects of state failure on
neighbouring countries, and therefore its theoretical underpinnings will stem from the
comprehensive understanding of state failure that this paper will explore. This is done in
order to isolate the effects of state failure per se from other possible influences such as
civil war or international conflict.

State failure has traditionally been researched from a state-centred approach,

paying little attention to the human-related consequences of state failure. Most of the
literature focuses on the institutions and the formal structures of the state (Bøås and
Jennings 2005, Cf. Howard 2008, Kasfir 2004, Rotberg 2004, Reno 2003). It has been
widely studied how and why these institutions collapse, but the effects of this collapse
and the actors involved once the state is gone have been researched less. Kasfir (2004)
provides an initial analysis of the consequences, arguing that collapse is followed by an
internal security dilemma that forces various groups to secure their survival. On the
same vein, Klare (2004) states that in the absence of the state, non-state actors tend to
take over the functions of the state. These non-state actors show an economic predatory
behaviour in order to carry out state-like tasks (Klare 2004). Kasfir (2004) and Klare
(2004) among others, point out the importance and potential danger of these groups.
They argue that non-state actors in a failed state may find incentives to expand their
activities beyond the borders of the state and at the same time, other groups may take
advantage of the chaos within the failed state to carry out their activities (Klare 2004).
Nevertheless, their state-centred approach neglects the informal scope of these
consequences. One of the results of this approach is a lack of systematic research on the
effects of state failure on neighbouring countries. It is these informal configurations, not
the failed structures of the state, which will be influencing neighbouring countries at the
point when the state has almost disappeared. Thus, in order to establish a
comprehensive theoretical framework, state failure has to be understood through the

-7-
lens of those remaining after failure and not from the point of view of failed structures
that are no longer functioning, and thereby, not influential.

The institutional focus of much of literature on state failure neglects a more

broad human-based understanding of the consequences of state failure. Yet, few
attempts have been made to set up analytical frameworks in which the state is not the
only focus, enlarging the scope to other possible intervening variables (Bøås and
Jennings 2005, Hameini 2007). Using the state as the unit of analysis reduces the range
and scope of theory when it comes to comprehending the effects of state failure outside
its borders. The institutional approach denies the importance of the informal
relationships of the population with the failure, their strategies to cope with its
consequences and the effects on neighbouring countries (Bøås and Jennings 2005,
Hameiri 2007, see also Esty et al. 1995, Howard 2008, Milliken and Krause 2002, Reno
2003, Rotberg 2002, 2004). As Bøås and Jennings (2005) point out, research on state
failure takes into consideration formal elements of the state but it does not take into
account informal configurations beyond the state’s scope. Following a logic thought,
state failure, first and foremost, means the failure of the formal state institutions and
normative behaviours; the failure of the state. The substance (ousía) of the state, in
Aristotle’s terms, has failed, but the matter (hylé), the citizens and members of the state,
remain. Hence, the unit of analysis in the research should be the matter of the state, that
remains, and not the substance, that fades out. It is nevertheless important to evince
attention to the state, which is an end and a mean, as a symbolic representation of
structural organization. The term state failure already implies a normative idea, and the
aim is to move the state from failure to strength. Notwithstanding this, the
constituencies of the state remain and are the best measure of the failure. A state does
not fail in a vacuum, it fails insofar somebody is suffering the failure. It does fail to its
citizens, and they have to cope with the consequences. Therefore, the understanding of
the consequences of state failure should stem from the human point of view rather than
from a symbolic and normative vision of the state.

Most of the literature on state failure points out that a state fails when it achieves
neither legitimacy nor efficiency. This definition is based on the work of Goldstone
(2008) but for the aim of this paper it needs further clarification. Goldstone (2008)
argues that a stable state must be effective and legitimate in order to remain stable; by

-8-
the same token, a failed state can be identified when it lacks effectiveness and
legitimacy. These two characteristics are necessary for a stable state, and the lack of
them will prompt state failure. Hence, it becomes relevant to understand these two
characteristics in order to explore the causal mechanisms that, stemming from the lack
of legitimacy and efficiency, will affect neighbouring countries. In a nutshell, only by
knowing the causes of state failure will we be able to understand the effects of state
failure on neighbouring countries. Goldstone (2008) observes both legitimacy and
efficiency from an institutional point of view, lacking a more human-based approach
towards failure (Bøås and Jennings 2005). For instance, Rwanda during the genocide in
1994 will not be considered, under Goldstone‘s understanding, a failed state since it was
able to achieve a relatively high degree of effectiveness in carrying out its objectives
and these objectives were carried out with the compliance of some segments of the
population. Notwithstanding, Rwanda was clearly failing for most of its population in
1994, since the basic political goods were not being delivered.

States achieve effectiveness and legitimacy insofar they are able to deliver
political goods to its constituencies. Holsti (1996) argues that states achieve legitimacy
based on their performance delivering political goods2. Moreover, those receiving
political goods must observe the state as the entitled entity in charge of the delivery
process and they will do so if they are receiving political goods. As Rotberg points out,
political goods are ‘those intangible […] claims that citizens […] make on states’
(2004:2). Hence, when a state is able to receive input from its citizens – the claims they
make on the state – and is capable and willing to transform these claims into output, it
will achieve effectiveness and be observed as legitimate by its constituencies. The lack
of capability and/or willingness will render the state ineffective and illegitimate, and
ultimately, failed. Nonetheless, the deliverance of political goods must be understood
from a human perspective and not solely from an institutional point of view. Efficiency
and legitimacy, from a human-based perspective, imply three elements: the capability to
deliver security – monopoly over the use of legitimate violence from the human security
perspective –, representation – the inclusion of all identities within the symbolic image
of the state – and wealth – the entitlement of the population and the enforcement of the

2
Holsti (1996) recognizes the religious-based legitimacy of older forms of government, still beheld in
some countries such as Saudi Arabia; nevertheless, he argues that performance-based legitimacy
nowadays also apply to those countries.

-9-
normative economic rules (Rotberg 2004, Reno 2003, Milliken and Krause 2002, Cf.
Tilly 1990, Sen 1999, Bøås and Jennings 2005)3. The lack of these three elements is the
most acute consequences of state failure, and if we are to research any effect on
neighbouring countries, it must stem from them. Thus, it becomes crucial to fully
understand the nature of security, representation and wealth, and the consequences of
their absence. An absence of any of these factors can affect a neighbouring country in a
distinctive way.

Rotberg (2002, 2004) and Reno (2003), among others, argue that security is the
principal political good that states should deliver to its citizens4. Security is understood
by these authors under the conceptualization created by the seminal work of Max Weber
(1958). States, argued Weber (1958), can be considered as such only if they are able to
wield a monopoly on violence with the legitimacy of its constituencies. The exercise of
this monopoly within concrete borders is what confers stateness to the state. The idea
Weber proposed portrays the state as the sole source of legitimate use of violence. In
case any other entity, e.g. armed rebel groups, is able to exercise violence within the
established borders of the country and without being counteracted, the state is not
fulfilling its requirements as a state – or more in line with Rotberg and Reno’s thoughts
– it is a failed state. This understanding of the state entails a vision in which the
coercive forces are the core of state creation (Cf. Tilly 1990). War-making, and the
entities in charge of war-making, create states, as proposed by Tilly (1990).
Nevertheless, the understanding of legitimate monopoly on violence used throughout
this paper goes a step beyond Weber’s premise. A state is delivering security to its
citizens insofar those citizens can feel safe from other actors and from the state itself.
Unbridled repression and state violence can be used to securitize a regime, but it does
not securitize its citizens. The idea of legitimate monopoly over the use of violence
acquires new dimensions if approached from a human security perspective (Bøås and
Jennings 2005). A state that preys upon its citizens can be labelled as secure using an

3
Holsti (1996) distinguishes between vertical and horizontal legitimacy, but both concepts are included
within this theoretical definition of state failure. Vertical legitimacy implies the securitization of citizens
and the creation of wealth. Horizontal legitimacy, on the other hand, implies the inclusion of all ethnic
groups within the idea of the state. In addition, there is, argues Holsti (1996), and international dimension
of legitimacy. This idea does not need further exploration due to the nature and approach of this paper.
4
The core idea behind the concept of security as a political good is found in the classic sociological
literature. Hobbes, Locke and Rousseau conceptualized the idea of social contract. The population, in an
attempt to avoid an insecure scenario, relinquished some of their individual rights. In exchange, the state,
as the holder of those rights, provides security and order to its citizens.

- 10 -
institutionalist point of view. North Korea might be the best example. Nevertheless, it is
necessary to enlargen the scope of our understanding of security including a more
human-focused point of view. For the aim of this paper, the deliverance of security will
thus be understood from the human perspective. Hence security will be delivered
insofar citizens are being threatened neither by rebel groups nor by the state itself.

In addition to security, Milliken and Krause (2002) argue that the state should
‘represent the symbolic identity of its citizens’ (2002:757). This idea becomes
problematic when applied to post-colonialist countries due to the dynamic state
formation process which has occurred from colonisation onwards. Western states
underwent a normative process of state (Milliken and Krause 2002 Cf. Tilly 1990).
Borders and nationalities were defined through war-making and the constant
reconfiguration of contiguous boundaries in order to accommodate different claims and
state-making objectives. The reality in the post-colonial states was, nevertheless, starkly
different. As Clapham (2004) points out, Africa and Asia were defined by islands of
authority without contiguous borders. Executive power was exercised insofar the realm
of the ‘island’ was present. Beyond this realm, authority was gradually demising. Those
territories – and usually ethnic groups – outside these realms of authority were
unrepresented in the core. Colonization overturned this systemic configuration, creating
artificial contiguous states where some ethnic groups were de jure divided. The lack of
representation was curtailed during the first stages of independence through the
revolutionary legitimacy (Schutz 1995:110). Although not all segments of society were
properly represented, the successful revolutionary movements achieved comprehensive
representation and legitimacy through the mere overthrow of colonial powers. When
representation was not expanded towards the whole society, the broad support they
enjoyed plummeted. Rotberg (2004) argues that the dwindling of the revolutionary
legitimacy came when citizens perceived that their government was being used for the
benefit of only certain segments of society.

Besides security and representation, creation of wealth stands as the last political
good that the state should deliver (Milliken and Krasue 2002). The inherent weakness of
post-colonial states hampers the delivery of wealth. As noted by Rotberg (2004) and
Van de Walle (2004), this weakness is a man-made result of patronage-based economic
networks and faulty macroeconomic policies. In order to comprehend the relationship of

- 11 -
effectiveness and legitimacy in the creation of wealth and state failure it is necessary to
understand the role of the state in the creation of a legal framework overseeing
economic relations. To protect its citizens from lawless scenarios, the state must
establish a normative and legal framework, also applicable to economic forces (Milliken
and Krasue 2002). The non-fulfilment of this framework, allowed by the government,
hinders any prospects of effective economic development and redistribution. Van de
Walle (2004) identifies this shortcoming as a structural factor of state failure. Problems
of governance generate corruption, patronage, and the rise of informal networks of
production and exchange, hampering the already low extractive capacity of these states
even more (Bøås and Jennings 2005, Van de Walle 2004). Low extractive capacity and
the patronage-based networks of action hampered the ability and willingness of any
given state to enhance the dynamic development of its population and its entitlements
(Sen 1999)5.

2.2 Theoretical construct

In order to answer the question of why and how failed states affect neighbouring
countries, the paper will be lead by the hypothesis which states, based on the existing
theoretical inferences and stemming from them, that failed states spread conflict-
enhancing mechanism through the establishment of informal networks between the
failed state and the neighbouring state. As stated in Section 1, the hypothesis is partly
based on the conclusions of Iqbal and Starr (2008). They demonstrate, through a
statistical model, that a failed state does not spread more failure, but rather conflict-
enhancing mechanisms (Iqbal and Starr 2008). The outbreak of conflict will be, thus,
the dependent variable of this research. Outbreak of conflict is understood as the
moment in which a struggle of any kind becomes violent and reaches 25 battle-related
deaths, a criteria used by the Uppsala Conflict Data Programme (UCDP) (Eck 2005).

The causal mechanism explored in the following sections is based on the

particularities of state failure, and will be explored in the three elements characterizing

5
Sen (1999) defines development as the acquirement of entitlements. For Sen (1999) real development
means the enhancement of the freedoms of the population through the increase of their possibilities, such
as access to education, to health, to clean water or shelter. This approach entails an understanding of
development not only as macroeconomic wealth that can be measured using GDP data, but also as a
comprehensive enlargement of people’s entitlement to political goods of economic nature.

- 12 -
the phenomenon, namely, the lack of security, representation and wealth. The presence
of informal networks within these three elements will transform them into conflict-
enhancing mechanisms on the other side of the border. The consequences of these three
elements suggest a causal path towards neighbouring countries. In this way,
neighbouring countries will observe how the likelihood of conflict in their territories
increases when informal networks are present. It is worth noting that conflict-enhancing
mechanisms do not create conflict by themselves but rather enhance the reasons,
resources and resolve of armed groups (Ohlson 2008).

Since the understanding of these concepts stems from a human-based approach,

it is possible to recognise causal mechanisms beyond the institutional realm. The
consequence of these absences – security, representation and wealth – can cross borders
since they are not constrained by the institutional and normative behaviour of
international norms of states’ vis-à-vis relations. These three unique and distinctive
characteristics of state failure generate three conflict-enhancing mechanisms that will
affect neighbouring countries when informal networks are present; the three I’s:
Insurgency, Identity and Illicit trade. These three elements will be used as independent
variables for this paper. Lack of security will enhance the possibility of insurgency on
both sides of the border. Absence of representation will foster identity quarrels on both
sides. Lack of wealth will open a space for illicit trade, which can be used to finance,
sustain and prolong an armed struggle.

Insurgency
First, lack of security in a failed state can spread conflict-related effects beyond its
borders through the actions of uncontrolled armed groups. The informalization of
relationships and lack of enforcing agencies opens a multidirectional opportunity for
armed groups and non-state actors to expand beyond the given borders of a state
(Buhaug and Gleditsch 2008). The opportunity is multidirectional since it can be steered
in two directions; (i) groups from the failed state carrying out cross-border activities or
even settling down in a neighbouring country and (ii) groups from the neighbouring
country using the failed state as a sanctuary from where to carry out their attacks. Not
encountering counter-activities and being the executants of the monopoly of violence –
without legitimacy – blurred the importance of international borders and international
law and the possibility to enforce them upon armed groups. In the first direction of the

- 13 -
causal mechanism, an armed group already controlling a piece of land will find
incentives in enlarging their scope of action to other regions, regardless of international
borders. Cross-border activities and the destabilization of neighbouring countries can be
beneficial for an armed group’s goal mostly when it is driven by greed, but also when it
only waves grievance-based reasons. If an armed group is able to expand beyond the
borders of the original state, it can have access to new resources, gain control over new
population groups, and extend its informal network of contacts and supplies. State
failure, and the lack of security that it entails, open a window of opportunity for these
kinds of cross-border activities. Nevertheless, groups require connections on the other
side of the border in order to expand successfully.

In the other direction of the causal mechanism, state failure on the other side of
the border decreases the relative mobilization cost of a possible insurgent movement. As
showed by Salehyan (2007), is extremely useful for rebel groups to maintain operational
bases on the other side of their country’s border, and having sanctuaries increases the
chances of success. The state they are fighting will face a higher cost if it pursues the
fighting beyond its borders – thus entering other state’s soil and violating international
law. The fact that the hosting state has failed enormously decreases the cost related with
establishing bases, therefore, it will increase the rebel group’s capabilities and strength,
providing them a sanctuary far from the reach of the government they are fighting.
Failed states, thus, increase the prospects of civil conflict in neighbouring countries
widening the capabilities of armed groups to use their soil and also exporting their own
armed groups thirsty for resources, networks and informal channels. The lack of
security is what allows these cross-border activities by armed groups, since the rebel
groups will not encounter any legitimate force to curtail their activities. In addition to
the dangers of expansion of rebel groups’ activities, the issue of unoccupied soldiers
may increase the possibilities of conflict-enhancing effects crossing borders. Many
fighters are born and have grown up in war-like environments and the lack of activity –
which entails lack of wages and looting possibilities– is an incentive to spark a war in
the region. The situation they used to find themselves –living in refugee camps, or
maintaining weak or even nomad bases in forest regions– increases the incentives for
mobilization.

- 14 -
 In summary, insurgency problems enhance conflict in the neighbouring country
when armed groups use the failed state as a safe haven or armed groups from the failed
state cross the border in order to spread their power, influence and sometimes their
revenues.

Identity
Second, failed states lack representation and cross-border ethnic ties can transfer
conflict-related activities and grievances to the other side of the border. Greed,
grievances or fear can prompt an ethnic security dilemma (Kasfir 2004 Cf. Lake and
Rothchild 1998). In Africa and Asia several different ethnic groups have been divided
during arbitrary delineation of state borders, which potentially expands the security
dilemma beyond the border of the failed state. The presence of kin ethnic groups on the
other side of the border might push ethnically related armed groups in the failed state to
find a safe haven and/or expand their activities towards neighbouring states. The
presence of armed groups in any country is usually not viewed upon favourably by the
authorities, unless these groups are actually been financed by the government. When the
authorities do not control hostile armed groups using their territory, conflict can easily
emerge in a fight for authority.

In addition, state failure implies population movements, affecting the ethnic

equilibrium of several regions. In these regions ethnic equilibrium is extremely fragile
due to the arbitrary delineation of borders stated before, and state failure may break
down the fragile equilibrium. This breakdown will mean movement of population, shifts
in ethnic presence and reconfiguration of relations between ethnic groups. Another
danger is refugee movements. As showed by Glenditsch and Salehyan (2006), refugee
flows correlate with the spread of civil war. State failure and its consequences are likely
to produce refugee flows. These refugee flows will also transport hidden rebels,
criminal gangs and will mean a shift in population percentages, all these fostering the
likelihood of civil conflict.

