‘GIVE WAR A CHANCE’: ALL OUT WAR AS A
MEANS TO ENDING CONFLICT
Comparing Sri Lanka and Colombia
A Research Paper presented by:
Fabio Andres Diaz
(Colombia)
in partial fulfilment of the requirements for obtaining the degree of
MASTERS OF ARTS IN DEVELOPMENT STUDIES
Specialization:
Conflict Reconstruction and Human Security
(CRS)
Members of the examining committee:
Professor MansoobMurshed
Professor Helen Hintjens
The Hague, TheNetherlands
November, 2011
Disclaimer:
This document represents part of the author’s study programme while at the
Institute of Social Studies. The views stated therein are those of the author and
not necessarily those of the Institute.
Inquiries:
Postal address:
Institute of Social Studies
P.O. Box 29776
2502 LT The Hague
The Netherlands
Location:
Kortenaerkade 12
2518 AX The Hague
The Netherlands
Telephone:
+31 70 426 0460
Fax:
+31 70 426 0799
ii
Contents
List of Tables
List of Figures
List of Acronyms
Abstract
v
v
vi
vii
Chapter 1. Introduction
1.1 Introduction
1.2 Two countries in a nutshell
1.3 What it is all about: all-out War
1.4 Validity and justification
1.5 Objectives
1.6 Research questions and working hypotheses
1.7 Structure of the document
1
1
1
3
4
4
5
6
Chapter 2. Colombia and Sri Lanka: from Colony to all-out war
2.1 Introduction
2.2 Colonial times: seeds of destruction?
2.3 The struggle after independence: looking for a nation and finding
civil war
2.4 War as a phoenix phenomenon
2.4.1 In Colombia: deception, reincarnation and rebirth
2.4.2 In Sri Lanka: fight to the bitter end
2.5 Post 9/11: Justification, endogeinity, and discourse
2.5.1 Colombia: from guerrilla to narco-terrorist
2.5.2 Sri Lanka: from Tamil tiger to Tamil terrorist
2.6 Conclusion: new wars and the offspring of 9/11
7
7
7
9
11
11
13
14
14
15
15
Chapter 3. A theory of practice? The All-out war solution.
3.1 Introduction
3.2 All-out war: a solution for failed peace processes?
3.2.1 Benefits and costs of reneging
3.2.2 The discount rate
3.2.3 Inability to commit
3.2.4 Indivisibility issues
3.2.5 Credibility problems and trust
3.2.6 Difficulty of enforcing commitments
3.3 All-out war –a conceptual primer?
3.3.1 The empirical evidence
3.3.2 Back to basics: ‘old theories’ revisited
3.4 Achievement of ‘peace’through war: Sri Lanka and Colombia
17
17
17
17
18
18
18
18
19
19
20
20
20
iii
3.5
3.4.1 In Colombia: continuity yet military build –up?
3.4.2 In Sri Lanka: blitzkrieg and scorched earth
Conclusion
Chapter 4. Theory,contradictions and practice: ontologies, ‘state’
making, and development
4.1 Collateral damage: weak foundations of the all-out war theory
4.2 Big Bang: ‘‘Tillian’’ wars
4.2.1 Colombia: The good ‘‘Tillian’’ student with bad grades
4.2.2 Sri Lanka: The bad ‘‘Tillian’’ student with good grades
4.2.3 In summary: can Colombia and Sri Lanja be described as
‘successful’ ‘Tillian’ wars?
4.3 Death (practice): is war making a particular model of development
and state making?
4.3.1 The ‘liberal peace’
4.3.2 The ‘illiberal’ peace?
4.4 Conclusion: Biases and implications
22
23
25
26
26
27
28
29
31
32
32
33
35
Chapter 5. Conclusions and reflections
36
References .
38
iv
List of Tables
Table 2.1 An introduction to the Colombian and Sri Lankan conflicts
Table 2.2 Colombian and Sri Lankan conflicts through the lenses of
different questions
Table 3.1 Possible positive externalities of the All-out War
8
9
19
List of Figures
Figure 2.1 Map of Colombia showing the DMZ
Figure 2.2 Map of Sri Lanka showing the area claimed as Tamil homeland
Figure 3.1 Military spending as proportion of the GDP
Figure 2.2 Military expenditure of Colombia and Sri Lanka
Figure 3.3 Size of the insurgent forces
Figure 3.4 Size of military forces in Colombia and Sri Lanka
Figure 3.5 GDP growth in Colombia and Sri Lanka
Figure 3.6 Battle deaths related to the Colombian conflict (according to
two different sources)
Figure 3.7 Battle deaths related to the conflict of Sri Lanka
Figure 4.1 Quintessential elements of the ‘give war a chance’ theory
Figure 4.2 Size of the military forces, government expenditure and revenue
in Colombia
Figure 4.3 Size of the military forces and revenue in Sri Lanka
Figure 4.4 Revenueand taxes on income in Sri Lanka
Figure 4.5 Government debt
Figure 4.6 Economic openess
Figure 4.6 Political terror scale for Colombia and Sri Lanka
Figure 4.7 Displaced population of Colombia
v
12
13
21
21
21
21
22
23
24
28
29
30
30
33
33
33
34
List of Acronyms
FARC
ELN
AUC
CGSB
JPV
JHU
LTTE
SNL
UNP
PTOMS
SLMM
EU
US
OAS
DMZ
HRW
ICG
ICRC
UNHR
UNHCR
CSO
DDR
GDP
-Fuerzas Armadas Revolucionarias de Colombia
-Ejercito de LiberacionNaciona
-AutodefensasUnidas de Colombia
-CoordinadoraGuerrillera Simon Bolivar
-JanathaVimukthiPeramuna
-JathikaHelaUrumaya
- Liberation Tamils of Tamil Elam
- Sri Lanka NidahasPakshaya
-United National Party
- Post Tsunami Operation Management Structure
-Sri Lankan Monitoring Mission
-European Union
-United States
-Organization of American States
-De-militarized zone
-Human Rights Watch
-International Crisis Group
-International Commission of the Red Cross
-United Nation Human Rights Commission
-United Nations High Commissioner for Refugees
-Civil society organization
- Disarming, demobilization and reintegration
-Gross domestic product
vi
Abstract
This paper investigates the emergence of a military approach as a means to
solving protracted civil conflicts in the particular cases of Sri Lanka and Colombia. The article attempts a comparative study of the military alternatives
emerging as an end to civil war in both countries. The approach adopted is to
study the emergence of these military options within the context of each country’s history and to assess whether the call for war was merely a consequence
of the war on terror, or driven by internal elements. The paper explores the
epistemological groundings and pitfalls of the all-out war theory informing this
approach, before reassessing the significance and validity of the theory in relation to Sri Lanka and Colombia. Finally, the liberal peace framework is used to
approach an understanding of how development is being conceptualized
through the practice of the all-out war theory in these two countries.