Summing up, identity problems can be used by armed groups to increase the
possibilities of conflict when informal ethnic ties are present on both sides of the border
and are not limited by the presence of the internationally recognized boundary.

- 15 -
Illicit trade
Third, lack of wealth can entail conflict-related effects on neighbouring countries
through the informalization and illegalization of economic activities when armed groups
take advantage of the situation. The dwindling character of the state’s effectiveness
opens a window of opportunity for the rise of an informal and illegal character of
human economic relationships. Lack of effectiveness and legitimacy means a recession
of the state (Bøås and Jennings 2005). The state is entitled and supposed to deliver the
political goods named above, thereby; the lack of delivering entails a clear recession of
the state and its symbolic role. That space cannot remain unfilled, thus the networks of
relationship among the population will become informal and fill up the space left empty
by the state. The informalization of the economic and social relations leads to the
illegalization of activities for the profit of those controlling the activities, usually,
violently. Without enforcing authorities and laws, illegal trade becomes more profitable
and secure than legal normative economic activities. Goods such as weapons, timber,
diamonds, cocoa or different drugs can then be use to finance an armed struggle. An
armed group on the other side of the border can, therefore, use these informal networks
in order to fund its activities and/or sustain their war against their government.
Moreover, armed groups in the failed state might find incentives to expand their illegal
networks towards neighbouring countries. Thereby, the informalization of economic
networks opens a window of opportunity for rebel groups on both sides of the border to
finance their activities. In addition, as stated by Skaperdas (2001) and Sanderson
(2004), the relationship between criminal networks and insurgency movements is
aggravated when the state is unable or unwilling to control these links, thus facilitating
the intentions of armed groups even more.

In summary, illicit trade will enhance the prospects of conflict when armed
groups and criminal groups utilize the informal networks of the failed state to fund their
struggle and activities.

- 16 -
 3. Methodology and Case Selection

Aiming at answering the research question driving this paper, it becomes necessary to
establish a clear research strategy, clarifying the methodology and the reasons behind
the selection of the cases that will be used. To carry out the research, a qualitative
approach has been selected based on two reasons. Firstly, any research regarding state
failure becomes problematic due to the small number of available cases. This difficulty,
considering the amount of time and resources available for this paper, rules out the
possibility of a large-N study. Some large-N studies regarding state failure have been
carried out, more notably Howard (2008) and Iqbal and Starr (2008). For the interest of
this paper – the influence of state failure on neighbouring countries – Iqbal and Starr
(2008) already achieved the benefits of a large-N study; however, they lack the
systematic explanations about causality that only small-N studies can achieve (George
and Bennett 2005). Secondly, the causal mechanisms carefully described in the
theoretical framework can only be explored through the use of intensive small-N studies
(George and Bennett 2005). The epistemology behind large-N and small-N research
remains similar; nevertheless, the lack of systematic research in this particular field
leaves the researcher with the need to deeply explore causal mechanisms rather than
general non-causal trends – which, at the same time, was already done by Iqbal and
Starr (2008). The most accurate way to theoretically explore those causal mechanisms is
with a small-N study (George and Bennett 2005). It is worth noting that the election of
small-N studies implies the understanding of the trade-off between validity and
generalization. The case selection tries to reduce the problem of generalization without
affecting the validity of the research; nevertheless, the inherent weakness of small-N
studies when it comes to generalization will be always present to a certain degree
(George and Bennett 2005).

Small-N or case studies allow the researcher to find causal explanations and
paths in the cases at hand, with some generalization power. This technique has
traditionally been used under the umbrella of Mill’s methods of agreement and
difference, but nevertheless, the requirements for successful use of Mill’s methodology
are hardly possible to replicate with real social phenomena6 (George and Bennett 2005).

6
Mill’s methodology seeks to replicate the experimental epistemology on social events. The ‘method of
difference’ uses cases with different outcome on the dependent variable and the same outcome on all the
independent variables except one. The ‘method of agreement’ uses cases with the same outcome on the

- 17 -
Within the research of failed states, in which few cases can currently be used, Mill’s
method becomes almost impossible. In order to overcome the difficulties posed by the
experimental approach in this particular case, within case analysis can be used to
uncover the causal mechanism behind a particular phenomenon when driven by theory
(George and Bennett 2005). A perfect methodological tool to uncover the particularities
of every case and understand the causal mechanisms behind the relationships of the
variables beyond the scope of Mill’s methods can be found within case analysis (George
and Bennett 2005). In order to capture the nuances of each case, and avoid spuriousness,
process-tracing will be applied on the empirics.

According to George and Bennett (2005), process-tracing analysis can be

successfully used to investigate theoretically-driven hypothesis regarding causal
mechanisms. Process-tracing will allow light to be shed upon the causal mechanism
linking the outcome of the dependent variables and the independent variables (George
and Bennett 2005). George and Bennett (2005) point out four possible kinds of process-
tracing analysis, ranging from the pure descriptive historical narrative, in which the
author seeks to explain how an event came about, to an abstract general explanation of a
particular phenomenon on a macro-level. The objective of the research will decide the
kind of process-tracing that should be used; nevertheless, when several cases are
involved, it becomes problematic to draw the line between theory testing and theory
development due to the multidirectional process between theory and data (George and
McKeown 1985).

The use of case analysis and process tracing in the selected cases laid the
foundations for a comparative analysis of the results of each particular case. In order to
carry out the comparison accurately, the cases should be subject of the same inquiries,
structuring the comparison through them. The organization of the data must answer to a
certain epistemological intention driven by theory and common to all the cases (George
and McKeown 1985). In addition, theory must guide the path of the researcher,
selecting and labelling the importance of the information according to its relation with
the theory. To build, develop and test the strength of a theory, it is important to focus on

dependent variable and different outcome on all the independent variables but one. Both methods aim to
discover which independent variables have an effect on the dependent variable (George and Bennett
2005)

- 18 -
the factors that the theory identifies as crucial (George and Bennett 2005, George and
McKeown 1985). In summary, the comparison between the cases must be structured
and focused. It must ask the same questions and be driven by theory (George and
McKeown 1985).

Two failed states have been selected for this paper, Liberia and Afghanistan.
Considerable amounts of literature have been devoted to the identification of failed
states, without a concrete and universal classification of what a failed state is and what it
is not7. In the theory-section the underpinnings for the identification of a failed state has
been laid. In addition, every study of state failure has included these two cases as clear
examples (see Rotberg 2004, Howard 2008, Iqbal and Starr 2008). Nonetheless, both
Liberia and Afghanistan border several countries; therefore it is important to select
some of the bordering countries in order to create a feasible research design. In this
way, four cases will be selected by creating two pairs for each failed state. This strategy
will enhance the explanatory power of the within country analysis and at the same time,
increased the reliability of the structured focused comparison. The selection of the four
cases answers to two specific research requirements. Firstly, the number of cases should
be bigger than the number of variables. King et al (1994) state that it is impossible to
test a hypothesis if there are more variables to examine than cases to observe. In
addition, parsimoniousness is easily achieved when the research contemplates fewer
variables than cases. Secondly, the variation in the dependent variable should be the
driven-force behind the selection of cases (George and Bennett 2005). The keystone of
the research question is precisely the different outcome in the dependent variable
observed in the cases.

Therefore, the influence of Liberia’s failure will be studied in Sierra Leone and
in Guinea, being these the first two cases. For the Afghan’s failure effects will be
measured in Pakistan and Tajikistan, which are the other two cases. The selection of the
four cases, as stated above, is not happenstance. In two cases, Liberia-Sierra Leone and
Afghanistan-Pakistan, the influence of the failed state on the neighbouring country has
helped to create conflict or at least it was one of the main enabling forces behind

7
Probably one of the more accurate classifications is done by the Fund for Peace. The classification is
based on the score of all the countries in several categories. For more information, visit
http://www.fundforpeace.org/web/index.php?option=com_content&task=view&id=99&Itemid=140
(accessed March 6th 2009)

- 19 -
conflict. It can be argued that the conflict in Pakistan and Sierra Leone would not have
happened without the influence of the failed state. On the other two cases, Liberia-
Guinea and Afghanistan-Tajikistan, the influence of the failed state does not seem that
clear and no conflict has broken out in relationship with the failure of Liberia or
Afghanistan8. The election of these four pairs is made in order to test the strength of the
theoretical framework and increase the generalization capability of the research.

The selection of Liberia and Afghanistan and its specific pairs answers to the
necessity of using relevant cases and avoid spuriousness as much as possible. These two
cases, and the four pairs that will be used, are relevant for the study of the effects of
state failure for two reasons. On one hand, Liberia and Afghanistan are two completely
different countries with no common cultural or political characteristics. They share
neither religion nor a colonial history. Therefore, the robustness of the theory will
increase if the hypothesis holds for four completely different cases. In addition,
spuriousness based on cultural or religious elements can be avoided. On the other hand,
selecting four cases from different continents avoids bias in selection and relatively
increases the generalization power of the findings (George and Bennett 2005). Both
cases will be studied for a specific period of time, namely, since the failure of the
countries (see the empirics below).

The research question driving this paper cannot be answered through any
existing theory, which implies a theory building and development process. As noted by
Lijphart (1971), it is important to base the building process of the theory on existing
material, in order to discern possible causal mechanism among the massive amount of
interfering factors in a given phenomenon. The theoretical construct guiding this paper
has been created according to the possible paths pointed out by several researchers. The
theoretical framework aims to replicate what George and Bennett (2005:235)
conceptualize as ‘typology theories’. According to George and Bennett (2005),
typological theories aim to imitate the behaviour of studies in medicine conducted when
identifying a pathology, trying to ‘specify the pathways through which particular types
relate to specified outcomes’ (2005:235). A typological theory must identify and

8
It is important to note the particularity of the Afghanistan-Tajikistan case. The Tajik civil war did not
break out as a consequence of the Afghan failure, therefore, the variation on the dependent variable –
conflict in the neighbouring country due to the failure of the country- still remains within the same
category than Guinea. This issue is more explained in the Afghanistan-Tajikistan section.

- 20 -
categorize the independent variables that will be used to explore the particularities of
each case, understand the individual behaviour of the variables and, at the same time,
‘provide(s) […] contingent generalizations on how […] they behave in specified
conjunctions’ (George and Bennett 2005:235).

The data used for this paper included several reliable sources. In order to avoid
biased sources as much as possible, only well-known university resources – such as the
Uppsala Conflict Data Base – have been used to obtain information, while internet
resources and biased journalism have been limited as much as possible. A fundamental
difficulty exists when exploring informal networks and illegal activities, and therefore
all the information regarding these factors might be biased, incorrect and/or incomplete.
Luckily, the recent interest in failed states, and the enormous dimension of some of the
more profitable business carried out in Afghanistan and Liberia, namely, diamonds and
poppy cultivation, provide a fairly big amount of information on these illegal activities.
Still, governmental reports have been avoided, especially in the case of Afghanistan, in
order to avert any possible official propaganda. Most of the information has been
obtained from University Press-type books, which remain more trustworthy than other
publications, enhancing the impartiality of the findings. It is important to note that for
the two pairs where conflict did not occur, Liberia-Guinea and Afghanistan-Tajikistan,
the task of gather information has definitely been more difficult than for the other two
pairs, Liberia-Sierra Leone and Afghanistan-Pakistan. This difficulty explains the less
detailed information in the empirics of these two pairs. Any remaining mistake or
misinformation is solely the researcher’s fault.

- 21 -
 4. Liberia’s Failure and its Influence in Sierra Leone and Guinea

4.1 Liberia Failure: The three I’s origin

Liberia has never been controlled by a colonial power. It was funded by freed
slaves from the United States and remained ruled by their descendants until 1980, when
a military coup partially expelled the Americo-Liberians from power. In 1989 the
invasion of Liberia by an armed group commanded by Charles Taylor initiated a long
civil war with two distinct phases. From 1989 until 1995, the forces of Taylor’s
National Patriotic Front of Liberia (NPFL) fought against the weak and uncontrolled
Armed Forces of Liberia (AFL), the Economic Community of West African States’
peacekeeping mission (ECOMOG), the United Liberation Movement for Democracy in
Liberia (ULIMO) and different factions of each group9 (UCDP 2009b, Williams 2002).
After several years of warlordism in Liberia, the parties signed an agreement and
elections took place in 1997 (Reno 1998, Williams 2002).

Charles Taylor secured victory in the elections with the slogan ‘He killed my ma, he
killed my pa, but I will vote for him’. Nevertheless, warlordism only increased with the
victory of Charles Taylor (Williams 2002). The constant involvement of Taylor in the
affairs of neighbouring countries resulted in the creation of the Liberians United for
Reconciliation and Democracy (LURD) in Guinea, and later, the Movement for
Democracy in Liberia (MODEL), which came as a split of the LURD (UCDP 2009b).
Conflict broke out again until 2003, when Taylor stepped down from the presidency,
leaving for exile in Nigeria. A new transitional government was set up, and in 2005
relatively free and fair elections were held. The first female African head of state was
elected.

Notwithstanding what is stated above, Liberia failed as a state years before the
war broke out on Christmas Eve 1989. Patronage networks based on a particular ethnic
affiliation – the Kranh people – prompted the failure of the state. Political goods were
not being delivered anymore during the 80s, when Master Sergeant Samuel K. Doe
overthrew President Tolbert in a violent coup (Williams 2002). During the 80s the small

9
It is not the aim of this paper to identify all the different armed groups fighting in Liberia, for more
information about all the factions, see UCDP (2009b) and Williams (2000)

- 22 -
bureaucracy and the state institutions were dismantled and the, by then, small flow of
political goods streaming to the population disappeared. Doe sustained his regime with
the help of coercion and patronage networks, controlled by the Kranh minority (Reno
1998). Since the invasion of the NPFL in 1989 until the elections in 2005, the country
was immersed in further chaos and failure. Between 1997 and 2003, Charles Taylor, the
leader of the NPFL, held the presidency of Liberia but the country was still deeply
immersed in failure (Reno 1998). Taylor maintained the warlike structure, stressing the
failure and redirecting all the state goods towards him (Reno 1998). No improvement
was seen by the population, and Liberia was nicknamed ‘Taylorland’ by some authors
due to the derail of the state towards Taylor’s interests (Reno 1998).

The Liberian state did not deliver any political good to its citizens during more
than twenty years. Neither security, nor representation or wealth were being delivered
by the state through normal channels. Some groups enjoyed protection, representation,
or wealth during limited periods of time, but the citizens did not enjoy them whatsoever.
The state, barely present throughout this period, was unable or unwilling to provide
protection to its citizens, handing the control of the security to armed groups and private
violence-entrepreneurs (Reno 1998, Williams 2002). During Doe’s regime, the Krahn
minority and the Americo-Liberian descendants enjoyed overrepresentation in the
state’s crumbling institutions (Reno 1998). Alongside the Mandingo people, the Krahn
were targeted by the NPFL during the first stages of the quarrel (Gberie 2005). In the
second stage of the war, the Mandingo constituted the bulk of the LURD, probably as a
response to the prosecution of the NPFL and the Government of Liberia after Taylor
achieved the presidency.

The economic collapse in Liberia prompted a frenetic activity of the informal

networks in Liberia. Some of these networks enlarged their scope, affecting
neighbouring countries such as Sierra Leone. The NPFL, in acute need of funding for
their struggle, designed a private economy outside the bureaucracy, based on informal
and illegal transactions (Reno 1998). The informal networks were gradually merging
into armed groups, mostly with the NPFL, becoming an integral part of the groups and a
crucial element of the state failure and gaining control of the Liberian economy (Reno
1998). These groups, in a scenario of state failure, found the only channel to carry out
their business under the armed groups’ patronage, while still being free enough to

- 23 -
maintain the informal and illegal character of it. These characteristics did not stop when
the NPFL achieved power in 1997, instead they became deeper and more rooted in the
economy, completely collapsing the formal economy in Liberia and neglecting any kind
of political good to those outside the informal networks.

4.2 Liberia – Sierra Leone

Sierra Leone suffered a civil war between 1991 and 2002, remembered mostly
for its apparent lack of sense and for its brutality. In 1991 the Revolutionary United
Front (RUF), a group based on Liberia, launched an attack into Sierra Leonean territory
with less than 100 soldiers. The attack was directed against the autocratic All People’s
Congress (APC) regime. One year later, amid the failure of the RUF to fulfil their
objectives, a military coup took place in Freetown, overthrowing the APC regime
(Gberie 2005). The new junta, the National Provisional Ruling Council (NPRC),
increased the resources dedicated to fight the RUF (Nilsson 2008). Nevertheless, the
RUF was able to made an spectacular comeback in 1995 thanks to the ‘sobels’,
members of the Sierra Leone Army (SLA) that fought for the RUF during the night
(Nilsson 2008).

Having lost command of the armed forces, the NPRC organized democratic
elections in 1996. The RUF did not participate, and the victory went to the Sierra Leone
People’s Party (SLPP). A year after the elections the new government was overthrown
by a military coup due to the relationship between the SLA and the new government
(Gberie 2005). The new government, the Armed Forces Revolutionary Council (AFRC)
formed a unity government with the RUF, but soon faced the opposition of the
Economic Community of West African States (ECOWAS). The fierce battle ended up
with the SLPP back in power and a fresh attack of the RUF/AFRC that devastated
Freetown (Gberie 2005, Nilsson 2008). In July 1999 all the parties signed the Lomé
Accord. The UN began a DDR process in October 1999 that encountered stiff
opposition from the RUF since it threatened their control of diamonds mines. After the
intervention of British troops the DDR process was carried out successfully, putting an
end to the civil war (Gberie 2005, Nilsson 2008)

- 24 -
Insurgency
The civil war in Sierra Leone had its starting point in north-west Liberia. The original
command of the RUF, Foday Sankoh, Abu Kanu and Rashid Mansaray, took advantage
of their Libyan-created links with Charles Taylor to sow the seed of their struggle in
10
Liberia (Abdullah 2004) . At the end of the 80s several would-be revolutionaries
joined the pan-African path established by Muammar al-Qaddafi in his ‘Green Book’,
spreading the teachings of the book and travelling to Libya in order to acquire military
training (Abdullah 2004, Gberie 2005). Some years later, during Taylor’s tour to find
support for his upcoming struggle, the future commander of the RUF joined the
Liberian adventure11. On Christmas Eve in 1989 Taylor’s forces crossed the Liberia-
Cote d’Ivore border and started the civil war that definitively collapsed the country. The
more collapsed the Liberian state got, with four different groups fighting each other in
1990, the bitterer the effects on Sierra Leone were12. The NPFL carried out several
incursions in Sierra Leone, but in March 1991, the RUF made its appearance with a
100-strong man invasion of Sierra Leone at Bomaru in the Kailahun District, from
NPFL-controlled territory in Liberia13.

The group that entered Sierra Leone in March 1991 was composed mostly of
non-Sierra Leonean fighters, although the core of the group was Sankoh, Kanu and
Mansaray’s leadership (Nilsson 2008, Richards 1996)14. The remaining forces were
Sierra Leonean economic immigrants and soldiers on loan from NPFL (Nilsson 2008).
Despite the composition of the RUF during the first attack on Sierra Leonean soil, the
armed group can be traced back to the anti-regime movements in Sierra Leone during
the 70s and 80s. These movements were influenced by Qaddafi’s Green Book, and
assembled as student associations, mostly the PANAFU (Abdullah 2004). The student-

10
Charles Taylor and Foday Sankoh met in Ghana in 1987, but strengthen their link in Libya in 1988,
while pursuing ideological and military education provided by the Libyan regime and his foreign policy
of the ’80: spreads revolution (Abdullah 2004). Many coups and wars in Africa in the ’80 and ’90 have a
link with Libya (Gberie 2005, Ellis 1999)
11
Charles Taylor travelled different West African countries in order to find support for its struggle,
including Sierra Leone (Abdullah 2004)
12
In 1990 four groups controlled Liberia. Taylor’s NPFL, Johnson’s INPFL, ECOWAS forces and the
remaining of the AFL.
13
This statement is contradicted by Gleditsch et al (2008:485). They argue that the NPFL never attacked
Sierra Leone, but rather used the RUF as a proxy-army. Nevertheless, attacks from Liberia were already
reported at the end of 1990, before the ‘presentation’ of the RUF.
14
There is some polemic about the exact composition of the 100-man strong force that entered Sierra
Leone in 1991, mostly about the social origin of the Sierra Leonean immigrants recruited. For more
information see Richard (1996) and Abdullah (2004).

- 25 -
based organizations were gearing up rebellions against the neo-patrimonial regime in
Sierra Leone alongside lumpen and criminal ranks but the crackdown on these groups in
1985 sent several future rebels into exile 15 (Rashid 2004). The crackdown in 1985 also
united the lumpen groups with the students. This union did not bring lumpen youth
within the educated ranks of student unions, but allowed them to soak in the
revolutionary message that was about to be promoted by the RUF a few years later
(Rashid 2004)16. In addition to this, the leaders of the movement left the country,
radicalized their message and searched for safe havens from where to carry out their
revolution, finding a helping hand in the Liberian failure.

As the struggle consolidated in Sierra Leonean soil during the first months of the
quarrel, the RUF ranks were becoming more and more Sierra Leonean in character, but
the link with Liberia was not broken. Gradually, the RUF composition was mirroring
the original anti-government movements, combining lumpen and educated members.
With the time also the lumpen ranks within the SLA also joined the RUF as ‘sobels’
(Nilsson 2008). The RUF became even more Sierra Leonean when Charles Taylor
decided to withdrawal part of its troops from Sierra Leone in order to counterattack the
United Liberation Movement of Liberia for Democracy (ULIMO) (Nilsson 2008). This
decision temporally dried up Liberian supplies and the RUF survived only by the
desired of maintain the conflict expressed by Freetown government through the supply
of weapons to the RUF while they were fighting them (Nilsson 2008).

Yet, the relation and use of Liberia’s failure by the RUF was not limited to the
initial months of the struggle. The RUF struggle waned and waxed with different
degrees between 1991 and 2002, but not its beneficial use of the havoc in Liberia.
Although the RUF was poorly equipped and most of its ranks were abducted, its
struggle was made possible by the economic and logistical support anchored in Liberia.
RUF soldiers were trained in Liberia under the supervision of the NPFL (UN 2000).
Liberia was also a safe haven for RUF-members’ relatives, and even hospitals in Liberia
were used to treat war-related wounds (UN 2000). The main supply line of arms and

15
The unrest in 1985 was the peak of several years of Pan-African leftist movements fighting against
Siaka Stevens’ regime.
16
It is important to note that the main student’s union, PANAFU, refused in 1987 a call for recruitment
coming from the future RUF.

- 26 -
military-related equipment was coming from Liberia17. A panel of experts described the
links between Liberia and the RUF as ‘unequivocal and overwhelming’ when it comes
to arms trafficking (UN 2000:8, see illicit trade below). A lax air legislation and the
near-collapsed state of the Liberian state between 1989 and 2002 allowed heavy air-
traffic in the region, mostly used to sustain RUF’s struggle (UN 2000)18. Nevertheless,
the RUF’s use of Liberia spans all levels, ranging from the use of their uncontrolled
state apparatus to gain access to weapons, to cross-border ground attacks and retreats.

The clear link between the Sierra Leonean insurgency and Liberia needs further
clarification from the other side of the border, Liberia. During the first months and years
of their struggle the NPFL encountered fierce opposition from different groups, among
them, ULIMO and the ECOWAS troops. These two groups enjoyed the support of
Freetown’s government; therefore, it was in NPFL’s interest to destabilize Sierra Leone,
aiming to halt the support. Moreover, the destabilization of the already-weak border
between the countries would allow the NPFL access to the diamonds fields (Bøås
2001). Although Sierra Leonean in essence, the first RUF group that entered Sierra
Leone was mostly a spin-off of the NPFL seeking resources and destabilization and
turmoil in Freetown (Bøås 2001). The later developments of the RUF noted above
clearly shows the Sierra Leonean character of the RUF.

Identity
The failure in Liberia and the subsequent conflict had a clear ethnic component. Besides
the Americo-Liberian and Afro-Liberian cleavage, the different groups struggling for
power in Liberia relied upon ethnic affiliation. At the outset of the conflict, the Gio and
Mano ethnic groups lined up with Charles Taylor forces, while the Kranh and the
Mandingo joined the government forces. During the first months of the conflict both
groups committed atrocities based on ethnic affiliation (Williams 2002:78). The
influence of Liberia was also felt in Sierra Leone, as some of the ethnic tendencies
protruded the border. The Mandingo, traditionally merchant traders spread throughout
West Africa, were targeted by the RUF during the first stages of the war. This group
was suffering attacks by NPFL forces in Liberia (Gberie 2004:153). The Mandingo

17
Sierra Leone suffered an arms embargo during its civil war, skilfully ignored by several arms traders
and even nations (UN 2000).
18
Infamous names of the arms-trade business are related with the conflict in Sierra Leone, such as Talal
El-Ndine and Victor Bout. For more information see UN (2000).

- 27 -
were seen as an alien group in Liberia, and the NPFL forces associated their successful
businesses with the Kranh ethnic group.19 (Richard 2005). However, this group was
more integrated in Sierra Leonean society and the attacks were only a consequence of
the presence of Liberian fighters in Sierra Leone (Gberie 2004, Richard 2005). Yet, it
can be argued that the early attacks on the Mandingo population were part of a
comprehensive plan to control the informal economic networks between Liberia and
Sierra Leone, over which the Mandingo people enjoyed high influence (Reno 1998,
Wannenburg 2006).

Yet, the main division in Sierra Leone’s conflict was based on class rather than
ethnicity. Some of the ethnic groups composing the social fabric of Liberia are also
found in Sierra Leone, but their relationship during the war, with the exception of some
periods, was insignificant. The ethnic character of the conflict in Liberia remained
overshadowed by other issues, mostly the warlordist character of the failure of Liberia
(Reno 1998). The case of Sierra Leone followed the same path, witnessing ethnic-
oriented violence during the first stages – probably carried out by the Liberian soldiers
fighting along the RUF (Gberie 2004). Moreover, RUF’s recruiting techniques, based
on abduction, drug use and child force-recruitment, hastily blurred any ethnic line.

Illicit trade
The illicit commerce between Sierra Leone and Liberia flourished soon after the state
failure in Liberia and it was mostly, but not solely, based on illegal smuggling of
diamonds. According to Reno (1998) the Sierra Leonean economy collapsed when the
RUF crossed the Liberian border into Sierra Leone. The revenues flowing to the state
dwindled to extremely low levels as the war advanced, reducing the economy of Sierra
Leone to illegal trade of a handful of goods (Reno 1998)20. Two goods dominated the
illicit economy in Sierra Leone: diamonds and weapons. These two elements were
indeed connected with the failure in Liberia.

19
The Government in Monrovia was based on the Krahn ethnic group.
20
For more information regarding the economic collapse of Sierra Leone, see Reno (1998: 115-123) and
Gberie (2005)

- 28 -
 The illicit trade of diamonds has been seen as the main driving force behind the
civil conflict in Sierra Leone (UN 2000, Smillie et al 2000)21. The illicit traffic of
diamonds would not have been possible without the inestimable assistance of Liberian
informal business networks. Diamonds were smuggled over the Liberian border, and
generally ended up in Liberia. From there, they were moved on to the international
market (UN 2000, Wannenburg 2006). During the initial years of the war, the various
armed groups, especially the RUF, but also the government, used diamonds to fund
their enterprise; however, it was after 1995, when Liberia was mostly controlled by
Charles Taylor, that the illegal trade of diamonds reached significant dimensions (UN
2000, Wannenburg 2006)22. The RUF used diamonds as a way to obtain resources,
combining it with looting and pillage of the towns and villages they conquered. The
diamonds were smuggled into Liberia through Lebanese informal networks. During this
phase, the RUF was in charge of the mining, doing it themselves or using forced labour
– often children – to carry out the task (UN 2000).

Yet, the illicit trade of diamonds became highly profitable for the RUF and for
Liberian informal networks from 1995 onwards, increasing the RUF’s interest in the
diamond-rich Kono region (UN 2000, Smillie et al. 2000). The UN concluded that
digging diamonds became one of the main activities of the RUF, according the
ferocious hold and defence of the diamond-rich areas under their control (UN 2000)23.
In 2000, the United Nations Mission in Sierra Leone (UNAMSIL) began an incursion
into the RUF-controlled Kono region, where the bulk of the diamonds mining illegal
industry was situated. The RUF stopped the UN troops, provoking a major crisis and
forcing the British Army to intervene with force. As an addition to the British
intervention, the UN also toughened the sanctions on Liberia, deemed as the main actor
in the illegal diamond industry (Gberie 2005). The crisis gained global awareness when
the Washington Post published an article shedding light on the links between the illegal

21
The use of diamonds in Sierra Leone has been widely discussed in the academia, popular science and
the policy-making community. It has opened new paths of research – such as the debate over greed vs.
grievance – and has significantly changed many established ideas about incentives and strategic
behaviour of armed groups. For more information on this, see Berdal and Malone (2000), Collier and
Hoeffler (2004) and Smillie et al. (2000)
22
During those years Charles Taylor was trying to create a ‘Greater Liberia’, expanding its activities and
control beyond the Liberian borders.
23
Evidence shows that the RUF did not hold permanent control of the Kono region before 1995, and it
was not until the RUF gained substantial control over this region that the diamond-business reached its
enormous dimensions (Wannenburg 2006).

- 29 -
diamond industry in Sierra Leone and Al-Qaeda (Farah 2001). This link, created by
Lebanese merchants, has lately been extended to also inclube Hizbullah (Gberie
2005)24.

The core of the illegal diamond business was based, as stated before, on the
RUF-Liberian networks, with the indispensable assistance of the Lebanese networks.
RUF’s ranks were responsible for monitoring the diggers, which were often forced into
labour, but also consisted of local diggers that were allowed to dig for themselves one
day per week (UN 2000, Wannenburg 2006). The diamonds were stored in Buedu, and
moved on towards Monrovia from there. The transportation of diamonds was either
made by RUF members, or by utilizing Lebanese connections between Sierra Leone and
Liberia (Gberie 2005, UN 2000, Wannenburg 2006). The community of Lebanese
diamond traders was crucial in the transportation of the diamonds, and the activity
earned them millions of dollars. The UN (2000) pinpoints the importance of Lebanese
groups, controlled by Talal El-Ndine, in the cross-border illegal networks. According to
the UN (2000:37)

‘…One key individual is a wealthy Lebanese businessman named Talal El-Ndine. El-
Ndine is the inner-circle’s paymaster. Liberians fighting in Sierra Leone alongside the
RUF, and those bringing diamonds out of Sierra Leone are paid by him personally. The
pilots and crew of the aircraft used for clandestine shipments into or out of Liberia are
also paid by El-Ndine’.

Nevertheless, it was Charles Taylor who was the director of this perfectly
greased machinery. Liberia was in a state of failure, which allowed Charles Taylor and
his close allies to get rich through the diamond business. Liberia’s own diamond
industry never exported more than $10 million, but between 1994 and 1999 Liberia was
able to turn over more than $2 billion in diamonds, clearly beyond its diamond mining
capabilities (Gberie 2005). Very little of this profit, if any, was redistributed to the
Liberian population. Liberian networks were stationed in the Kono region, monitoring

24
The diamonds business, although dominated by the connection Liberia-RUF, was not exclusive to this
group. Executive Outcome, the private army hired by the Sierra Leonean government to fight the RUF,
was paid through stakes in the mining industry, allowing them to manage the diamond mines under their
control (Reno 1998). Executive Outcome was able to establish a nearly-legal diamond business. Some
evidences also suggest that UNASMIL was involved in the illegal business (Gberie 2005).

- 30 -
the whole process and working in close connection with the RUF and the Lebanese
community to secure the final destination for the profit of the diamonds: Charles Taylor.

Apart from diamonds, illicit trade of weapons experienced a boost during the
Sierra Leonean civil war. Liberia was the main channel of the weapons coming from
Eastern Europe, and the transportation was handled by almost exactly the same
networks that handled the diamonds. The UN (2000) discovered several pieces of
evidence of the constant flow of weapons from Liberia to Sierra Leone. Most of the
weapons, according to the UN (2000), have been transported through at least two more
countries before reaching Liberia and then entering Sierra Leone. The weapons trading
involved several dealers from different parts of the world with clear connections to
Charles Taylor (UN 2000). The illicit trade of weapons – it was illicit since all the
international organizations banned any weapons-related acquisitions to Sierra Leone –
was connected to other illegal activities in Liberia. As an example, Simon Rosenblum,
an Israeli businessman with stakes in the Liberia logging industry, supplied trucks to the
RUF for the transport of weapons (UN 2000). Moreover, Gus Van Kouwenhoven, a
Dutch businessman with connections to the timber industry in Liberia, acted as
‘responsible for the logistical aspects of many of the arms deals’ (UN 2000:43).
Weapons were paid for with cash or diamonds.

4.3 Liberia – Guinea

In 1984 General Lansana Conté seized power after a military coup in Guinea
after the death of Sekou Touré, who had lead Guinea since the country achieved
independence in 1958 (ICG 2005). The new cabinet moved Guinea away from the
socialist path imposed by Touré during his more than 25 years in power. General Conté
promised a democratic reform and economic aperture of Guinea, but clearly failed in
both tasks. During the 90s, after the failure of Liberia, Guinea sided with the anti-NPFL
front, sending troops to defend Monrovia and creating a long-standing enmity with
Charles Taylor. The enmity reached its peak in the year 2000 when fights broke out on
the border between Liberia and Sierra Leone, involving a mix of Liberian, Guinean and
Sierra Leonean fighters (ICG 2005, UCDP 2009). An indigenous group claimed
responsibility for the attacks, the Rassemblement des Forces Democratique de Guinée

- 31 -
(RFDG)25. Conté’s government labelled this group as part of Charles Taylor forces, but
it seems reasonable to assume that the core commanders of this group are Guinean
officials (UCDP 2009). The fight between the Government of Guinea and the RFDG
intermingled with the extension of the Sierra Leonean and Liberian wars, blurring the
distinction between the three battles (ICG 2005). Nevertheless, the Peace Agreements in
Liberia and Sierra Leone during the early 2000s decreased the violence in the border
regions to insignificant levels. In December 2008, hours after the death of President
Conté, a military coup seized power in Guinea. The effects of this coup, at the time of
writing, are still to be seen (ICG 2009).

Insurgency
The Guinean Government has been deeply involved in the failure of Liberia,
collaborating with troops and giving resources and diplomatic support to different
groups – from Doe’s government to the LURD – but especially the latter (HRW 2005,
ICG 2005). During the Liberian conflict the Government of Conté supported firstly the
armed groups fighting against Charles Taylor. After the leader of the NPFL was elected,
Guinea gave its support – in the form of arms, sanctuaries and logistical support – to the
LURD (ICG 2005, HRW 2005). This anti-Taylor support created enmity against
Conté’s regime in Liberia, especially among Taylor’s allies. The peak of this
confrontation was reached with the attacks on Guinea in 2000 and 2001. The groups
carrying out these attacks were supported by Charles Taylor and also by the RUF,
which was closely connected to – some may argued even controlled by – Charles Taylor
(ICG 2004, UCDP 2009). Although the RFDG claimed responsibility for the attacks,
the presence of Liberian fighters and regional warriors trained and armed by Liberia
shed light upon the link between the enmity between Taylor and Conté and the attack
(HRW 2005, ICG 2005). The attacks, however, did not escalate further, and can easily
be labelled as part of the Liberian war.