In order to do so, this document performs a comparative analysis, as well
as an historical study of the evolution of both conflicts, incorporating elements
of discourse analysis. The document also explores the notion of ‘‘Tillian’’ wars
from an agent based perspective, not only to establish the logic and validity of
these approaches, but also as a means to understanding possible solutions to
protracted and intractable wars.
Keywords
Civil war, conflict resolution, peace settlement, military victory, agent based
theories, Colombia, Sri Lanka, protracted conflicts.
vii
Chapter 1. Introduction
1.1 Introduction
Protracted and intractable conflicts (Coleman, 2000; Zartman, 2005) present a major
challenge to scholars studying civil wars, and ‘complex’ wars1 (Kaldor, 1999; Kalivas, 2001; Kalivas, 2010). Their resolution and comprehension tends to be difficult and complicated, with no
accepted frameworks for understanding why peace initiatives so often fail (Kalivas, 2001). The
usual frameworks proposed for conflict solution include: negotiations, third party intervention
and mediation, development programmes, as well as policies geared towards social change which
eliminates the social injustices and structural violence thought to underpin violence and grievances (Ballentine, 2003; Jacoby, 2008). However, these efforts have not proven to be effective
methods of achieving peace in very protracted conflict situations like Sri Lanka and Colombia
(Sørensen, 2001; Toft, 2010; Walter, 2004; Lutwakk, 1999). Therefore, this study reassesses an
old framework. What about reconsidering the military solution’s ‘end’ to civil violence; the imposition of peace through force of arms? In other terms, this study will look at war as a means to
achieving peace in civil war2environments.
To explore this question, I present here a comparative study of war as a means to achieving peace in Sri Lanka and Colombia. The military solution is considered by some to have
brought about an ‘end’ to 26 years of civil war in the former (Arambewela, 2010; Shastri, 2009;
Devotta, 2009). In the case of the latter, the Colombian conflict has dragged on for more than 50
years(Chernik, 2005; Sanin, 2005; Bushnell,1996), with a policy of all-out war having been in
place for the last 8 years (2002-2010). This study will explore, compare and contrast both cases.
The aim is to deepen our insight into the underpinnings of how military approaches have gained
momentum in each country, what the logic was behind this shift in policy towards ‘peace’ (in
both cases the military option emerged after a failed peace process). The examples demonstrate,
in the case of Sri Lanka, a ‘successful’ military option, and in the case of Colombia, what might
be considered a ‘marginally successful military option.’
1.2 Two countries in a nutshell
Sri Lanka’s civil war can be defined as a ‘classical’ civil war (Korf, 2006; Orjuela, 2003).
Sri Lanka gained independence from Britain in 1948 (Devotta, 2000: 60). Prior to independence,
the country was governed in the main by a Tamil minority3.The years following independence
saw a drive to balance ethnically-based inequalities (Frerks, 2005) that gradually reduced the
Tamil minority to a less empowered condition. ‘Positive discrimination’ policies (Shasthi, 2009)
were introduced with the intention of marginalizing the previously advantaged section of the
population (Devotta, 2005; Frerks, 2005) andTamil was not recognized as the official language of
the country.Gradual impoverishment in Sri Lanka, coupled with the incapacity of the political
system to embrace and solve social disputes, increased the tensions that existed between the Sinhalese and the Tamil in the 1970’s and 1980’s, and as a consequence saw different riots in which
Sinhalese or Tamil were victims(Devotta, 2000).
Civil wars or complex wars can be two different categories ascribed to the same phenomena, but literature appears to draw a distinction between the two. For a discussion on the topic see Kalivas, 2010.
2The document will refer to civil wars as war.
3In Sri Lanka, the Tamil population of Sri Lankan origin is around 12%, while the Sinhalese population is
close to 74%. Other ethnic groups are the Burger, Malay and Indian Tamils (Devotta, 2005).
1
1
As the environment became more radical, armed groups emerged among the Sinhalese
and the Tamil, the Tamil with the LTTE and several other groups, and the Sinhalese with the
JPV (Shastri, 2009; Hoole, 2009). Both groups to some extent claimed universality on their right
for a Tamil state (or a greater Tamil autonomy within a Federal Government) (Hoole, 2009;
Korf 2006) or an exclusively Sinhalese state (Orjuela, 2003). Since its inception in 1983, the
LTTE waged an irregular-regular war against the government (Uyangoda, 2007), while other
groups such as the JPV became political parties or were destroyed in the carnage of the conflict.
The LTTE controlled territories on the north and the east of the country, where the majority of
the country’s Tamil population reside (Korf, 2006).