The clearest link between Guinea and Liberia is shown through the Guinean
support of the LURD (HRW 2002). The LURD’s bases were established in Guinea,
although the government denied it. The support became more apparent after the attacks
in 2000 and 2001 (ICG 2005). After the Comprehensive Peace Agreement in Liberia,

25
RFDG stands for ‘Rally of Democratic Forces of Guinea’ in English.

- 32 -
Guinea faced several problems with its support to military factions, endangering its own
security and stability (ICG 2005). Several groups remained active in Liberia hoping to
overthrow Conté, as well as recruiting Conté supporters. Several reports (HRW 2005,
ICG 2005) argued that different groups were recruiting former regional soldiers in
Liberia in order to set up an armed group able to fight against the Conté regime. No
attacks have taken place so far, and the new political situation in Guinea creates a new
unknown scenario for this group. Moreover, ICG (2005) points out the risk of ex-LURD
soldiers, both Liberian and Guinean, returning to Guinea. These soldiers, after the end
of the conflict in Liberia, returned to the Forest Region in Guinea – bordering Liberia –
and remained inactive and, most importantly, they have not been demobilized26.

Despite what is stated above, Guineans did not engage in informal relations with
the Liberian failure. Most of the links between Guinea and Liberia that concerned any
form of insurgency followed formal channels, mostly through the Guinean government.
The two groups sponsored by the government of Guinea –ULIMO and LURD – did not
follow informal paths for their creation, their main asset being official support.

Identity
The ethnic character of the first stages of the Liberian conflict soon crossed the border
towards Guinea. The hardest attacks were directed towards the Krahn and Mandingo
people, thus, those groups represented the bulk of the refugee community settling in
Guinea (ICG 2005, Reno 1998). This especially holds true in the case of the Mandingo
people. The Mandingo, spread all over West Africa, have traditionally not been
welcomed in Liberia, since they are considered a foreign group despite having a
presence in the country for centuries (Reno 1998). The attacks on the Mandingo, and the
links this group had with Guinea, led to the creation of the first group that challenged
Taylor’s control over Liberia. The ULIMO, which was mostly comprised of Mandingo,
was able to expel the NPFL from the northwest region, bordering Guinea (Reno 1998).
This group was established in the Guinean Forest Region but lacked a substantial social
base in Liberia (Reno 1998). The support of the Guinean government allowed ULIMO
members to freely act within Guinean borders, but also attracted the support of some

26
The Forest Region probably contains several armed groups that have sought a sanctuary far from the
reach of any other army or government. These groups are composed mostly of Liberia-war fighters in
retreat. These groups have constantly joined different armed groups or acted independently. Nevertheless,
it is extremely difficult to assess their importance, their numbers, or their strength.

- 33 -
anti-Conté groups. It is said that the 1996 coup against Conté would not have been
possible without the rebel officials getting help from ULIMO (Reno 1998). After
Taylor’s victory in the elections, ULIMO transformed into the LURD, maintaining the
links to and support of the Guinean government (ICG 2003).

After the end of the war, most of the refugees remaining in Guinea were Mandingo
(ICG 2005). These refugees were not welcomed in Liberia anymore, and they did not
consider any attempt to re-settle in Liberia safe. In addition, their presence in the Forest
Region of Guinea was seen as a threat by the Foresters, who claim to be the original
inhabitants of the region (ICG 2005, IRIN 2005). The intentions and symbolic
representation of some LURD’s ranks aggravates the tensions between the Mandingo
and other ethnic groups in the border regions between Liberia and Guinea (ICG 2005).
The presence of Mandingo people has certainly shifted the ethnic balance in the region
and some Mandingoes, ex-combatants of the LURD, maintain the idea of the restoration
of the old Mande Empire (ICG 2005). These groups are uncontrolled and probably
small and undisciplined, but they have been able to create ethnic tension in the Forest
Region in Guinea (ICG 2005, IRIN 2005). The ethnic tension has also been felt in
Liberia, where the Mandingo are encountering problems in re-settling in the Nimba
Region. This fuels ethnic tensions on both sides of the border (IRIN 2007).

Illicit Trade
Informal networks in Guinea have been involved in cross-border activities over the
border of the Forest Region (IRIN 2005, 2007). The trade between the Forest Region in
Guinea and the Nimba Region in Liberia has been always active, but trade has been
based on the peasants’ trade through the porous border. Guinea has been used
extensively as an export point for RUF diamonds (Gberie 2005). Although the bulk of
the diamonds business was channelled through Liberia, Guinea was used as an exit
point for diamonds as well. ULIMO, the armed group supported by Guinea, got
involved in the mining and trafficking of diamonds in Guinea in order to fund their
fighting (Reno 1998). The revenues Guinea got from the diamond business dropped 40
percent in 1994 (Reno 1998). The networks created by ULIMO to fund their quarrel
against the NPFL were also used by other Guineans, increasing the illegal activity in the
country during those years (Reno 1998).

- 34 -
 Nevertheless, the main illicit trade between Liberia and Guinea has been based
on weapons traffic in the Forest Region (ICG 2003, 2005). The enclave, bordering
Sierra Leone, Liberia and Cote d’Ivore has suffered a constant flow of weapons, from
and to Guinea. Guinean officials and informal networks – probably with the
acknowledgment of the government – have been deeply involved in illegal arms trade
with Liberians, providing fake documents, brand-new weapons and combat-related
material to different armed groups, mostly ULIMO and LURD. The heavy traffic of
weapons, the settlement of weapons-networks in the region and the presence of soldiers,
former soldiers and soldiers-to be have led to the region being flooded with all kinds of
weapons. The Guinean government has tried to curtail the flow of weapons by
tightening their control over the border; nevertheless, these measures have only reduced
the population’s access to goods (IRIN 2005, 2007). The presence of weapons
combined with the ethnic tensions generated by population movements surely
represents a threat to Guinea’s stability, but mostly to the inhabitants of the Forest
Region.

4.4 Within-country Analysis: The Implications of Liberia’s Failure on its Neighbours

Why did conflict erupt in Sierra Leone, but not break out in Guinea after
Liberia’s failure? In theory, both countries had authoritarian regimes and the failure of
Liberia created great opportunities for the full-fledged outbreak of conflict, but that was
not the outcome. The reason lies in the difference in the informal character of relations
between both countries.

Insurgency in Sierra Leone was informally linked to the vacuum of power in

Liberia. The creation of RUF was possible since no authority was able to enforce law in
the country. The security situation in Liberia, combined with the informal networks
created between the Sierra Leonean insurgency and the Liberian insurgency, led to the
birth of the RUF. The ranks of the RUF were Sierra Leonean immigrants in Liberia and
the lumpen youth population of Sierra Leone, which mirrored the anti-government
movements of the 1970s and 1980s. In the case of Guinea, the armed groups involved
with the failure in Liberia were mostly controlled by the government. The ULIMO and
the LURD were allegedly formed and funded by the Guinean government as an attempt
to get a stake in the Liberian conflict. These groups followed formal channels and were

- 35 -
therefore easily controlled by the Guinean government. The fact that both ULIMO and
LURD were controlled by the Guinean government averted any possible extension of
these groups’ struggle inside Guinea. Once the Guinean government lost interest in
Liberia, the lack of informal networks and the drying up of resources coming from the
Guinean government condemned the LURD to extinction. The RUF on the other hand
was able to maintain its struggle due to its informal connections with Liberia. The
difference between the informal relations of RUF with Liberia and the formal relations
of the ULIMA and LURD with Liberia are also observable in the processes of them
acquiring arms. The RUF enjoyed a synergic informal relation with Charles Taylor,
which allowed a constant flow of weapons in both directions, but mostly from Taylor
towards the RUF. Moreover, RUF’s forces were able to establish strong informal links
with Liberia and use the havoc in the country to train, hide and regroup troops.

Identity did not play an important role in the building up conflict-enhancing

mechanisms. In both cases, Mandingos were prosecuted and targeted. More concretely,
Mandingo people were being targeted during the first phases of war in Liberia and
Sierra Leone. Nevertheless, the Mandingo did not create networks of defence and
mostly sought refuge in Guinean’s Forest Region. With the end of the Liberian failure, a
considerable percentage of the Mandingo people in Guinea did not return to Liberia due
to the fear of being prosecuted. Nowadays, tensions are arising between the ‘original’
inhabitants of the Forest Region and the Mandingo refugees. Notwithstanding this, the
identity-component did not play a significant role in any of the cases.

Diamonds, the most profitable resource of West Africa, was to a great extent
present in the cases, being crucial to the development of events in Liberia, Sierra Leone
and, to a lesser extent, in Guinea. Sierra Leone and Liberia created an intricate and
strong informal network of diamond trafficking. The RUF funded its struggle with
diamonds, and the conflict in Sierra Leone broke out several times thanks to the fresh
money coming from the diamond business. However, the original outbreak of conflict
in Sierra Leone was, apparently, not related to diamonds. The diamonds were, according
to several authors, the driving force in the Sierra Leonean conflict, but it is not clear if it
was the original reason behind the outbreak of conflict (see UN 2000, Smillie et al
2000, Wannenburg 2006). However, it seems clear that all the comebacks and outbreaks
of the Sierra Leonean conflict were related to the informal business of diamonds.

- 36 -
Without these informal networks, the RUF would probably not have been able to make
all the comebacks and sustain its struggle. In the case of Guinea, diamonds were also
present, and informal networks also flourished around the cross-border activity. ULIMO
was using diamonds to fund their fight and using informal networks in Guinean where
its base was established. Nevertheless, the tight grip that the Guinean government
enjoyed over ULIMO reduced the capability of this group to act freely in the diamonds
business.

Apart from diamonds, weapons were also moved around the three countries
thanks to the state of failure in Liberia. Networks were created between Liberia and
Sierra Leone and between Liberia and Guinea. The RUF used Liberia as the main entry
point for its weaponry. The networks created for the diamonds business were at the
same time used for weapons-related traffic, and even diamonds were use as currency to
close the deals. In the case of Guinea, a similar situation developed during the failure of
Liberia. Several informal networks have been established in the region in order to
manage the illicit trade of weapons. Moreover, the involvement of the Guinean
government in the ULIMO and the LURD make essential the presence of government-
related businessmen in the informal networks. The illicit trade did not however lead to
an outbreak of conflict in Guinea, and was apparently not the main reason behind the
first outbreak of conflict in Sierra Leone; although it was the driving force during the
conflict and the reason behind the successive outbreaks of conflict in Sierra Leone. Still,
the illicit trade of diamonds and weapons are an important and sometimes crucial factor
in the outbreak of wars, and can be perfectly used as an early warning measure.

- 37 -
 5. Afghanistan’s Failure and its Influence in Pakistan and Tajikistan

5.1 Afghanistan’s Failure: The three I’s origin

Land-locked Afghanistan gained independence in 1919. In 1973, after a coup,

Afghanistan became a republic and the influence of the Soviets became even bigger.
Five years later, a communist government took control of Afghanistan, facing fierce
Islamic resistance since almost the beginning. Soviet troops were sent to Afghanistan to
help the communist regime in Kabul against the so-called mujahedeen; Islamic fighters
waging a jihad against the communist regime27. In 1989 the Soviet troops left the
country, although maintaining the support to the government. The withdrawal of the
soviet troops finally collapsed the state (Rais 2008). Since then until today, Afghanistan
has been suffering constant fighting between different groups, and the state has been
absent, being considered a clear case of state failure.

From 1989 until 1996 the different mujahedeen groups fought each other
without any clear winner. In 1994, a new group appeared to again shift the balance of
power in Afghanistan. The security situation, particularly in the Pashtun areas, the
strategic image portrayed by the group, and the vacuum of power in Kabul allowed this
new group to rapidly expand over two-thirds of the Afghan territory and stand as a
serious candidate to control Kabul in the forthcoming years (Rais 2008). This group,
named the students or Taliban, was able to gather the sympathy of a big sector of the
Pashtun population28. The origin of this group is partly found the in the Afghan diaspora
in Pakistan and the network of madrasas established in the tribal areas of Pakistan
(Hussain 2005, Rais 2008)29. Although only three countries recognized their
government, the Taliban were able to enter Kabul in 1996, establishing an extremely
weak government that was not able to solve the problem of the Afghan failure (Murshed
2006, Rais 2008). In 2001, after the Taliban-protected Al Qaeda launched an attack in
the United States, a coalition of armies, guided by the US and UK, entered Afghanistan

27
The term mujahedeen comes from the Arabic word mujahid, and can be translated as the one fighting
for a revolution, struggle or jihad. The term jihad refers to the struggle in the way of Allah in the Muslim
religion. The scope of the ‘struggle’ is not clear, and it can be understood as a personal struggle or, as
usually portrayed by Western media, a holy war.
28
The word Taliban is the Pashtun plural for the Arabic word Talib, which means ‘student’.
29
The term madrasa comes from Arabic and literally means ‘school’, nevertheless, a misused of the term
in popular science has transformed its meaning into ‘Islamic school’

- 38 -
and supported one of the anti-Taliban groups30. The coalition and its supporters soon
reached Kabul and ousted the Taliban government. Nowadays, Afghanistan remains in a
state of failure, with a constant fight between all the different actors going on in the
country.

Since 1989, Afghanistan has not been able to provide any of the basic political
goods to its citizens. With the withdrawal of the Soviet forces, no entity has been able to
provide any sort of security for the whole country, rendering the security a private good
in the hands of warlords. The fights in the different parts of Afghanistan have left the
whole country in the hands of warlords, even during the Taliban years (Rais 2008). In
addition to the lack of security, the different ethnic affiliation of the armed groups
present in Afghanistan created a situation of insecurity and ethnic tension between the
ethnic groups of Afghanistan. The most relevant one was between the Tajik groups in
the north and the Pashtun groups in the rest of the country; however, this tension was
only one among many other ethnic problems. Obviously, the economy of a ravaged
country like Afghanistan has been unable to perform in a normative way. More than 30
years of civil war in Afghanistan has left the economy basically reduced to informal
networks of commerce, in many cases based on the illegal trade of poppy,
archaeological relics, and weapons.

5.2 Afghanistan – Pakistan

The relationship between Afghanistan and Pakistan has always been turbulent.
Since Pakistan gained independence in 1947, Afghanistan has been the strategic depth
that Pakistan lacks (Hussain 2005)31. During the Soviet invasion of Afghanistan,
Pakistan funded, armed, and helped the insurgency movements, or mujahedeen, to fight

30
Al Qaeda, or The Base, is an armed group made up of Muslim fighters that supported the mujahedeen
during the Soviet invasion of Afghanistan. Al Qaeda is a loosely linked network of groups, commanded
by the Saudi national Osama Bin Laden, who, according to Kepel (2002), was able to create a database
with the information of these fighters. The scope of their action is worldwide, and they claim
responsibility for terrorist attacks in cities such as New York, London, Casablanca, Madrid, and on Bali.
For more information see Kepel (2002).
31
Pakistani officials assessed it to be dangerous not to have a friendly Afghanistan. The geostrategic
reason was the lack of ‘strategic depth’ or retreat capacity, in case of an Indian invasion in the East. For
more information about the Pakistani motivation and interest during the Soviet invasion, see Hussain
(2005).

- 39 -
against the Soviet-like regime in Kabul. Thousand of fighters, both Afghans and
Pakistani, were trained in tribal areas in the north-west of Pakistan (Hussain 2005).
After the withdrawal of the Soviet Army and the total collapse of Afghanistan, Pakistan
maintained its interest in the country, supporting the Hezb-e Islami group against the
other militias fighting for power. The Pakistani support shifted towards another group
which originated within Pakistani territory, the Taliban. The Taliban entered Kabul in
1996 with the support of Pakistan and with many Pakistani nationals within its ranks
(Hussain 2005). With the American- and UK-led invasion of Afghanistan, Pakistan
shifted its support again, this time towards the American and British army. However,
the nexus between the Taliban and its supporters in the tribal areas of Pakistan did not
follow the new policy of Islamabad (Hussain 2005). The Taliban used the tribal areas to
retreat, challenging the authority of the Pakistani government.

The Taliban in Pakistan have transformed into a powerful force since then, both
in terms of military and support, gaining control of several regions in the tribal areas.
The government, unable to tackle the insurgency and the increasing power of the
Taliban, has entered into several negotiations with them while at the same time
launching a fierce military operation against them. As the time of writing the conflict
remains unsolved. The Taliban now control several regions, especially in Waziristan,
and have been able to impose Sharia law in a considerable number of regions.

Insurgency
The relationship between armed groups and insurgencies in Afghanistan and Pakistan
during the last 40 years is deeply rooted in networks of collaboration and common
ideologies (Hussain 2005). Since the failure of Afghanistan in 1989, cross-border
activities have taken place almost on a daily basis between Afghanistan and Pakistan.
Nevertheless, armed groups already engaged in cross-border activities active during the
Soviet invasion of Afghanistan (Hussain 2005). Geostrategic and religious
considerations by the Pakistani leadership resulted in a close collaboration between the
anti-soviet groups and the Pakistani state (Hussain 2005). The origins of the Afghan
‘jihad’ between 1979 and 1992 are found in Pakistan, as described by the soviet

- 40 -
ambassador in Afghanistan at the time (Hussain 2005)32. These groups were being
protected and sometimes assisted by the Pakistani army in order to avoid the communist
takeover of Afghanistan, among other reasons (Hussain 2005). Peshawar, the capital of
Pakistan’s North West Frontier Province (NWFP), became the operational centre of the
Afghan Islamic insurgency (Hussain 2005).

The withdrawal of the Soviet Union and the collapse of Afghanistan did not
lesser the involvement of Pakistan in the Afghan milieu. During the fierce power
struggle, Pakistan maintained a clear support to the Hezb-e Islami fight against Ahmad
Shah Masoud’s militia, the United Islamic Front for the Salvation of Afghanistan
(UIFSA) (Hussain 2005). A relative loss of influence over the events in Afghanistan
forced Pakistan to shift their support to an emerging Pashtun group spawned in the
madrasas of the Federally Administrated Tribal Areas (FATA), the Taliban (Hussain
2005)33. This group seized power amid the collapsed of Afghanistan in 1996, but was
overthrown by the UIFSA at the end of 2001; nevertheless, the relationship between the
Taliban and Pakistan was not over.

The Taliban enjoyed the international support of Pakistan, but the state of
collapse in Afghanistan fostered an increase in the informal relations between the
Taliban and their counterparts, the Pashtuns in Pakistan. They especially found support
among the FATA, without the knowledge of Pakistani authorities (ICG 2006). The
UIFSA threw the Taliban from power with the support of an American- and UK-led
international coalition. This enhanced the state of collapse in Afghanistan and pushed
some of the Taliban towards their origins, the FATA and the NWFP in Pakistan
(Hussain 2005). Moreover, ousting of the Taliban had another crucial effect on
Pakistan. The Pakistani government sided with the anti-Taliban forces in order to avoid
the military fury of a wounded US government and engaged, at times just verbally, in
the anti-Taliban fight (Sinno 2008)34.

32
Alexandre Puzanov, Soviet ambassador to Afghanistan in 1978 reported ‘propaganda, dropping in of
weapons and diversionary groups which are being prepared in Pakistan to topple the PDPA regime’
(quoted in Hussain 2005:102)
33
The FATA and the NWFP are the tribal areas in Pakistan bordering Afghanistan. These areas enjoy a
special status and do not follow provincial or federal legislation (Gunaratna and Nielsen 2008). In
addition, Baluchistan also borders Afghanistan in the south.
34
The US, after the attacks of September 2001, launched a ‘war on terror’; their first target being the
Afghan government and the terrorist network Al-Qaeda. Pakistan, in a drastic turn in its policies and in
fear of the US reaction, supported the ‘war on terror’ and became one of its major defendants.

- 41 -
 It is crucial to understand the aim of the Taliban and the groups related to them.
Their ultra-orthodox religious ideas go beyond established borders, and the whole
Muslim world is their possible stage of action (Kepel 2002, Rashid 2002). Their
intentions, according to their statements, reach beyond power, and are found on a more
spiritual level (Kepel 2002). The main benchmarks of their ideology are the imposition
of sharia law, and creating a normative Islamic society that live according to the rules
of the Koran. Nevertheless, as will be pointed out later (see identity section below), the
Taliban, being mostly a Pashtun group, also created regional ethnic tensions, (Hussain
2005). Most likely, they aim at controlling all the Pashtun regions and want to impose
their radical religious view there (Hussain 2005, ICG 2006, Kepel 2002). Gearing up
with Pakistani groups and creating a somehow homogeneous actor points to this
objective (ICG 2009b).

The Taliban fighters entering Pakistan in 2001 and 2002 were mostly Pashtun,
although Arabic and Central Asian fighters were also among those crossing the porous
border (Gunaratna and Nielsen 2008). The involvement of Pakistani nationals, and more
importantly, the increasing involvement of Pakistani indigenous groups have shifted the
outlook of the Taliban and their objectives substantially. Several Pakistani groups,
religiously related to the ultra-orthodox Deobandi Islam preached by the Taliban, have
taken advantage of the surge of militants in the FATA and NWFP to increase the
strength of their struggle (ICG 2009b). Nowadays, all these groups have merged and
represent a loosely homogeneous actor with stakes on both sides of the border. In March
2009, Taliban groups in Pakistan were acting on both sides of the border and entailed
several different nationalities, although they were predominantly Pashtun (ICG 2009)35

The ouster of the Taliban has been felt extensively in Pakistan, leading to the
FATA and the NWFP gaining more strength (ICG 2006, 2009b). Overthrowing the
Taliban pushed several members of the Taliban and Al-Qaeda towards Pakistan, in
search of a safe haven (ICG 2006)36. The Taliban merged with the local population of

35
It is possible to find Uzbeks, Arabs, Muslim Europeans and East Asians among the foreign fighters
(ICG 2006, 2009b)
36
The treatment and categorization of the Taliban as a terrorist group makes it difficult to distinguish the
line that separates Al-Qaeda and the Taliban, if there is any; therefore, this paper will treat both groups as
a unique unit of analysis. The close links created between both groups before and after the ouster of the
Taliban allows for the homogenization of their activities. ICG (2009b) assesses that a group called

- 42 -
these tribal areas, receiving support and shelter from different local militias and tribal
leaders (ICG 2006). Taliban leaders used their connections and networks in the FATA
and NWFP to gain power in the provinces. The network of madrasas and kindred
militant groups has been utilized to recruit disfranchised young people to fight the
Soviet after their invasion of Afghanistan (Hussain 2005). The Taliban used these
networks to impose their power over some provinces in the FATA and the NWFP (ICG
2009b). The compliance of the local tribe leaders, or maliks, is difficult to assess37.
Evidence points at a close collaboration between maliks and the Taliban (ICG 2006).
Nevertheless, these local tribal leaders have received large amounts of money for their
services (ICG 2006). In addition, Taliban fighters have killed several maliks recently,
and their authority has dwindled gradually as the Taliban strength increased (ICG 2006,
2009b).

In June 2002, less than a year after the ouster of the Taliban, the Pakistani
intelligence (Inter-Services Intelligence, or ISI) had already gathered information about
the activities of Taliban fighters on Pakistani soil, moving troops to the FATA in order
to curtail the activities of these groups (ICG 2006). The Taliban in Pakistan have been
active since then, expanding the war in Afghanistan inside Pakistani borders and
engaging the Pakistan Armed Forces into the struggle. The Taliban’s intentions and
range of action have been decisive in their relationship with the Pakistani government.
As part of the ‘war on terror’, the Pakistani army moved the FATA and NWFP in order
to hunt down the leaders of the Taliban and Al-Qaeda in their territory (ICG 2006).
Nevertheless, time demonstrated that these leaders and fighters were not only a
temporary presence in Pakistan.

The de facto authority of Taliban groups in the FATA and the NWFP, now
under the loosely unified command of Baitullah Mehsud and the Tehrik-i-Taliban in
Pakistan, has been challenged by the Pakistani army, creating a long conflict throughout
the region38. The reason behind the Pakistani army’s involvement in the fight against

‘Tehrik-i-Taliban Pakistan’ has been loosely coordinating all the activities of all the Taliban-oriented
groups in the FATA and NWFP since 2007.
37
The maliks system was inherited from British colonial rule. The British created a special position in the
tribal areas in order to forge friendly elites. The maliks receive money from the state and they maintain a
powerful status in the daily decision-making process of the communities (ICG 2006)
38
The authority of Mehsud is questionable since several confrontations between different groups under
his auspice have occurred since 2007(ICG 2009b)

- 43 -
the Taliban is twofold. On one hand, Islamabad saw that it was losing its weak control
over the FATA and NWFP, and these uncontrolled areas were also being used by other
groups, especially in Punjabi areas and in Khasmir (ICG 2009b). On the other hand,
increased US pressure on the Pakistani government accelerated the military
intervention, leaving no space for early political solutions (ICG 2009b). The war
between the Pakistani army and the Taliban, sometimes supported by NATO air strikes,
has seen full-fledged conflict during the last years, and remains unsolved (ICG 2009,
UCDP 2009c).

It is worth noting that the Taliban in Afghanistan and the Taliban in Pakistan are,
in theory, two separate organizations with two separate leaders (Acharya et al 2009).
Both movements consider themselves pieces of a bigger mechanism, the Taliban, and
collaborate in several situations. Their objectives are the same, and they fight together
in Afghanistan. Nevertheless, the Pakistan Taliban also runs their own agenda, targeting
objectives outside Afghanistan and the FATA and NWFP region (Acharya et al 2009).
It is impossible to draw a line between the groups, mostly due to their shared Pashtun
identity and to the centuries of cross-border activities between Afghanistan and
Pakistan. At best, it is possible to distinguish which operations are part of the resistance
against the ISAF-NATO forces, and which operations target Pakistani-only objectives.

Identity
Although it has a global vision, the Taliban is mostly a Pashtun group, and the group is
mostly influential in Pashtun areas (Gunaratna and Nielsen 2008). The Taliban argues
that Islam, rather than the Pashtun identity, is what unites and glue together all of its
members, nevertheless, its scope of action has been felt mostly in Pashtun areas –
FATA, NWFP and Afghanistan. The strong Pashtun identity can be found at both sides
of the border and it has had dire consequences, especially for Pakistan. After the
American-led invasion in 2001, the Taliban sough refuge with their ethnic counterparts
in the FATA and NWFP (ICG 2006). At the same time, the invasion and ouster from
power created grievances that were also felt in the FATA and NWFP (Gunaratna and
Nielsen 2008). The close alliance of the Pakistani government with the US-led invasion
of Afghanistan meant an extra grievance for the government in Islamabad. The Taliban-
like groups in Pakistan sided with the global struggle of the Afghan Taliban, pointing at
the Pakistani government as the closest enemy of the fight. The discourse of the

- 44 -
Pakistan Taliban and the Afghan Taliban, and their objectives, shares a multitude of
similar traces (ICG 2008). The conjunction of discourses is a clear sign of the ethnic ties
and the shared grievances on both sides of the border. The Pakistan Taliban has adopted
the behaviour, grievances, ideology and methods of their ethnic counterparts in
Afghanistan (ICG 2006, 2009b)

Finding a safe haven in Pakistan’s FATA and NWFP was possible because of
the ethnic ties uniting both the Taliban and the population in these two regions. The
Afghan armed groups moved to an already-known scenario where they maintained
informal connections with the leaders in 2001. The Taliban have been able to merge and
hide within the local population perfectly due to their predominantly Pashtun identity,
making it seemingly impossible to differentiate an Afghan Pashtun from a Pakistani
Pashtun (ICG 2006, 2009b).

Illicit trade
The failure in Afghanistan has created business opportunities on both sides of the
border, mostly related to the illicit trade of heroin, weapons and archaeological relics.
The illegal trade of these goods has been used as one of the main funding opportunities
for the Taliban in Afghanistan and Pakistan, but the illegal activities of the Taliban in
Pakistan are not only limited to smuggling heroin and archaeological relics (Acharya et
al 2009). It is important to note the longstanding informal networks present in the tribal
areas of Pakistan and Afghanistan. Their nexus has always existed, but the failure of
Afghanistan after the Soviet withdrawal increased the salience of the link. The two main
informal economic activities used by the Taliban in Pakistan are related to the failure in
Afghanistan: heroin and weapons traffic.

The CIA encouraged poppy cultivation in Afghanistan during the Soviet

invasion, as a funding opportunity for the mujahideen (Halali 2002). Nowadays, poppy
cultivation represents one of the main funding opportunities for Afghan and Pakistani
armed groups, which are in charge of the security and control the traffic lines. One of
the main three routes of the heroin transportation crosses the FATA and NWFP towards
the Karachi port (Priego 2008). From there, refined and unrefined heroin is transported
to the Arabic Peninsula, where it continues its journey towards Turkey and Europe
(Priego 2008). As noted before, Taliban control over several regions in both

- 45 -
Afghanistan and Pakistan allow for a smooth flow of heroin through the FATA and
NWFP. The Taliban enjoy tight control over the transportation of all goods across the
Afghan-Pakistani border (Acharya et al 2009). A relatively complex taxation system
provides revenue and helps to identify any goods, including heroin, crossing into
Pakistani tribal areas (Acharya et al 2009). Although there is an obvious lack of
information about the drug trafficking, the Taliban in some regions of Pakistan and
Afghanistan also monitor – and profit from – the drug traffic (Acharya et al 2009, Shaw
2006). Taliban rulers in the FATA and NWFP are probably deeply involved in heroin
smuggling from Afghanistan, providing protection and taxing the activity. In addition to
merely participating in transportation, the FATA and NWFP have a network of
refineries where the raw poppy plants are transformed into consumable heroin. Again,
the dominance of the Taliban in certain regions of the FATA and NWFP indicates that
the Taliban are using the drugs to fund their war.

In addition to the drug business, the failure in Afghanistan, ironically, flooded

Pakistan with weapons, creating a ‘Kalashnikov culture’ (Hilali 2002). Pakistani
weapons ‘expertise’ is an indirect consequence of the war in Afghanistan and its
subsequent failure (Bhatia 2008). Most of the weapons currently present in Afghanistan
have at some point been in the hands of Pakistani weapons dealers, and many of them
are even returning to Pakistan (Bhatia 2008). The CIA used to ship their weapons
through the port of Karachi, thus increasing the corruption of army officials in Pakistan,
and creating a perfect opportunity for business in the FATA and NWFP (Hilali 2002).
Some villages in the FATA and NWFP have become some of the biggest unofficial
arms markets in the world (Hilali 2002). The business does not promote any particular
group, but is established as a free market where goods are sold to and bought by
anybody able and willing to pay the price. The streets of several towns, including
Peshawar, are filled with arms-related business where almost any kind of light weapons
can be purchased.

Although the main illegal activities of the Pakistan Taliban in relation with
Afghanistan are drugs and weapons, the dense traffic between the two countries through
the FATA and NWFP currently provides a clear funding opportunity for the Taliban in
Pakistan (Acharya et al 2009). The Taliban taxes all the road traffic in the FATA and
NWFP (Acharya et al 2009). These taxes include the permission to use the roads, and

- 46 -
also protection of trucks in the Taliban-controlled territoreis on both sides of the border
(Acharya et al 2009). The ISAF-NATO trucks are also an objective of the Taliban, and
the group ambushes them with relatively high frequency. The network of Taliban-
supporters in Karachi helps to identify the trucks, which are robbed when they are in the
FATA and NWFP (Acharya et al 2009).

5.3 Afghanistan-Tajikistan

Before embarking on any analysis regarding the relationship between

Afghanistan and Tajikistan, it is important to shed light upon the civil war that took
place in Tajikistan between 1992 and 1997, and the role Afghanistan played in the war.
In 1992, Afghanistan was already a failed state, and Tajikistan was a recently created
republic after the breakdown of the Soviet Union (Jonson 2006). During Gorbachov’s
government, Tajikistan acquired several rights within the frame of the Soviet Union,
and definitively created a national identity (Joson 2006). The civil war that took place
during the first years of independence was a direct consequence of the power vacuum
created by the breakdown of the Soviet Union. Nevertheless, the Afghan factor did have
an influence on the duration of the conflict, and some authors argue it also played a part
in the motives behind it, but not in the creation and escalation of it (ICG 2001, 2009c,
Jonson 2006:40-43, Rashid 2002). The reasons behind the civil war in Tajikistan,
according to ICG (2001), Jonson (2006) and Rashid (2002), were local, and
Afghanistan’s failure did not influence the outbreak of the conflict. The conflict in
Tajikistan, using counterfactual logic, would have broken out even without the failure
of Afghanistan a few years earlier. Therefore, in order to avoid misunderstanding and
biased analysis of a country already plunged into a conflict, the analysis of Tajikistan
will be done starting at 1997, the last year of the civil war. Moreover, the final objective
of this paper is to shed light on the effects of state failure on the outbreak of civil war in
neighbouring countries; therefore, the study of the Tajik civil war remains outside the
scope of this research.

Since the end of the civil war, Tajikistan has plunged into an unstable situation.
The political system, commanded by president Emomali Rakhmon, has progressively
dwindled and become more and more corrupt, raising the already-high level of poverty
in the country (ICG 2009c). Ethnic tensions have arisen throughout the country but no

- 47 -
policies have been enforced in order to tackle them (ICG 2009c). The violent civil war
and the mismanagement of Rakhmon’s government have destroyed what little
infrastructures were left behind by the Soviets (ICG 2009c). Nowadays, most of the
country faces energy blackouts during most of the winter, and the mountainous
geography of Tajikistan increases the isolation of several regions (ICG 2009c). As a
consequence of the unstable situation in Tajikistan, thousands of young Tajiks have
emigrated as seasonal workers to its wealthy neighbours. Nevertheless, the current
economic crisis is reducing the amount of available employment abroad, thus also
reducing the remittances that were sustaining a considerable percentage of the Tajik
population (ICG 2009c)

Insurgency
During the years of ethnic-Tajik control of the government in Kabul, Tajikistan’s anti-
government opposition sought refuge in Afghanistan (Jonson 2006). The Taliban’s
ousting of the ethnic Tajik government in 1996 ironically meant a bigger effort was
needed by the international community to put an end to Tajikistan’s civil war (Jonson
2006, Rashid 2002). The threat of the Taliban increased soon after the General Peace
Agreement was signed in June 1997 (Rashid 2002). Masood’s UIFSA received
increased support from Tajikistan and other countries after the Taliban advanced into
Tajik areas in Afghanistan. At some point during 2000, Taliban forces were eyeball to
eyeball with the Russian troops protecting the Afghan-Tajik border (Rashid 2002)39.
Tajikistan was then used as the main entry point for supplies directed at Masood’s
UIFSA, and the Tajik government offered all its support to the group (Jonson 2006).
With the intervention of the US and UK-led troops in Afghanistan after the attacks on
New York and Washington in September 2001, the prospects of the UIFSA – renamed
Northern Alliance – drastically changed, and so did the perception of the Taliban threat
in the eyes of the Tajik government.

The fast victory of the Northern Alliance, with the decisive support of the US
and UK-led international force, meant a relief for the Tajik government. The Taliban
left Kabul and withdrew to the east and south of the country, far from Tajikistan. With

39
Russia deployed troops on the Afghan-Tajik border in 1993 as part of a bilateral agreement. However,
the deployment has been highly polemical on both sides of the border and Russia’s intentions are still
unclear. For more information see Jonson (2006) and Rashid (2002).