As the conflict unfolded, it became apparent that the Tamil would be difficult to halt or
defeat, given their combination of irregular- regular tactics, such as suicide bombings, large scale
military attacks on government garrisons, naval attacks, etc. A number of times, matters reached
an apparent ‘mutual hurting stalemate’ (Zartman, 1995), that resulted in several failed negotiation
attempts.The most notable of these occurred in 2002, and saw the international community playing an important role(Lewis, 2010). Following this peace process, public opinion and politicians
changed inattitude towards the Tamil and the LTTE.In 2007, hard line policies and support for a
military solution to the ‘Tamil issue’gained momentum, resulting in the military defeat of the
LTTE in 2009, the imprisonment of the majority of their soldiers, and the death of most of their
leaders.
One of the major concerns around the policy of war as a means to achieving peace has
been around the highly gendered and militaristic role of the army. Sri Lanka is an example of a
country where militaristic action, couched in humanitarian language, has seemed to raise the status of the army and of the military (de Mel, 2009). Similarly, in Colombia, nationalism and praise
of the military forces increased in the period of 2002-2010 (Rojas, 2009).
On the other hand, since 1964, the Colombian armed conflict has involved violence between left-wing guerrillas Richani, 2002), and the Colombian government, the FARC4, and the
ELN5. For some authors, the efforts of the guerrillas have always been interpreted as an attempt
to overthrow the state. In contrast to Sri Lanka, Colombia can be framed as a ‘new war’, where
the existence of armed groups might be explained by grievances related to land ownership
(Richani, 2007)and previous inter-party violence over the control of the political power in the
country (Palacios, 1995).Armed groups resort to kidnapping, drug trafficking, and taxation of
commodities (Leal, 2006).
The violence also stems from right wing groups- emerging bands known as paramilitaries
(Chernik, 2005;Pardo, 2007; Romero, 2007). These groups appeared as a response to left wing
guerrilla violence, but also as a by-product of drug lords’ attempts to secure their assets from the
government and the guerrillas and in some extent of their typical use of violence to acquire
wealth (Duncan, 2005). The groups are quite dispersed, and the label of ‘paramilitaries’ functions
more as a franchise than as a simple unified category. A comparison of the Colombian and Sri
Lankan conflicts seems to suggest that the Colombian case is characterized by a higher degree of
complexity. Drug production and trafficking have become major financers fuelling the fighting
in Colombia (IEPRI, 2006).
As the conflict waged, the nature of the state changed.Policies aimed at creating a greater
legitimacy and efficiency in the government (decentralization)had the side effect of exacerbating
the effective co-optation by armed groups of state institutions at local and regional levels
(IEPRI, 2006, Duncan, 2004; Lopez, 2010). For some authors, the Colombian conflict is under-
Fuerzas Armadas Revolucionarias de Colombia
5 Ejército de Liberación Nacional
4
2
stood as being grievance motivated, and for others as being greed driven, or even as both (Posada-Carbó, 2001; Bushnell, 1993; Pecaut; 2008). Similarly to Sri Lanka, in the Colombian case,
after several failed peace attempts and different sets of negotiations (roughly one round of negotiations per decade), in 2002, the Colombian government finally opted for a military-focused policy.
This military option was supported by external donors, particularly the U.S. The strategy
involved raising the strategic capabilities of the military, attacking theguerrilla’s military units, and
weakening their income structure, which was strongly based on illicit crops and cocaine (Rojas,
2007) thereby forcing the guerrilla groups to negotiate a peace settlement. The implied argument
behind this policy was that once the guerrillas had been defeated, paramilitary forces would no
longer have any objective reason to exist, and the path to peace could be paved. The current results of these policies can be seen in a weakening of the guerrillas (Romero, 2007).On the other
hand, the spaces abandoned by guerrillas have been co-opted by paramilitaries that have extended their influence across the country, even permeating the political system (Duncan, 2005). Despite the weakening observed, it appears that a military defeat of the guerrillas will be hard to
achieve. On both sides, tactics and strategies have adapted to the new environment, making a
possible peace hard to foresee.
1.3 What it is all about: All-out War
Some theorist of conflict and state-making suggest that civil war will only end when the
costs, however defined, of waging war become prohibitive for the major parties involved(Toft,
2010;Walter, 2004; Fortna, 2003; Fearon, 2004). According to this view, if, and only if, one side
can militarily dominate other parties, then the level of violent interactions between state and
non-state combatant forces will tend to decline or even end (Cohen, 1981: 904; Jacoby, 2008;
Luttwak, 1999; Tilly, 1991). According to the military logic of this argument, in order to achieve
a decline in or termination of hostilities, it is necessary to use large scale violence against the military opponents (Sørensen, 2001;Fearon, 2004).One reason for the long duration of civil wars is
that neither side can disarm or defeat the other. This causes a military stalemate which can be
broken relatively quickly when conditions favor an apparent victory by one side or the other
(Fearon, 1995) or a possible marginal gain on the breakout from the negotiations (Murshed,
2008; Beardsley, 2008; Collier, 2008).
Luttwak (1999) argues that war can actually also be a means for social parties (the question is which ones) to achieve new societal settlements. War may be even more effective in obliging warring parties to comply to the settling of a peace agreement. This agreement is likely when
a party realizes that they cannot win,and cannot accept the price they are paying for losing(Cohen, 1981;Huth, 1988; Korf, 2006; Goodhand, 2009).
Some studies actually argue that peace settlements achievedthrough military victory are
more lasting, and have a lower probability of recurrence/re-emergence, therefore proving to be
more ‘sustainable’ than normal peace processes (Toft, 2010; Walter, 2009;Quinn, 2007). Additionally, a military effort can support the process of state making; money is needed to wage war,
and countries that increase their military capacity therefore accompany these processes with
more effective taxation and a more visible state presence (Tilly,1991; Mann, 2002).
It is possible that these policies could have the negative effect of creating an inertia surrounding the conflict which would make it difficult to de-escalate matters (Garrison, 2009;
Zartman, 1991). Additionally, the winning of a war does not necessarily make a military option
the better solution, nor does it mean that violence is the only option for settling disputes within a
country. The military path can also drive societies towards a militarization of their state, putting
in place the makings of an autocratic regime (Korf, 2006; Höglund, 2010). In other cases, groups
may continue, and even escalate, fighting, if this is strategically more advantageous than exposing
3
their weaknesses at the negotiation table (Harris, 2011; Kriesberg, 1998). Despite the dangers of
escalation, advocates frequently intentionally escalate a conflict, because although war is a great
evil, it can also result in the greater virtues of resolving political conflicts and leading to peace
(Kurtz, 2009; Luttwak, 1999).
1.4 Validity and Justification
An important element to consider in comparing Colombia with Sri Lanka is the value
gained in terms of knowledge. It is interesting to note that in both cases the solution ‘implemented’ by the governments denies/denied any political legitimacy to the claims of the irregular
forces, labelling these groups as ‘terrorists’ or ‘narco-terrorists’.In both cases the decision to engage in full-on war followed a failed negotiation attempt andwas supported by, and frames itself
within, the discourse of the war on terror. Despite the fact that Sri Lanka and Colombia experienced what appeared to be an ethnic based conflict and a class based conflict (in one of the most
unequal countries in the world) respectively, in both cases inequality seems to have increased after the war6. The policies implemented in these countries have become a case study that has informed discussions on how to deal with irregular forces in military academies (Sepp, 2005;
Kiszely, 2006; Gorka, 2011; Mills, 2007; Metz, 2007; Kinroos 2004; Kilcullen, 2006). They have
likewise informed the policy making processes of other countries facing security issues/insurgencies7 (for example, the Colombian government now advises central American governments on strategies for fighting drug traffickers and terrorists).
Therefore, assessing the military campaigns to end violence in Sri Lanka and in Colombia
can shed light the effectiveness of these policies. In addition, the study aims to critically question
such approaches, and the underlying assumptions that allow the comparison of one kind of military approach with another in terms of their ‘efficacy’ in ending warfare or violence. To some
extent this study will suggest that whilst such approaches may end war, they may not resolve
many of the underlying elements that fuelled violence in the first place (chapter 4).
The aim of the study is also to contrast the ‘facts’ and evidencethat inform different performance indicators. This investigation follows from the hypothesis (not to be assessed in this
work) that the way we measure results, a consequence of our epistemological paradigms and biases, affects our understanding of the efficacy and efficiency of policies. An example of the way
in which these embodied biases affect our understanding can be foundin the difficulty of defining ‘war’, a complex and ambiguous construct (see Sambanis, 2004 for a discussion on the topic).
The difficulty in arriving at an absolute definition of ‘war’ lies not only in the nature or ‘what is’
element, but also in thecircumstantial, the ‘when/why’ of a war. The nature of conflict means
that the accuracy of information pertaining to wartime is not always reliable, and it is therefore
wisest to analyse conflicts from several epistemologies, databases and angles in an attempt to
shed more light on them.
1.5 Objectives
The investigation aims to compare and analyse the military solution to protracted conflicts
theory and relate it other approaches to peace-making. In particular, the investigation aims to
explore the particular cases of Sri Lanka and Colombia. In achieving this aim, I propose to fulfil
the following objectives:
6Worldbank
(2011), Data retrieved from http://data.worldbank.org/country/
In this work I wish to clarify that for me the war on Afghanistan and in Iraq are not conventional guerrilla/separatists movements, and belong to particular dynamics and logics beyond the scope of this investigation.
7
4

Trace the emergence and implementation of the all-out war theory in the Colombian and
Sri Lankan contexts, and compare and contrast the practice and effects of the military
solution in Colombia (2002-2010) and in Sri Lanka during the period of 2007-2009.

Review some of the literature around the all-out war theory as a means to ending civil
wars, like those in Sri Lanka and Colombia.

Analyse and explore some of the implications and critiques of the all-out war theory and
practice, and assess which model of development seems to be informing this approach.
1.6 Research questions and working hypotheses
To achieve the aforementioned objectives the investigation poses the following questions:
Main question:
How has the all-out war policy emerged and been operationalized in Colombia and Sri
Lanka, and with what results? How can this be interpreted in relation to domestic and international factors, and through on-going debates in the field of conflict and development?
Sub-questions:

Was the military solution to civil war in these two countries a conjunctural reaction to the
post 9/11 environment or was it mainly domestically influenced?

What are the theoretical groundings of the all-out war theory, and how can it be related
to some of the current theories on war and ending conflict? Is it a new theory, or is allout war theorymerely the extension of ‘mainstream’ theories of conflict?

Has the application of the theory managed to achieve ‘victory’in both cases, and how was
this accomplished?

Is the all-out war a development model? Or does it inform a particular approach to development in each case?
As a working hypothesis, military victory can appear to ensure a more sustainable peace8, but
the effects of ending the war in this way can increase structural violence, thus adding to the
grievances/needs of those defeated (see chapter 4).Last resort military alternativesare probably
preceded by several attempts at negotiation; in the case of Colombia and Sri Lanka, it appears
that the military solution emerged as the consequence of several failed peace processes.The sustainability, desirability and the reach of these strategies can be misleading, depending on the
manner in which the results are assessed and analysed. On the one hand, if warring parties learn
how to adapt to these new military methods, then their defeat or victory may be less likely. Additionally, the effects of these policies on development can be mixed; if they are analysed from an
economic perspective, they can show some positive short term results, but from a human rights
perspective, the outcomes appear altogether more grim. This analysis depends strongly on the
measurement strategy and time horizon.
8Galtung
peace.