- 48 -
the set-up of a new government with Tajik representation, the situation partly improved
and the Taliban were no longer a security threat to Tajikistan (Jonson 2006). The
situation has worsened over the years, and Tajikistan has not been able to play the role
of ‘middleman’ that Russia and the US intended (Jonson 2006). The Afghan-Tajik
border remains porous and unpatrolled, and Afghanistan remains unstable.
Nevertheless, the Tajik national identity still remains stronger than any possible support
for the Taliban orthodox Islam. As a consequence, no anti-government groups have
created informal networks. Simply speaking, Tajiks won the Afghan war against the
Taliban, and no group has, so far, sought the support of the Taliban and taken advantage
of the security situation in Afghanistan to initiate a conflict in Tajikistan. Moreover, the
northern part of Afghanistan remains slightly more stable than the rest of the country,
rendering it more difficult for any armed group to seek refuge and to establish informal
networks in the region. As an example, the most Islamist militant group during the civil
war in Tajikistan, the Islamic Renaissence Party (IRP), which shared the same orthodox
interpretation of Islam with the Taliban, did support the UIFSA during the Afghan
conflict (Rashid 2002). Therefore, it is so far unimaginable to see an Islamic group
seeking the support of the Taliban to fight the Tajikistan government.

Nevertheless, Tajikistan still faces another threat with a direct link to

Afghanistan, namely the Islamic Movement of Uzbekistan (IMU) and the Hizb ut-
Tahrir al-Islami (Party of Islamic Liberation, HT). These two groups, listed as terrorist
groups by several organizations, maintain a direct link with the Taliban and have a Pan-
Central Asian scope of action, which obviously includes Tajikistan (Rashid 2002). The
threat of the IMU, the most active group in the region, is exacerbated by the tensions of
Tajikistan with Uzbekistan and Kyrgyzstan. However, IMU is said to have its
operational base on Afghan soil, and enjoys the support and help of the Taliban (Cornell
2005). Indeed, the IMU was formally founded in Afghanistan (Cornell 2005). In
addition to their use of Afghan soil, the IMU is believed to have bases in the Tajik
section of the Ferghana Valley (ICG 2002). The position of the IMU in Afghanistan was
initially weakened during the first years after the ouster of the Taliban; nevertheless,
with the resurgence of the Taliban, the IMU has again gained a sanctuary for its central
command. The support of the IMO in Tajikistan remains low, according to ICG (2002),
but its Pan-Central Asian agenda and the increasing strength of the Taliban in
Afghanistan and in Pakistan threaten the stability of Tajikistan. Moreover, the possible

- 49 -
future presence of American bases in Tajikistan could further endanger the weak
stability of the country (ICG 2009c). The presence and interest of the IMU in Tajikistan
is unclear, and the lack of control of the Tajik government of the Ferghana Valley
increases the lack of information regarding IMU’s presence in the country.

Identity
The Tajik identity is a key piece of the puzzle that is Afghanistan, and probably stands
as the reason behind the lack of outbreak of conflict related to state-failure- in the Tajik
case. There are more ethnic Tajiks in Afghanistan than in Tajikistan (Jonson 2006) and
it would be logical to assume than informal networks of ethnic Tajiks on both sides of
the border enhance the risk of conflict in Tajikistan. Nevertheless, as exposed above, the
Tajik ‘won’ the war in Afghanistan, therefore is in their own interest to maintain the
formal character of the ethnic relations on both sides of the border. These are two of the
scenarios in which ethnic ties might prompt conflict in a neighbour of a failed state. On
one side, armed groups could find safe haven in a neighbour where their ethnic group is
also present. This situation can mean a challenge for the government insofar it does not
accept the presence of this group within its borders. In the Tajik case, the presence of
the mostly-Tajik Northern Alliance is viewed favourably by the government (Jonson
2006, Rashid 2002). The interests of the Tajik government in Afghanistan relied upon
the situation of the Tajik minority. After the ouster of the Taliban in 2001, Tajiks have
enjoyed an important presence in the Afghan government (Jonson 2006). This has also
improved the situation of the Tajik government vis-á-vis Afghanistan. Even before the
ouster of the Taliban, the support that Tajikistan gave to the Northern Alliance – then
the UIFSA– averted any sort of conflict-enhancing influence of the failure in
Afghanistan on Tajikistan.

On the other hand, ethnic ties can affiliate the transmission of conflict-enhancing
mechanisms when big groups of people move between the failed state and a
neighbouring country. This situation took place between Tajikistan and Afghanistan
after the civil war in Tajikistan. The refugees were mostly Afghan-Tajiks crossing into
Tajikistan to seek a safe refuge (Jonson 2006, Rashid 2002). Obviously, despite all the
negative effects that refugee flows may have, any alteration of the ethnic equilibrium
did not take place in Afghanistan. The Afghans crossing into Tajikistan were ethnic

- 50 -
Tajik, which allowed easier settlement in Tajikistan albeit they did face hard conditions
in the war-ravaged country (UNHCR 2005).

Illicit Trade
The failure of Afghanistan led to a boost to poppy production and exportation to the
Middle East, Africa, Europe and Asia. Tajikistan, due to its location and mountainous
geography, is one of the main points of transport in the drug route towards Russia and
Europe (Priego 2008). Tajikistan has been labelled a ‘narco-state’ due to the heavy
reliance of its economy on the drug business (Greenfield et al. 2007). According to
Greenfield et al. (2007), the drug industry adds up to a 30% to the GDP of Tajikistan
and supports thousand of Tajiks. As pointed out by Engwall (2005), the narcotics meant
a serious threat to Tajikistan security on several levels, most prominently, related to
conflict.

When researching the influence of Afghan’s poppy production on Tajikistan’s

security, all the literature points at one particular group, the IMU (Cornell 2005,
Engwall 2005, ICG 2002). The armed group, based in Afghanistan, dominates a fairly
big share of the drug business in Central Asia. The fact that Tajikistan is the main entry
point for the Afghan drugs towards Central Asia effectively links the destinies of the
IMU and Tajikistan (Cornell 2005, Priego 2006). The IMU profits extensively from the
drug business, it being their main source of income. The line between their profitable
activity and their pursued goal has blurred with the years, and nowadays it is hard to
separate one from the other (Cornell 2005, Engwall 2005).

The IMU’s actions, objectives and methods are, according to Cornell (2005),
highly related to the drug trafficking, and sometimes drugs are even the main reason
behind them. The route between Khorog, a city near the Afghan-Tajik border and one of
the main entry points for narcotics, and Osh, the main city in south Kyrgyzstan has been
labelled as the main drug-route between Afghanistan and Central Asia, and has been
strictly monitored by different organizations (Cornell 2005). Nevertheless, the IMU has
been able to open other routes from Afghanistan. These routes lie near their main
strongholds, and give the IMU comprehensive control of the business (Cornell 2005).
Tajik warlords are also involved in the illicit business, usually supporting the IMU, or
with their ‘permission’ (Cornell 2005, Engwall 2005).

- 51 -
 The heavy presence of narcotics in Tajikistan creates a paradox for the Tajik
government. On one hand, effectively curtailing the business and putting an end to the
informal networks that manage it will affect thousand of citizens and low-paid
government workers. Corruption is widespread throughout Tajikistan’s government at
all levels (ICG 2009c), and putting an end to the drug business will decrease the income
of a considerable percentage of the population. On the other hand, the illicit networks
controlling the transit of narcotics from Afghanistan are gradually increasing their
power and capabilities, the IMU being the clearest example of this. This as pointed out
by Cornell (2005) and Engwall (2005) endangers the security of Tajikistan and might
create forces powerful enough to challenge the already weak authority of the Tajik
government as well as overcome the trauma of civil war (Cornell 2005, Jonson 2006).
Nevertheless, due to the IMU’s own interests, their involvement in the informal
networks of the drug business does not directly threaten Tajikistan (see Cornell 2005,
ICG 2002). According to different sources (Cornell 2005, ICG 2002, Rashid 2002) the
IMU aims at overthrowing the government of Uzbekistan, at least as a first step towards
the creation of a Caliphate in Central Asia. Nevertheless, the Tajikistan government,
where Islamic parties bear a great influence, is rather an example than an enemy to the
IMU.

5.4 Within-country Analysis: the Implications of Afghanistan’s Failure on its

Neighbours
Afghanistan’s failure seems to be the driving force behind the conflict in
Pakistan, and it can be argued that the conflict in Pakistan would have not broken out
without the failure in Afghanistan. Nonetheless, the failure in Afghanistan did not,
apparently and keeping in mind the particularities of the Tajik case, help to create
conflict in a highly unstable country such as Tajikistan. What are the reasons behind this
different outcome? According to the logic exposed above, Pakistan has created strong
informal networks in Afghanistan, and these networks took advantage of the lack of
security, representation and wealth in order to carry out their own fight. In the case of
Tajikistan, these informal networks have not been present, and most of the relations
between Tajiks and Afghans are done through formal networks, even during the Taliban
years.

- 52 -
 The success of insurgency in Pakistan is clearly related to the creation of
informal networks between Pakistan and Afghanistan. These networks are mostly based
on the porous border between the two countries – more specifically, along the borders
between FATA, NWFP and Afghanistan. The two directions of the insurgency have
taken place in Pakistan. Afghan groups, namely the Taliban and Al Qaeda, sought
refuge in Pakistan due to the establishment of informal networks during the war against
the Soviet presence in Afghanistan. These networks have challenged the Pakistani
authority in the FATA and NWFP, and this has prompted a fierce conflict between
Pakistan and these groups. In addition, Pakistani groups have used these informal
networks to carry out their struggle against the Pakistani government. The lack of
authority in Afghanistan allowed both groups to acquire the resources and grievances
needed for the outbreak of conflict in Pakistan. However, in the case of Tajikistan, these
networks have been channelled though the Tajik government’s support to the UIFSA-
Northern Alliance. After the failure in Afghanistan this group and the relationship it
enjoys with the Tajik government have averted any incentive to challenge its authority
within Tajikistan. This behaviour has been seen both during the Taliban years and after
the ouster of the Taliban. However, in the Pakistani case, these groups did not rely upon
the Pakistani government in such a manner that they would have been able to act
independently and thus be able to challenge the Pakistani authority when needed.

Identity also plays a fundamental role when it comes to the type of influence that
Afghanistan wields over Pakistan and Tajikistan. In Pakistan, the creation of informal
networks among the Pashtun is seen as a fundamental factor behind the outbreak of
conflict in the region. These networks transferred the grievances and actions of the
Afghani Pashtun population towards Pakistani Pashtun. Moreover, Pashtuns at both
sides of the border have collaborated closely and the informal networks that organize
such collaboration enjoy widespread power throughout the FATA and NWFP. On the
contrary, although networks were created between Afghanistan and Tajikistan, these did
not enjoy the informal character of the Afghanistan-Pakistan networks. The bulk of the
interaction between Tajiks in Tajikistan and Tajiks in Afghanistan has been dominated
by the formal influence Tajikistan and its allies wields over the different Tajik groups –
mostly over the UIFSA-Northern Alliance. In the Pakistani case, the grievance of the
Pashtun population in Afghanistan rapidly spread across the border towards Pakistan
thanks to the informal networks of collaboration created before between both Pashtun

- 53 -
groups. Nowadays, Afghani and Pakistani Pashtuns argue that they are fighting the
same war against the same enemy. Not only grievances crossed the porous border
between Afghanistan and Pakistan, ideologies and methods originally developed in
Afghanistan are now found among Pakistani Pashtuns. Tajiks, on the contrary, have not
felt the urgency to adopt the grievances of their counterparts in Afghanistan since Tajiks
in Afghanistan can be labelled as the winners, and they enjoy a powerful position within
the Afghan institutions.

Poppy cultivation has cursed Afghanistan and its neighbours, and to this day, it
remains one of the fundamental problems of the country, as well as one of the main
funding opportunities for armed groups. Tajikistan suffers the dire consequences of
poppy cultivation through the actions of IMU, which stands as one of the main security
risks for the country. Nevertheless, at the time of writing, IMU is not directly
threatening Tajikistan’s security and it remains active only in relation with heroin traffic
in the region. Pakistan insurgency seems to be widely benefiting from the heroin
business and it is being one of the financial means used by the Taliban. The close
collaboration of Afghan Taliban and Pakistani Taliban also extends to the drug
business, the FATA and NWFP being the main entrance points into Pakistan. It is
unclear what the exact relation between the Pakistani Taliban and the heroin trade is,
and how involved they are in the business. Evidence points at an indirect benefit from
heroin traffic. On one hand, Pakistani Taliban exercises a tight and complex taxation
system of the FATA and NWFP roads, which probably involves heroin taxation. The
taxation system is also found, although to a lesser degree, in cities and villages in the
region. On the other hand, the close links between Afghan and Pakistani Taliban
ensures a common use of resources, such as weapons or intelligence. A considerable
percentage of these resources are financed through the heroin business. Nevertheless,
heroin does not seem to be a root cause of the outbreak of conflict in Pakistan. The
control wielded by the Taliban encircles both illegal and legal resources, heroin being
one among several different means of funding.

- 54 -
6. Analysis: Assessing the Implications of State Failure on Neighbouring Countries

Based on the results of the case analysis and the empiric information systematically
gathered for this paper, it is possible to find some empiric evidence that would lead to a
refinement of the theoretical construct of how state failure affects neighbouring
countries. In the section 2 of the paper three independent variables were exposed,
Insurgency, Identity and Illicit trade. The theoretical strength of these three variables
and of the theory can now be assessed regarding the case analysis and the empirical
evidence. In addition, this will allow an explanation of the variation on the dependent
variables observed in the four units of analysis. The analysis will firstly summarise the
outcomes of the four cases, drawing preliminary conclusions about the strength of the
theoretical framework. Secondly, the strength of the independent variables will be
assessed within a comprehensive and detailed analysis. Lastly, the main weakness and
limitations of the theoretical framework will be exposed.

The four units of analysis observed have underlined the link between state
failure and conflict in a neighbouring country. The presence of informal networks
taking advantage of the failure of Afghanistan and Liberia partly explains the different
outcome in the 4 cases. The variable Insurgency seems to have great explanatory power.
The presence of informal networks between Liberia and Sierra Leone and between
Afghanistan and Pakistan seems to explain the outbreak of conflict in Sierra Leone and
Pakistan. These networks were able to use the lack of security in the failed state to
create conflict in the neighbouring state. At the same time, Guinea and Tajikistan did
not suffer the actions of informal networks benefiting from the lack of security in the
failed state, which seems to explain the lack of conflict outbreak in their territory. The
comprehensive picture of the mechanism seems to point out Illicit Trade as part of the
insurgency efforts, occurring after the conflict has broken out. Identity plays a strong
part in the Afghan case, but there is no empirical evidence of it being influential in the
Liberian case, rendering the interpretation of its results complex.

The conflict in Pakistan is mostly caused by the Taliban actions inside Pakistan.
Taliban groups have been able to mingle with the population and continue their struggle
due to the use of informal networks between both countries. These informal networks
are based on considerable segments of population on both sides of the border sharing

- 55 -
the same identity, ideology and religion. These similarities between inhabitants of both
sides of the border have enabled the Taliban to spread their struggle to Pakistan,
challenging the authority of the Pakistani government and creating a Pakistani version
of the Afghan Taliban. Sierra Leone observed an outbreak of conflict with a clear and
strong link with Liberia. The informal networks created among the leaders of the anti-
government movements in Sierra Leone and the insurgency in Liberia paved the road to
outbreak of conflict in Sierra Leone. These networks were not based on identity, but
rather on a shared ideology and class identification. The links between both groups
expanded to unprecedented size in correlation with the diamonds business. Diamonds
were the main sustaining force behind the conflict in Sierra Leone, and without the
informal links with Liberia, the diamond-based economy would have been impossible.

Tajikistan, on the contrary, has not suffered from conflict during the period
studied (1997-2009). No informal networks have been created between Afghanistan and
Tajikistan, and the contact between the countries seems mostly formal. Only the heroin
business has created informal networks, following the trend observed in all the cases
regarding illicit trade. Nevertheless, informal networks involved in illicit trade have not
demonstrated any explanatory power. All the cases investigated suffered the presence of
informal networks involve in illicit trade, but the relationship between illicit trade and
conflict seems to go further than pure enrichment and control of resources. In the Tajik
case identity plays an important role, since the links between the Tajik communities in
Tajikistan and Afghanistan are mostly based on official channels. The grievances and
ideologies of the Afghan Tajiks have not crossed the border, and when they have, they
have rapidly been assimilated by the government.

The situation observed in Guinea seems to belong to the extension of the Liberian
conflict rather than an internal conflict. It is debatable if the small conflict was part of
the state failure-related effects or only pure internationalization of a civil conflict. The
connections between Liberia and Guinea have followed formal channels, having the
Guinean government behind two different armed groups in Liberia. Notwithstanding
this, the failure in Liberia did bring informal networks to Guinea, but these networks
have not been able to take advantage of the Liberia failure. Nowadays, some of these
networks remain active in the Forest Region of Guinea, but the new political situation in
the country and the improvements in Liberia, which is not a failed state anymore, situate

- 56 -
the current situation outside the scope of this paper. As a reminder, it is important to
keep in mind the lack of access to information encounter in any research regarding
Guinea. The outbreak of conflict registered by the UCDP (2009) needs to be considered
carefully due to this lack of information.