(1969) would call it negative peace. From here onwards the document will assess it generically as
5
1.7 Structure of the document
In chapter two, the document presents the history of both conflicts, tracing their ‘origin’
to 1948 and comparing the manner in which both conflicts have emerged and evolved between
then and 2010 (for a detailed analysis, see appendix 1). Chapter three comprises of a theoretical
analysis of the all-out war theory, providing an understanding of its logic and emergence in the
cases of Colombia and Sri Lanka. Chapter four critically assesses the basis for all-out war as a
‘theory of practice’ and examines its logic, addressing the validity of some key theories (especially
the liberal peace approach) that inform the all-out war approach. This chapteralso proposes a
taxonomy which links approaches to ending civil war with approaches to development more
broadly. The last chapter returns to the central questions and main hypotheses and sums up the
key lessons of this research, presenting future research questions.
6
Chapter 2. Colombia and Sri Lanka: from colony to all-out
war
2.1 Introduction
To what extent is it valuable and useful to compare two different countries, especially
when they have different histories, languages, dynamics and contexts? This chapter will try to
compare Colombia and Sri Lanka, and discuss how their histories are interrelated/determined by
their internal conflicts.
Preliminarily, both conflicts appear to be more different than similar (see table 2.1), while
one is portrayed as an ethnic conflict(Devotta, 2005), the other can be considered to be classic
guerrilla warfare (Rangel, 2001). In Sri Lanka, the rebels were looking for a separate state
(Ropers, 2008; Shastri, 2009), in Colombia they were searching for control of the state apparatus
(IEPRI, 2006). TheSri Lankan conflict is often framed as a civil war, whileColombiais framed as
a ‘new war’9. The Sri Lankan conflict is identified as an ethnically based civil war10, but Colombia
is somewhat ‘poetically’ labelled under milder categories and referred to euphemistically as ‘the
situation’, ‘the violence’ or the ‘internal conflict’11. In both cases, history is determinant and
therefore essentialfor understanding the dynamics of violent conflict.
The media narratives/discourses of both conflicts has been affected by changes in the international environment. It is therefore interesting to observe in Colombia and Sri Lanka the effect of post 9/1112 attitudes within the national discourses around peace and conflict (Korf,
2006). In both cases, the war on terror discourse and the label of ‘terrorist’ became instrumental,
first as justifications for external support, and also as meansof levering negotiations through the
proscription of these groups. Aside from the different intentions and strategies surrounding the
use of the discourse of terrorism, in both cases this semantic change was related to a global homogenizing discourse about armed illegal groups (McGuinty, 2009; Metz, 2007; Tilly, 2004). Independently of the justification for these changes, or the validity of these labels, in both cases,
labelling the armed groups as terrorists has affected the nature of conflict, the manner in which
war has been conducted, as well as its outcome.
In exploring these dynamics, the chapter is structured as follows; the colonial history of
both countries, as well as the cessation of colonialism each country, is outlined. The chapter then
presents the emergence of conflict in each particular context, and continues on to discuss the
evolution of the conflicts before and after 2001, in light of 9/11.Finally, changes in the discourses surrounding the rebels are discussed.
2.2 Colonial times: seeds of destruction?
The demographics of Colombiachanged with the arrival of the Spaniards, whose introduction of weapons and diseases decimated the population (Posada, 2004). In spite ofthediversity of cultures and populations in Colombia,there was a ‘melting pot’ environment among ethnici-
I do not consider the category of new war to be sufficient, as it describes changes in the environment,
ignoring the effect of external dynamics on the internal strategies and tactics of some groups. I argue that
it is an old phenomenon in a new environment. For a discussion of these ideas, see Kalivas (2010).
10For a discussion on the nature of ethnic wars see Sambanis(2001) and Fearon (2003).
11 For an internal discussion on the Colombian case seePosada-Carbo (2001).
12This is used to refer to the 9th of September, 2001 attacks on the Twin Towers of New York.
9
7
ties and races. In spite of this, there was a strong class component, with an elitist social system,
which favoured the ‘criollos13’. Despite the fact that the Spaniards had ‘outsourced’ administration to these ‘criollos’, they still held the majority of the power, and the presence of a Spanish
military garrison was common(Bushnell, 1996).
Table 2.1
An introduction to the Colombian and Sri Lankan conflicts
DESCRIPTION OF THE
COUNTRIES
Area of the country
Size of the military (before the offensive)*
Size of the rebel forces (before the
offensive)**
Presence of paramilitary forces
Main political parties within the
country
Year in which the country achieved
independence
Was independence achieved
through a peaceful movement or an
armed insurrection?
Ethnic distribution of the country
Origin of the conflict
Political discourse/ ideologies of rebels
Sources of rebel funding
How soon after the end of colonialism did conflict emerge?
COLOMBIA
SRI LANKA
1,141,748 square km
65,610 square km
441.000 (262.000)
223.000 (213.000)
8.000 (11.000)
Unknown (9.000)
No formal presence; there are groups
called paramilitaries, but they are not independent of military units.
Liberal Party, Conservative Party
1819
Present, but most were former Tamil
groups that inclined towards the government in the internal struggles between
Tamil groups
SNL, UNP
1948
Armed insurrection
Peaceful transition, preceded by
strikes and riots
Mestizo 58%, white 20%, mulatto
14%, black 4%, mixed black-Amerindian
3%, Amerindian 1%
Grievances around land and class
inequality.
Sinhalese 73.8%, Sri Lankan Moors
7.2%, Indian Tamil 4.6%, Sri Lankan
Tamil 3.9%, other 0.5%, unspecified 10%
Redistribution of power and wealth
after independence.
Marxist Leninist discourse. Mainly
based on the class inequalities within the
country.
Extortion, kidnapping, drug trafficking
Ethno nationalistic discourse. Not a
strong discourse on power and inequality
beyond the categories of ethnic groups.
Extortion, diaspora remittances
Officially in 1964 with the origin of the
guerrillas, but affected by prior events.