6.1 Insurgency
For the four cases analysed in this paper, the creation of insurgency using the informal
networks between the failed state and the neighbouring state seems to be crucial to the
outbreak of war. As the theoretical framework pointed out, the creation of informal
networks taking advantage of the lack of security in the failed state opens an
opportunity for armed groups on both sides of the border. The use of this opportunity
will create conflict-enhancing mechanisms in the neighbouring country. The outbreak of
war in Pakistan is primarily due to the creation of informal networks between the
insurgency in Afghanistan and groups in Pakistan. The creation of informal networks
between these two groups can be observed in both directions of the mechanism. On one
side, groups from the failed state cross the border into the neighbouring countries,
enhancing the prospects of conflict outbreak. Afghan groups were able to settle in
Pakistan thanks to the existence of informal networks. These networks allowed the
Afghan groups to find refuge and hide among the population. The protection and help
provided through these informal networks created a safe haven for the Afghan groups in
Pakistan. The same situation is observed in Sierra Leone. The informal networks linking
Charles Taylor’s insurgency with the RUF allowed the NPFL several entrances into
Sierra Leone. NPFL troops were fighting alongside the RUF in Sierra Leone and the
withdrawal of these troops was directly felt by the RUF. This shows the importance of
the informal links between both groups for the outbreak and continuation of conflict in
Sierra Leone. It is worth noting that the actions of these networks were only possible
due to the lack of security in the failed state.

Insurgency groups from neighbouring countries use the failed state as a

sanctuary for their task. Both Pakistani and Sierra Leonean groups have used the failed
state as a safe haven, again utilizing the informal networks that were created in the
countries. Sierra Leone’s RUF was created in Liberia and launched its first attack on
Sierra Leonean soil from Liberia. Most of RUF’s troops were trained in Liberia and
used the failed state as a place to retreat. Pakistani insurgency movements did

- 57 -
apparently not launch their first attacks from Afghanistan; nevertheless, it is important
to note that most of the Taliban fighters in Pakistan received their baptism by fire in
Afghanistan. In addition, several Pakistani fighters have been found fighting along the
Taliban in Afghanistan. The networks created between the groups in Afghanistan and
Pakistan do not follow any official channel, and did not even do so under the Taliban
government. The link is based on the madrasas in Pakistan and the historically porous
border separating the countries. In the cases of Sierra Leone and Liberia, the informal
networks used were originally created in Libya, during the ‘revolutionary’ training of
Taylor and Sankoh. None of these networks follow any official path. The madrasas have
never been controlled by any other authority except the religious groups behind them.
The ideology and close collaboration between Afghan and Pakistani Taliban is deeply
rooted in the nexus created in the madrasas. The Libyan link is, on the same vein,
informal, since Libya has not been able to retain any control over the leaders ‘educated’
within their territory. Therefore, it can be argued that the insurgencies in Pakistan and
Sierra Leone were possible thanks to the informal networks created between these
groups and the groups in the failed state.

Tajikistan and Guinea have not seen insurgency movements in their territory in
relation to the failed state next to them. Tajikistan movements in Afghanistan have
crossed the border between both countries in both directions; nonetheless, the links
between groups in Tajikistan and groups in Afghanistan are purely formal. The
government of Tajikistan assisted and sustained the UIFSA-Northern Alliance,
formalizing the networks of this group with Tajikistan. Because of these official links,
no reason could be found by the UIFSA to challenge the authority of the Tajik
government in Tajikistan. In the case of Guinea, the same pattern is observable with the
ULIMO-LURD. The link between the group and the Guinean government had the same
effect as in the Tajik case, rendering any insurrection against their main sustainers
without motive. The very survival of these two groups was tied to the formal character
of these links. Interestingly enough, the trend found in Pakistan and Sierra Leone, where
armed groups used informal networks to create an insurgency, might be taking place in
Guinea. Some reports argue that ex-soldiers are being recruited in Liberia to fight
against the Guinean regime, taking advantage of the state of havoc still at place in some
parts of the country. The shift between formal and informal networks can threaten the
Guinean government.

- 58 -
 It is worth noting the important difference between the two possible directions of
the mechanism in Sierra Leone and Pakistan. In the Pakistani case, the main trigger
behind the outbreak of conflict has been the presence of Afghan Taliban in the FATA
and NWFP. The Taliban, perfectly intermingled with the local population and using the
informal networks to carry out their global struggle, challenged the authority Pakistan
held over FATA’s and NWFP’s territories. Pakistan suffered when groups from the
failed state entered their territory and enlarged the scope of their struggle. These groups
have become more Pakistani with time, and to this day, both Taliban – Pakistani and
Afghan – remain theoretically separated. This does not mean that the other direction of
the mechanism, groups from the neighbouring country using the failed state as a
sanctuary, did not play an important part in Pakistan. Pakistani Taliban have been
crossing the border almost on a daily basis, and have probably at some point been using
the ‘safety’ of Afghanistan to hide, retreat or acquire more supplies. In Sierra Leone, the
most important direction of the mechanism was the opposite, namely, armed groups
from the neighbouring country using the failed state as a safe haven. The conflict in
Sierra Leone broke out when RUF’s troops entered the country from Liberia. The RUF
used the havoc in the failed state to establish a safe haven. In Liberia, they were able to
train and retreat far from the reach of the SLA. Both directions of the causal mechanism
are related to the informality of the networks. Groups have entered Guinea and
Tajikistan from Liberia and Afghanistan, but the lack of informality of the networks that
allowed for such entrance has spared the countries from conflict.

Recapitulating, insurgency seems to demonstrate its importance as an explanatory

variable. The different outcome of the cases can partly be explained by the presence of
informal networks in Pakistan and Sierra Leone, and the lack of such networks in
Tajikistan and Guinea. It seems clear that informal networks create conflict-enhancing
mechanisms in two different directions in the neighbouring countries; armed groups use
the failed state as a safe haven and armed groups from the failed state enlarge the scope
of their actions and enter a neighbouring country. In the cases with no outbreak of
conflict, the baseline for conflict was relatively similar – nevertheless, the lack of
informal networks explains why both directions of the causal mechanisms are not
present. Without informal networks, armed groups cannot use the failed state as a safe
haven, which is clear in the Guinean case. Moreover, without informal networks, armed
groups from the failed state might not, find incentives to expand their struggle into the

- 59 -
neighbouring state, such as in the case of Tajikistan. It is not the existence of networks –
which existed in all the cases with varying intensity – that explains the different
outcome, but rather the informal character of these networks. Informal networks were
only found in Sierra Leone and Pakistan.

6.2 Identity
The influence of identity-related issues in the four cases varies. In the Afghan case,
identity is an important factor, and helps explain why the outcome is different than in
Pakistan and Tajikistan. Nevertheless, identity does not play an important role in
Liberia, explaining why the outcome is different than for Sierra Leone and Guinea. The
outbreak of conflict in Pakistan can partly be explained by identity factors. Pashtun
groups are present on both sides of the border, and the informal networks created among
both groups have been able to spread ideologies and grievances across the border.
Pashtun groups in Pakistan have adopted the struggle of their counterparts in
Afghanistan as their own battle. This element becomes crucial in explaining the
transmission of conflict-enhancing mechanisms from Afghanistan to Pakistan. Several
reports (ICG 2006, 2008, 2009b) argue that the Pashtun identity allowed a soft landing
of the Taliban in Pakistan, enabling the Taliban to mingle and hide within the Pashtun
population of the FATA and NWFP. The informal networks created and reinforced
during centuries between Pashtun enabled the transmission of conflict-enhancing
mechanisms from Afghanistan towards Pakistan. In Tajikistan, the lack of informal
networks averted the transmission of conflict-enhancing mechanisms from Afghanistan
towards Tajikistan. The Tajiks in Afghanistan have officially been supported by
Tajikistan, and the grievances of the Tajik population in Afghanistan were assimilated
by the Tajik government, whereas Pakistan did not assimilate the Afghan Pashtun
grievances. After the ouster of the Taliban in 2001, the Tajiks stand as the ‘winners’ of
the Afghan war, and Tajikistan’s government remains closely linked to this group in
order to extract benefits from their powerful position. Nevertheless, the situation is
starkly different in Pakistan, where the Afghan Pashtun are the ‘losers’ of the conflict.

Notwithstanding what is stated above, Sierra Leone and Guinea did not witness
identity-related conflict-enhancing mechanisms. The Liberian conflict did have an
ethnic component during the first stages of the war, and the failure of the country was
felt more prominently among certain ethnic groups. This ethnic character crossed into

- 60 -
Sierra Leone, but it was a mere reflexion of the Liberian conflict and did not last long
enough to be considered a crucial element. Although informal networks between ethnic
groups were present, the Mandingo people’s networks being the most important, the
inherent character and situation of this group averted any conflict-enhancing effect into
Sierra Leone. The failure in Liberia, with its subsequent conflict, and the conflict in
Sierra Leone were based on class rather than ethnicity. The identity variable does
therefore not wield an important influence in this case.

The only case where refugee-related effects are found is in Guinea. The
Mandingo population sough refuge in Guinea and a considerable percentage of them
remain in the country even after the end of the Liberian failure and conflict. This
presence has altered the ethnic composition of the Forest Region in Guinea, creating
tension between the Mandingo and the original inhabitants in the region. In addition to
this, informal networks of Mandingo people act on both sides of the borders, mostly in
business-related activities. This fact actually reinforces the importance and uniqueness
of the influence on state failure. The fact that Liberia is not a failed state anymore averts
the use of informal networks to create conflict-enhancing mechanism in the Forest
Region of Guinea.

The strength of the Identity variable can be debatable. Its influence and crucial
role in the case of the Afghan pair is clear: the informal character of the Pashtun
networks and the formal character of the Tajik networks partly explain the different
outcome observed in these two countries. Nevertheless, Identity did not play a crucial
role in the Liberian pair. It can be argued that the particularities of the Liberian case,
were the affiliations were mostly based on class – although some ethnic divisions
remain important and sometimes follow the same lines – might explain the lack of
influence of informal networks based on Identity. The lack of Identity-related issues
presents a clear weakening of the explanatory power of the proposed theoretical
framework

In summary, the creation of identity-based informal networks helps to explain

the outbreak of conflict in Pakistan since it created conflict-enhancing mechanisms.
Moreover, the lack of informal networks seems to be a factor in the lack of outbreak of
conflict in Tajikistan. However, the generalization power of this finding is limited since

- 61 -
identity-based informal networks did not play a significant role in either Sierra Leone or
Guinea, and it is not possible to explain the different outcome in the two West-African
countries using only this variable.

6.3 Illicit Trade

Diamonds and heroin have been pointed out as the underlying reasons behind the
conflict in Sierra Leone and Pakistan. Yet, Guinea and Tajikistan respectively have been
able to engage profusely in the diamonds and heroin business. Despite this, neither
diamonds nor heroin seem to be behind the outbreak of war in Sierra Leone and
Pakistan. Both resources have an impact in the prolongation of the armed strife,
diamonds having more salience in Sierra Leone than heroin in Pakistan. Apparently,
and based on the behaviour of the RUF during the first stages of the war, the conflict in
Sierra Leone did not break out as a conflict over control of resources. The RUF and its
informal networks in Liberia did not attempt to control the diamonds industry until
1995, almost 5 years after the outbreak of the conflict. The Taliban has a complex
relationship with heroin in Pakistan. The comprehensive control the Taliban exercise in
the FATA and NWFP includes control over the heroin. Heroin is taxed like any other
good crossing these regions, and the Taliban are obtaining benefits from this business.
In addition, the same networks that are behind the outbreak of war in Pakistan also
control the traffic of heroin and weapons between Afghanistan and Pakistan. However,
the Taliban’s ideology does not allow the cultivation or use of heroin, and heroin
cultivation was severely prosecuted during the time they controlled the government in
Afghanistan. Notwithstanding this, heroin does not seem to be a factor in the outbreak
of war in Pakistan, but rather a source of income for the Taliban.

The Illicit trade variable is apparently not a strong factor when it comes to the
outbreak of conflict. This idea gains strength when we look at Tajikistan’s involvement
in the heroin business. Informal networks have been created in Tajikistan in relation
with the heroin, this country being one of the main exit points of the Afghan heroin.
Nevertheless, no group has used this situation to fund their struggle against the Tajik
government. The IMU seems to be interested in the trade, and uses it as the main source
of income; however, it has not directly challenged the authority of Tajikistan and its
intentions towards the country remain unclear. A similar situation is observed in
Guinea, where informal networks were created in relation with the trade of diamonds;

- 62 -
these informal networks did nonetheless not provoke any outbreak of conflict in the
country. It can be concluded that illicit trade might be a good funding opportunity
during a conflict – as can be observed in Pakistan and Sierra Leone – but the presence
of informal networks within the illicit trade does not create conflict.

In summary, illicit trade seems to have little explanatory power when it comes to the
outbreak of conflict; nevertheless, armed groups take advantage of the havoc of the
failed states to fund their activities once the conflict is in progress. Informal networks
are present in this funding enterprise, but they seem to be used by armed groups rather
as a mean to continue their struggle than as a reason to challenge the authority of the
neighbouring state.

6.4 Critical Evaluation of the Findings

The research presented above needs to be evaluated critically in order to strength its
validity and generalization power. The outbreak of conflict in Pakistan and Sierra Leone
can be understood as an accumulation of grievances and protected conflict since 1950
(Fearon and Laitin 2003). At the same time, the cases of Pakistan and Sierra Leone will
perfectly fit Ayoob’s theoretical framework (1995). These two explanations stress the
importance of decolonization and inherent post-colonial weakness as a cause of war. In
addition, the cases of this study could be contemplated as part of a Regional Security
Complex (Buzan and Weaver 2003). The security of Pakistan and Sierra Leone is
intimately linked to their neighbours. Based on this idea, failure on their neighbours will
endanger their security and might generate outbreak of conflict. Nevertheless, the
theoretical approach of this study aims at enhancing the understanding of a particular
event –state failure– and tries to present its influence as a possible influential element in
the outbreak of conflict. Therefore, the findings of this study should not be considered
against other theories, but rather as a useful addition to them.

Concerning the methodology used, the use of qualitative studies implies a trade-
off between validity and generalization. In order to avert the lack of generalization
power as much as possible, the study attempted to study completely different cases.
Using this technique, regional, cultural and economic particularities could be avoided to
a certain degree. Nevertheless, the research findings have to be contemplated keeping in
mind the limitations of qualitative studies. The findings of this research point out some

- 63 -
causal mechanism that seems to be above the concrete particularities of each case, the
insurgency variable being the strongest one. However, the particularities of each case
make it impossible to draw a general extension of the conclusions to pertain to other
situations. The cases in this study were selected based on previous knowledge of the
circumstances and bearing in mind the resources and time limitations of the enquiry.
Notwithstanding this, the small amount of failed states and the early stages of the
research on the effects of state failure should bestow certain degree of power of
generalization to the findings of this research.

A second limitation of this project is found in the empirical information used to

analyse the underpinnings of the theoretical construct. Informality is already an evasive
subject of analysis, and even more so when only secondary sources can be used. In
addition, questions can always be raised regarding the impartiality and accuracy of any
secondary source used. This study has tried to avoid these shortcomings by using a
considerable range of sources with different origins.

Overall, this paper can be classified within the early stages of research on the
effects of state failure on neighbouring countries, and its systematic focus has been on
the causal mechanism affecting neighbouring states. The use of secondary sources, the
trade-off of generalization power, and the case selection has been done in order to better
use the resources available to shed light upon the field of research. The research lacks
the parsimoniousness of well-established theoretical constructs and it should be
considered as a first attempt to shed light upon the relationship between failed states and
their neighbouring countries from the perspective of their citizens. The robustness of
certain results, both positive and negative, should be further tested in other cases and
against other theoretical frameworks in order to achieve the richness of the
accumulative power of science.

- 64 -
 7. Conclusions

This research has attempted to enhance the scientific knowledge of state failure;
more concretely, to explore the influence of failed states on neighbouring countries. The
theoretical framework leading this paper argues that state failure spreads conflict-
enhancing mechanism through the creation of informal networks in three different areas,
Insurgency, Identity and Illicit Trade. The research has been guided by the hypothesis
explored in section 2, which states that the creation and use of informal networks
between the failed state and its neighbouring countries explains the outbreak of conflict
in some neighbouring states. Logically, according to the hypothesis, the lack of informal
networks explains the lack of conflict in some neighbouring states. In order to test the
hypothesis and refine the theoretical framework explaining the causal mechanism, two
cases have been analysed, Liberia and Afghanistan. The selection of the cases is based
on the dependent variable, the outbreak of conflict. As explained in Section 3, each
case has been divided in two units of analysis, one case with conflict and one case
without conflict. The unit of analysis is the relationship between the failed state and one
of its neighbours. In the case of Liberian, the selected units of analysis are Sierra Leone
and Guinea. In the case of Afghanistan, the units selected are Pakistan and Tajikistan. In
order to test the hypothesis and the theoretical framework, the three variables have been
scored in every unit of analysis, aiming at explaining the different outcome of the
dependent variable.

The empirical evidence gathered for this research show different validity for
each explanatory variable. The creation of insurgency-based informal networks seems
to explain the different outcome in the units of analysis. Pakistan and Sierra Leone, the
cases with conflict outbreak, suffered a widespread creation of insurgency-oriented
informal networks towards Afghanistan and Liberia, respectively. These networks
between the failed state and the neighbouring state seem to explain the outbreak of
insurgency in the country. At the same time Tajikistan and Guinea, the cases without
conflict, also lack insurgency-based informal networks with the failed state. In
Tajikistan and Guinea the insurgency movements acting in the failed state were
controlled or supported by the government, therefore, the networks were formal. The
governmental support averts any possibility of challenging the authority of the
government in the neighbouring country.

- 65 -
 The variable Identity shows a different outcome in the four cases. Its influence is
remarkable in the Afghan case, where the creation of identity-based informal networks
between Afghanistan and Pakistan party explains the outbreak of conflict. At the same
time, the lack of informal networks between Afghanistan and Tajikistan entails a lack of
conflict-enhancing mechanisms in Tajikistan. Nevertheless, the Liberian case does not
follow this trend, leaving the variable with an insignificant explanatory power in the
case. The understanding of identity-based informal networks leaves a open space for
further research

The last variable, Illicit Trade, does not show any explanatory power in any of the
cases. Informal networks seem to use illicit trade as a funding mean, but they do it
during the conflict rather than prior to it. The comprehensive picture of the structural
actions of the variables seems to indicate that illicit trade is used as part of the
insurgency strategy, and not as the main reason for the outbreak of conflict. However,
the strong influence of diamonds in Sierra Leone’s conflict might also be caused by
other intervening variables not contemplated in this research. Hence, further research is
needed in order to understand the complex relation between state failure and informal
trade in relation to conflict scenarios.