Conflict began officially 35 years after
independence, but the violence that
fuelled it can be traced back to the years
following independence.
Source: *Figure 3.6 **Figure3.5
The drive toward independence demanded a reconfiguration of the distribution of power
rather than a revolution. An example of this is the pleading of offenses (memorial de agravios)
that criticized the exclusion of ‘white’ Americans, and defended the rights ofcriollos to govern
the country (Posada, 2004).
Since the initial stages of Colonization, rebel movements existed, but were not strong
enough to contest the power of Spaniards until 1810, when the power of Spain, in an international context, was declining. Independence was finally achieved in 1819 after the battle of Boyacá, were Spanish forces were defeated by the insurgent army.
Since the 14th and the 15th centuries, Sri Lanka had been an important port for commerce, having been colonized by different countries: the Portuguese (1505-1638), the Dutch
(1638-1796) and finally the British (1796-1948). Miscegenation in Sri Lanka was less common
than in the Spanish colony of Colombia, yet administrative control was held by the Tamil Nadu
13Criollo
was the descendent of a Spaniard born in Colombia. Also criollo referred to the sons of a Spaniard and a Native.
8
and the Tamil Ceylon.The former were mostly ‘introduced’ to the country to work in plantations,
and the latter were a minority present on the island long before it became a colony, but were given a preponderant role in the administration of the colony (Devotta, 2000).
As in Colombia, grievancesrelated to mistreatment and discrimination gave rise to several
struggles for independence, but the grip on the colony remained strong. Rebellions in 1817, riots
in 1915, and 1919’s claims for a greater autonomy, among other events, demonstrated the national desire for independence. While a rebel army was not formed in these times, and there was
not an armed struggle for independence, plans existed for rebel forces trained in neighbouring
countries such as India to arrive and liberate the country (Ibid).
The dynamics of the independence struggle changed in 1942 with the entry of the Japanese intothe SecondWorld War. The Japanese bombed Colombo and reducedSri Lankato booty
to be contested among the allies and the axis powers (Uyangoda, 2007).
After the end of the war, several mutinies showed the British that their grip on the country was slipping, as political parties now called for complete freedom, rather than a greater autonomy, and staggering strikes across the country occurred with more frequency. Consequently,
in 1948 the British granted independence to the then dominion of Ceylon (Shastri, 2009).
2.3 The struggle after independence: looking for a nation and finding civil
war14
Internal political settlements and the dynamics of internal social, economic and political
struggles within both countries determined the elements favouring the emergence of conflict.
Internal grievances, ethnical (or class) inequalities, and the lack of a strong state defined the elements that lead to the emergence of civil war(see table 2.2).
In Sri Lanka, conflict emerged relatively soon after independence, as the post decolonization years saw the country moving towards a natural change in structures and power distribution.
These changes created new grievances among the former elite of the country. The Sinhalese demanded a bigger share of power in the country, corresponding to the demographic composition
of the country.The former elite (Tamil) resented these changes, claiming to have been marginalized within a national project more centred on the Sinhalese ethos than an ethnic-pluralism
(Devotta, 2005; Höglund, 2011). As the internal politics undermined the possibility of building a
plural nation state, the initial social movements of the Tamils became increasingly radical, as did
those of the Sinhalese, thus fuelling the emergence of armed groups that embraced war (Frerks,
2005).
Consolidation of the state did not occur in the ‘early’ period ofColombian independence.
The civil war did not emerge immediately after independence, but was the product of multiple
transformations within the Colombian nation and its state. The conflict was also a consequence
of several previous civil wars, which occurred between 1834 and 1903 and had impeded the consolidation of a strong political system and a strong state presence. Hatreds seeded on previous
conflicts and partisanship by politicians weakened the state, This lack of state presence created a
political economy of violence, where violence became a means to acquire local power.
Table 2.2
Colombian and Sri Lankan conflicts through the lenses of different questions
14To
see a detailed chronology of the history of both countries, see appendix 1.
9
KEY QUESTIONS
Are there elements related to micro foundations
(Kalivas) in the conflict?
Is it a grievance based
conflict?
Did the economy play
an important role in the rebels functioning?
COLOMBIA
SRI LANKA
Yes.
Yes. It can be asserted that is a political
conflict, born from the lack of representation of a
particular class.
Definitely, in fact several authors argue that
thegrowth and strength of FARC is justified by
their strong economic structure.
Ethnically based?
No.The insurgents came from particular areas or social classes, mainly the poorer areas of
the country.
Nature of the strategies
and tactics used
Mainly guerrilla tactics and use of terrorist
attacks usually targeting military bases/police
posts. They have not used strategic suicidal attacks, but in the late 1990s used more classic
warfare strategies.
No, but in case of death or 'retirement' they
offer welfare programs.
Not primarily, but it appears that some of
their cadres are forcibly recruited
Strong hierarchical structure, with concentration on the secretariat. However, their units
have strong autonomy within their operations.
Yes. Two left wing, one 'Right' Wing
Do they pay their fighters?
Do they use forced recruitment?
Organization of their
leadership (rebel forces)
Are there other armed
groups within the conflict?
Are the rebels based in
a particular territory?
Can this be called a
proxy war in the Cold War
era?
Which countries have
supported the government?
Is(was) it a regional
conflict?
Were (are) there associations among the rebel
groups?
Which countries supported the rebels?
No.Guerrillas are highly mobile, but violence
can be traced historically to particular areas of
the country.
No.
U.S.A
Not clearly, but possibly present in the disorders in the 1980s, which were used to solve personal disputes.
Yes.As the power was redistributed, changes in
the country, lack of representation and the presence
of violence within the political system contributed to
the idea that war/revolution was the only way to
express the opinions and interests of the Tamil
population.
Yes.The rebels depended mostly on taxation
and remittances from the Tamil diaspora to fund
their war. At some stages they also relied on support from India.
Yes.It can be argued that ethnicity actually provides a unifying element for a broad group sufferring
grievances. Tamils can be divided in two or further
sub groups.
Mix of irregular tactics (ambushes), terrorist
tactics and conventional warfare tactics.
No. In case of death they offer a coconut tree.
Yes.
Strongly hierarchical structure, concentrated on
the leader.
Yes, but most groups defending the Tamil
cause were annihilated, absorbed or joined the
government paramilitary forces.
Yes, mostly the northeast of the country.
No.
India, Pakistan, China, U.S.A
No, but the conflict has had consequences
forneighbouring countries.
No, but neighbouring countries have interfered,favouring some of the warring factions
Yes.The CGSB was an attempt to coordinate efforts and carry out joint operations.
No. There has been co-optation of some
groups under the LTTE.
Cuba supported the rebels in the 1960s and
1970s, but ideological differences made the
guerrillas independent.
In the 1970s and 1980s, India provided training
and weapons for the rebels.
The violence in Colombia reached its climax in ‘la violencia’, an episode in which almost
2% of the population of the country died (Palacios, 1995). The eventual result of this bloodbath
was a population that did not trust state institutions, for whom self-defence organizations and
the privatization of security became common practice. The whiplash effect which followed gave
10
rise to multiple guerrilla groups (Maoist, Marxist, charismatic) and to right wing ‘paramilitary15’
groups (Bushnell, 1996).
The nature of the Colombian conflictis complex, due to the existence of several groups
with different ideologiesand the influence of drug trafficking. Therefore it can be seen as a civil
war in which the limits between grievances (IEPRI, 2006), opportunism (Posada-Carbó, 2001),
politics (Richani, 2007), warlordrism (Duncan, 2005), ideology and micro foundations (Andrade,
2010) become blurred, making it a difficult phenomenon to comprehend.
In Sri Lanka, the conflict can be described as a more ‘classic’ conflict, having its originsin
an ‘ethnic’ post-colonial civil war. The conflict had different actors (armed groups claiming to
represent the Tamil interests) as well as some pro-Sinhalese armed groups (Devotta, 2005;Korf,
2006), but the LTTE managed to gain a violent monopoly on the armed agenda of the defence
of the Tamil cause.
The LTTE was an armed group that acted, within their controlled areas, as a regular army, but when outside of them used guerrilla and irregular warfare tactics. The LTTE appears
more radical than the Colombian guerrillas, as they used violence to kill any political opponents
and also forcibly recruited children as common practice (UNICEF, 2011). Although coca and
poppy cropshares are not common in Sri Lanka, the conflict is also to some extent associated
with mafias and drug trafficking (Jayasundra, 2011). The Sri Lankan conflict was of shorter duration than inColombia, but saw a similar tension between war and peace, where the final support
for the all-out war was born out of the weariness of the population and the LTTE’s lack of
commitment to the peace process (Devotta, 2011).
2.4 War as a phoenix phenomenon
In both Colombia and Sri Lanka, war emerged as a by-product of a failed peace process,
where the commitment of the actors, the difficulty of building trust and enforcing agreements
drove public opinion and politicians towards the idea that a military option could provide a useful solution to the problems of the peace process. It is interesting to observe that war was considered to be a solution, rather than an attempt at improving the flaws in the peace process, presented in section 3.2. The outcome was a revolution in the management of peace processes,
where a victor’s peace is comparable to a peace process.
2.4.1 In Colombia: Deception, reincarnation and rebirth
By 1998, public opinion in Colombia favoured a peace process. The peace process started
in 1998 and ended in 2002with the breakdown of negotiations.From its inception, it was evident
that the peace process had its own problems and difficulties. As negotiations were held without a
ceasefire or a clear negotiation agenda, military actions from each side became instrumental in
the negotiation itself.As the logic of the peace process became one of war, the credibility of the
process itself was undermined (Tokatlian, 2004; Pecaut, 2008).The negotiation can be described
as a ‘broken phone’, as the actors at the table approached talks with different time frames, logics,
and discourses in mind. This finally appears to have eroded the possibility of clear communication among them.An example of such breakdown in communication is the issue of the DMZ;
military forces wereopposed to the establishment of a the DMZ in the middle of the country (see
image 2.1),as it would provide a training and preparation area for guerrilla fighters, but the guerrillas considered the DMZ vital to the commencement of negotiations, as the area provided a
safe place to negotiate without any military interference.
Paramilitaries can be seen as an extension of the state apparatus, and therefore similar to military forces.
In the Colombian case, they are independent organizations.
15
11
Figure2.1
Map of Colombia showing the DMZ
Source:roebuckclasses.com
Retrieved on august 10nd 2011 from
<http://www.roebuckclasses.com/maps/topicmap/Colombiadrugwar/Colombia_political_DMZ.jpg>
This lack of trust and ‘commitment’ was latent during the whole peace process, as military
forces accused the FARC of bringing kidnapped persons to the DMZ, and using the area to prepare attacks. The FARC continuously complained about the military forces’ lack of resultsagainst
paramilitarism (Tokatlian, 2004: 640).
The peace process never managed to progress beyond the negotiation of the agenda.Not
only were the definitions and terms of negotiation troublesome, but theyevidenced both parties’
lack of experience in or commitment to the management of a peace process. These difficultiescan be seen in the failure even to establish a verification commission, a simple element
such as the naming of a friendly third party to solve disputes(Ibid: 639).
One of the elements that drove the Colombian public opinion towards peace was their
weariness of the consequences of war; kidnappings, attacks on small towns, death of civilians
and members of the armed forces.The public expected that the peace process would decrease
these grievances, but when it failed to do so, support for the peace process began to dwindle
(IEPRI, 2006).