Policy Implications
The interest in failed states has been growing during the last five years, fostered by
events like the 9/11 and the problem with piracy in Somalia. This research points out to
some possible policy implications, mostly based on the role of informal networks
between failed states and neighbouring countries. It seems clear that state failure
increases the likelihood of conflict in neighbouring states, and this research has showed
an initial understanding of the causal mechanism behind this transmission. Thus, policy
makers should take into consideration the possible channels of transmission, increasing
their understanding of informal networks between the countries, mostly when these
informal networks are managed by armed groups. The systematic control and
understanding of armed groups and informal networks becomes crucial in order to avoid
conflict in neighbours of failed states.

- 66 -
Bibliography

Abdullah, Ibrahim (2004) ‘Bush Path to Destruction: The Origin and Character of the
RUF/SL’ in Abdullah, Ibrahim (ed) Between Democracy and Terror: The Sierra
Leone Civil War. Dakar: CODESRIA

Acharya, Arabinda, Syed Adnan Ali Bukhari and Sadia Sulaiman (2009) ‘Making
Money in the Mayhem: Funding Taliban Insurrection in the Tribal Areas of
Pakistan’ Studies in Conflict & Terrorism, 32:95–108,

Ayoob, Mohammed (1995). The Third World Security Predicament: State Making,
Regional Conflict, and the International System. Boulder, CO, Cologne and
London: Lynne Rienner

Bates, Robert H. (2008). When Things Fell Apart: State Failure in Late-Century Africa.
Cambridge: Cambridge University Press

Berdal, Mats and David M Malone (eds) (2000) Greed and Grievance: Economic
Agendas in Civil Wars. Boulder, CO: Lynne Rienner

Bhatia, Michael (2008) ‘Small Arms Flows into and within Afghanistan’ in Bathia,
Michael and Mark Sedra Afghanistan, Arms and Conflict: Armed Groups,
Disarmament and Security in a Post-war Society. New York: Routledge, pp 38-
71

Bøås, Morten (2001) ‘Liberia and Sierra Leone—dead ringers? The Logic of
Neopatrimonial Rule’ Third World Quarterly, Vol 22, No 5, pp 697–723

Bøås, Morten and Kahtleen M. Jennings (2005) ‘Insecurity and Development: The
Rhetoric of the ‘Failed State’. The European Journal of Development Research
17:3, 385-395

Buhaug, Halvard and Kristian S. Gleditsch (2008) ‘Contagion or Confusion? Why

Conflicts Cluster in Space’. International Studies Quarterly 52, 215-233

Buzan, B. and O. Weaver (2003), Regions and Power: The Structure of International
Security, Cambridge: Cambridge University Press

Carment, David (2003) ‘Assessing State Failure: Implications for Theory and Policy’.
Third World Quarterly Vol 24, No 3 407-427

Clapham, Christopher (2002) ‘The Challenge to the State in a Globalized World’.

Development and Change 33(5) 775-759

Clapham, Christopher (2004) ‘The Global-Local Politics of State Failure’ in Rotberg,

Robert I. (ed) When States Fail: Causes and Consequences. Princeton and
Oxford: Princeton University Press, pp 77-93

Collier, Paul and Anke Hoeffle (2004) ‘Greed and Grievance in Civil War’ in Oxford
Economic Papers 56, 563-95

- 67 -
Cornell, Svante E. (2005) ‘Narcotics, Radicalism, and Armed Conflict in Central Asia: The
Islamic Movement of Uzbekistan’ Terrorism and Political Violence 17:4,619 — 639
Crocker, Chester A. (2003) “Engaging Failing States”. Foreign Policy Vol. 82 No. 5,
32-44

Eck, Kristine (2005) ‘A Beginner’s Guide to Conflict Data: Finding and using the Right
Dataset’ UCPD Papers Nº 1. Uppsala: Department of Peace and Conflict
Research

Ellis, Stephen (1999) The Mask of Anarchy: The Destruction of Liberia and the
Religious Dimension of an African Civil War. London: Hurst & Company

Engwall, Johan (2005) ‘Stability and Security in Tajikistan: Drug Trafficking as a

Threat to National Security’ Arbetsrapporter, Working Paper No. 86. Uppsala:
Department of East European Studies

Esty, Daniel C. et al. (1995) State Failure Task Force: Phase I Findings. McLean, VA:
Science Applications International Corporation

Farah, Douglas (2001) ‘Al-Qaeda cash tied to diamonds trade: sale of gems from Sierra
Leone rebels raised millions, sources say’ Washington Post 2 November 2001,
available at http://www.washingtonpost.com/ac2/wp-dyn/A27281-2001Nov1
(accessed March 12th 2009)

Fearon, J.D. and D.D. Laitin (2003) ‘Ethnicity, Insurgency and Civil War’ American
Political Science Review Vol. 97 (1) pp 75-90

Gberie, Lansana (2005) A Dirty War in West Africa: The R.U.F. and the Destruction of
Sierra Leone. London: Hurst & Company

George, Alexander L. and Andrew Bennett (2005) Case Study and Theory Development
in the Social Science Cambridge and London: MIT Press

George, Alexander L. and Timothy J. McKeown (1985) ‘Case Studies and Theory of
Organizational Decision Making’ in Advances in Information Processing in
Organization. Volume 2. Greenwich: JAI Press Inc.

Glenditsch, Kristian S. and Idean Salehyan (2006) ‘Refugee Flows and the Spread of
Civil War’ in International Organization

Glenditsch, Kristian S. et al (2008) ‘Fighting at Home, Fighting Abroad: How Civil

Wars Lead to International Disputes’ Journal of Conflict Resolution; 52; 4, 479-
506

Goldstone, Jack A. (2008) ‘Pathways to State Failure’ in Conflict Management and

Peace Science, 25:4, 285-297

Greenfield, Victoria A., Letizia Paoli, Irina Rabkov and Peter (2007) ‘Tajikistan: The
Rise of a Narco-State’ Journal of Drug Issues Volume: 37:4 , 951

- 68 -
Gunaratna, Rohan and Anders Nielsen (2008) ‘Al Qaeda in the Tribal Areas of Pakistan
and Beyond’ in Studies in Conflict & Terrorism, 31:775–807

Hameiri, Shahar (2007) ‘Failed States or a Failed Paradigm? State Capacity and the
Limits of Institutionalism’ Journal of International Relations and Development
10, 122–149.

Herbst, Jeffrey (2004) “Let Them Fail: State Failure in Theory and Practice:
Implications for Policy” in Rotberg, Robert I. (ed) When States Fail: Causes and
Consequences. Princeton and Oxford: Princeton University Press, pp 302- 318

Hilali, A.Z (2002) ‘The Cost and Benefits of the Afghan War for Pakistan’
Contemporary South Asia , 11(3), 291–310

Holsti, Kalevi (1996) The State, War and the State of War, Cambridge: Cambridge
University Press

Howard, Tiffiany O. (2008) ‘Revising State Failure: Developing a Causal Model of

State Failure Based Upon Theoretical Insight’. Civil Wars, 10:2, pp 125-146

HRW (2002) ‘Guinea: Liberian Refugees in Guinea: Refoulement, Militarization of

Camps, and Other Protection Concerns’ Human Rights Watch Vol. 14, No. 8
(A)

_____ (2005) ‘Youth, Poverty and Blood: The Lethal Legacy of West Africa's Regional
Warriors’, Human Rights Watch, vol. 17, no. 5(A), 13

Hussain, Rizwan (2005) ‘Pakistan and the Emergence of Islamic Militancy in

Afghanistan’. Hampshire & Burlington: Ashgate

ICG (2001) ‘Tajikistan: An Uncertain Peace’ Asia Report Nº 30. Osh & Brussels:
International Crisis Group

_____ (2002) ‘The IMU and the Hizb-ut-Tahrir: Implications of the Afghanistan
Campaign’ Asia Briefing Paper, Osh & Brussels: International Crisis Group

_____ (2003) ‘Guinee: Incertitudes Autour d’une Fin de Regne’ Africa Report Nº 74.
Freetown & Brussels: International Crisis Group

_____ (2005) ‘Stopping Guinea’s Slide’ Africa Report Nº 94. Dakar & Brussels:
International Crisis Group

_____ (2006) ‘Pakistan’s Tribal Areas: Appeasing the Militants’ Asia Report N°125.
Islamabad & Brussels: International Crisis Group

_____ (2008) ‘Taliban Propaganda: Winning the War of Words?’ Asia Report Nº 158.
Kabul & Brussels: International Crisis Group

- 69 -
_____ (2009) ‘Guinea: The Transition Has Only Just Begun’ Africa Briefings 58.
Brussels & Dakar: International Crisis Group

_____ (2009b) ‘Pakistan: The Militant Jihadi Challenge’ Asia Report Nº 164. Islamabad
& Brussels: International Crisis Group

_____ (2009c) ‘Tajikistan: On the Road to Failure’ Asia Report Nº 162. Dushanbe &
Brussels: International Crisis Group

IRIN (2005) ‘Guinea: Living on the Edge’. IRIN Web Special Report. Integrated
Regional Information Network

_____ (2007) ‘Liberia: Refugee returns creating ethnic “time bomb”. 8 February.
Integrated Regional Information Network. Date of Retrieval: 12-03-2009

Iqbal, Zaryab and Harvey Starr (2008). ‘Bad Neighbors: Failed States and Their
Consequences’. Conflict Management and Peace Science, 25:4, pp 315-331

Jonson, Lena (2006) Tajikistan in the New Central Asia: Geopolitics, Great Power
Rivalry and Radical Islam London & New York: I.B Taurus

Kasfir, Nelson (2004) ‘Domestic Anarchy, Security Dilemmas and Violent Predation:
Causes of Failure’ in Rotberg, Robert I. (ed) When States Fail: Causes and
Consequences. Princeton and Oxford: Princeton University Press, pp 53-76

Kepel, Gilles(2002). Jihad: The Trail of Political Islam. Cambridge: Belknap Press of
Harvard University Pres

King, Gary, Robert O. Keohane, and Sidney Verbe (1994) Designing Social Inquiry
Princeton: Princeton University Press

Klare, Michael T. (2004) ‘The Deadly Connection: Paramilitary Bands, Small Arms
Diffusion, and State Failure’ in Rotberg, Robert I (ed). When States Fail: Causes
and Consequences. Princeton and Oxford: Princeton University Press, pp 116-
134

Lake, David A., and Donald Rothchild. (1998). ‘Spreading Fear: The Genesis of
Transnational Ethnic Conflict’ In Lake, David A and Donald Rothchild (ed) The
International Spread of Ethnic Conflict. Princeton, NJ: Princeton University
Press.

Lijphart, Arend (1971) ‘Comparative Politics and Comparative Methods’ American

Political Science Review 65 (3) pp 682-693

Milliken, Jennifer and Keith Krause (2002) ‘State Failure, State Collapse and State
Reconstruction: Concepts, Lessons and Strategies’. Development and Change
33(5). pp 753-774

- 70 -
Most, Benjamin A. and Harvey Starr (1980) ‘Diffusion, Reinforcement, Geopolitics and
the Spread of War’ The American Political Science Review Vol 74 nº 4 pp 932-
946

Murshed, S. Iftikhar (2006) Afghanistan: The Taliban Years London: Bennett and
Bloom

Nilsson, R. Anders (2008) Dangerous Liaisons: Why Ex-Combatants Return to

Violence, Cases from the Republic of Congo and Sierra Leone. PhD
Dissertation. Uppsala: Department of Peace and Conflict Research

Ohlson, Thomas (2008) ‘Understanding Causes of War and Peace,’ European Journal
of International Relations, vol. 14, no. 1, pp. 133-159.

Priego, Alberto (2008) ‘The Drug Business in Central Asia’ (El Negocio de la Droga en
Asia Central) Date of retrieval: 23-03-2009. Madrid: Real Instituto Elcano ARI
132/2008.

Rais, Rasul Bakhsh (2008) Recovering the Frontier State: War, Ethnicity and State in
Afghanistan Plymouth: Lexington Books

Rashid, Ahmed (2002) ‘Jihad: The Rise of Militant Islam in Central Asia’ New Haven
& London: Yale University Press

Rashid, Ismail (2004) ‘Student Radicals, Lumpen Youth and the Origins of
Revolutionary Groups in Sierra Leone, 1977-1996’ in Abdullah, Ibrahim (ed)
Between Democracy and Terror: The Sierra Leone Civil War. Dakar:
CODESRIA, 66-89

Reno, William (1998) ‘Warlord Politics and African States’ Boulder & London: Lynne
Rienner.

_____ (2003) ‘Sierra Leone: Warfare in a Post-State Society’ in Rotberg, Robert I. (ed)
State Failure and State Weakness in a Time of Terror. Washington D.C:
Brookings Institution Press. pp 71-100

Richards, Paul (1996) Fighting for the Rainforest. War, Youth & Resources in Sierra
Leone. Oxford & Portsmouth: James Currey & Heinemann

_____ (2005) ‘West-African Warscapes: War as Smoke and Mirrors: Sierra Leone 1991-2,
1994-5, 1995-6’ in Anthropological Quarterly, Vol. 78, No. 2 (Spring, 2005), 377-402

Rotberg, Robert I. (2002) ‘The New Nature of Nation-State Failure’. In The Washington
Quarterly 25:3 pp 85-96

Rotberg, Robert I. (2004) ‘The Failure and Collapse of Nation-States: Breakdown,

Prevention, and Repair’ in Rotberg, Robert I (ed). When States Fail: Causes and
Consequences. Princeton and Oxford: Princeton University Press, pp 1-50

- 71 -
Salehyan, Idean (2007) ‘Transnational Rebels: Neighboring States as Sanctuary for
Rebel Groups’ World Politics 59 ( January 2007), 217–42

Sanderson, Thomas M. (2004) ‘Transnational Terror and Organizaed Crime: Blurring

the Lines’ SIAS Review, vol 25, no 1, Winter-Spring 49-61

Schutz, Barry (1995) ‘The Heritage of Revolution and the Struglle for Governmental
Legitimacy in Mozambique’ in Zartman, I. William (ed) Collapsed States: The
Disintegration and Restoration of Legitimate Authority Boulder, CA/London:
Lynne Rienner, pp 109-124

Sen, Amartya (1999) Development as Freedom. New York: Anchor Books

Shaw, Mark (2006) ‘Drug Trafficking and The Development Of Organized

Crime in Post-Taliban Afghanistan’ in William A. Byrd and Doris Buddenberg
(ed) Afghanistan’s Drug Industry: Structure, Functioning, Dynamics and
Implications for Counter-Narcotics Policy. Washington D.C: The World Bank &
UNODC, pp 189-213

Sinno, Abdulkader H. (2008) ‘Organizations at War in Afghanistan and Beyond’ Ithaca

and London: Cornell University Press

Skaperdas, Stergios (2001) ‘The Political Economy of Organized Crime: Providing

Protection when the State does not’ Economics of Governance 2:173-202

Skocpol, Theda (1979) States and Social Revolutions: A Comparative Analysis of

France, Russia and China. New York: Cambridge University Press

Smillie, Ian, Lansana Gberie and Ralph Hazleton (2000) The Heart of the Matter:
Sierra Leone, Diamonds and Human Security. Ottawa: Partnership Africa
Canada.

Tilly, Charles (1990) Coercion, Capital and European States. Oxford: Blackwell

UCDP (2009) ‘Guinea: Conflict Summary’. Date of retrieval: 12-03-2009. Uppsala:

Department of Peace and Conflict Research, Uppsala University

_____ (2009b) ‘Liberia: Conflict Summary’ Date of retrieval: 10-03-2009. Uppsala:

Department of Peace and Conflict Research, Uppsala University

_____ (2009c) ‘Pakistan: Conflict Intensity and Warring Parties’ Date of retrieval: 17-
03-2009. Uppsala: Department of Peace and Conflict Research, Uppsala
University

UN (2000) ‘Report of the Panel of Experts Appointed Pursuant to Security Council

Resolution 1306 (2000), Paragraph 19, in Relation to Sierra Leone’
S/2000/1195. United Nations Security Council.

- 72 -
UNHCR (2005) ‘UNHCR deplores forced return of Afghan refugee family from
Tajikistan’ Geneva: UN High Commissioner for Refugees. Available at
http://www.unhcr.org/cgi-bin/texis/vtx/afghan?page=news&id=432ad0ef4
(accessed March 24th 2009)

Van de Walle, Nicolas (2004) ‘The Economic Correlates of State Failure: Taxes,
Foreign Aid and Policies’ in Rotberg, Robert I. (ed) When States Fail: Causes
and Consequences. Princeton and Oxford: Princeton University Press, pp 94-115

Wannenburg, Gail (2005) Africa´s Pablos and Political Entrepeneurs: War, the State
and Criminal Networks in West and Southern Africa’ Johannesburg: The South
African Institute of International Affairs

Weber, Max (1958) From Max Weber edited by H.H Gerth and C.W Mills. New York:
Galaxy

Williams, Gabriel I.H (2002) ‘Liberia: The Heart of Darkness’ Victoria: Trafford

Yannis, Alexandros (2002) ‘State Collapse and its Implications for Peace-Building and
Reconstruction’. In Development and Change 33(5), pp 817-835

Zartman, I. William (1995) ‘Introduction: Posing the Problem of State Collapse’ in

Zartman, I. William (ed) Collapsed States: The Disintegration and Restoration
of Legitimate Authority Boulder, CA/London: Lynne Rienner, pp 1-14

- 73 -
- 74 -