The involvement of the international community in the peace process was limited. The
USA’s so called ‘Plan Colombia’, aimed to strengthen the military capacities of the government
against drug production and drug trafficking, under the argument that the conflict was driven by
a lack of government strength and revenues from illicit crops and drug trafficking (Pecaut,
2008:3). The European Union accompanied the process, but did not have a strong role as mediator, and their economic support for the peace process was limited.
The limited involvement of the international community in the failed Colombian peace
process presents a stark contrast with the Sri Lankan peace process.The Colombian government
made efforts to involve the international community, even, ‘touring’ Europe with some guerrilla
leaders to gather support for the peace process (IEPRI, 2006). In spite of these efforts, the peace
process was mostly ignored by the international community, and the international commitment
which might have bonded the process and the actors never became a reality.
12
The end of the peace process was supported by Colombian public opinion; several attacks
by guerrillas and the lack of results from the process had destroyed the remaining public support
for peace.
2.4.2 In Sri Lanka: Fight to the bitter end
Towards the end of 2001, the LTTE expressed their intention to explore options for a
peaceful settlement to the conflict, an initiative that was promoted by Norway and CSO’s, and
was the result of a process that started in 2000 (Höglund, 2010).
After the elections, the UNP came to power, supported by a pro-peace agenda as in Colombia. By the beginning of 2002, the LTTE and the government had signed a Memorandum of
Understanding (MoU), and agreed on a ceasefire (Jayasundara, 2011: 16). The parties also appointed a committee, headed by Norway,that would monitor the ceasefire (Höglund, 2010). This
body, the SLMM, created a promising environment where the costs of spoiling from the process
were ‘high’, and commitment technologies appeared to be strong.
Figure2.2
Map of Sri Lanka showing the area claimed as Tamil homeland
Source:kasamaproject.org
Retrieved on august 18nd2011 from <http://kasamaproject.org/2009/05/23/sri-lankan-communist-on-the-tamil-conflicts/>
The future seemed to be promising, expectations were high and substantial financial
support was offered by the international community(Höglund, 2010). These expectations were
reinforced by the unilateral actions of the actors, as the LTTE initially agreed for the first time
on the possibility of a federal solution to the conflict(Korf, 2006: 286). See figure 2.2 for the areas of influence of the LTTE.
As in Colombia, endogenous factors worked against the initiative for peace, and the
peace process was held captive by only one political party. It is therefore understandable that
part of the political establishment opposed the peace process (Shastri, 2009). While good in theory, in practice the process was flawed, and although a ceasefire was signed, hostilities continued
on both sides, albeit on a smaller scale (Höglund, 2010).
As there was not a broad consensus in support of settlement,possible peace gestures,
such as the LTTE’s proposalto form an interim government in the north were misinterpreted as
aggressions. These tensions influenced the political landscape, favouring the arrival of Rajapakse,
who representedthe political change that was occurring within Sri Lanka.Some sectors of society
were frustrated by the peace process and parties such as JPV and the JHU (nationalist Sinhalese
parties) gained power and influence (Jayasundara, 2011). Changes also came within the LTTE, as
there was an internal fracture, unrelated to the peace process, in the east, which lead to the defection of 5000 soldiers (Godhand, 2009: 348).
13
By 2005, the presidency of the country had been won by a narrow margin by MahindaRajapaksa. The victory actually appeared to have been determined by the LTTE boycott on
elections in Tamil areas (Jayasundra, 2011; Devotta, 2011). This strategy had been usedpreviously
in elections as a means to recompose political orders within Sri Lanka (Shastri, 2009).
After the tsunami struck the country in 200416 the peace process drove both parties to
agree on a Post-Tsunami Operational Management Structure (P-TOMS), that emerged after
harsh negotiations on how to redistribute resources (Shastri,2009:92; Goodhand, 2009:349). The
agreement was finally scrapped, as opposing parties declared it to be secessionist and unconstitutional (Ibid:91). Additionally, a series of military attacks eroded support for the process, and diminished the international credibility of the LTTE as a peace-willing organization (Shastri, 2009).
These military attacks occurred during a period of ceasefire, not during wartime negotiations as in Colombian. By 2006, the SLMM reported more than 3400 violations of the ceasefire
on the part of the LTTE and around 270 from the government (Shastri, 2009: 90). This can be
considered as evidence of the LTTE’s lack of commitment to the peace process, or as evidence
of the difficulty of enforcing ceasefires. To the government of Sri Lanka, these violations clearly
evidenced the former.
By late 2006, the LTTE officially and indefinitely pulled out of peace talks (but not out
of the peace process), and accompanied this decision with a series of attacks. This triggered a
strong military reaction from the government. The condemnation of theSLMM went unheeded
by actors, and the path to war was paved once again.
2.5 Post 9/11: Justification, endogeinity, and discourse
By the end of 2001, the international atmospherehad changed completely, withirregular
armed groups across the globe being demonized, regardless of their tactics. The effect of 9/11
was not only political, but also influenced discourses in such a way that the word ‘terrorism’ became a loaded and hated word (Tilly, 2004). On October 5th2001, both FARC and LTTE were
included in the US updated terrorist organizations list(U.S. State Department, 2001).
It would be naive to believe that the effect of the 9/11 dynamics alone determined the results of the Sri Lankan and Colombian conflicts and peace processes. Both conflicts had initial
attempts at peace processesprior to 2001, in which economy, politics, and internal conditions
affected the stakes surrounding war and peace. Therefore it can be saidthat 9/11 brought about
changes in the international attitudes to conflict, rather than completely changing the nature of
the conflict itself (see table 2.3). The biggest change in international attitude was the creation of
“the war on terror” discourse post 9/11 (Kaldor, 2005). The same wars were being fought with
almost the same actors, butthe labels had changed. Within this new discourse, war was presented
as a virtuous campaign , where the violence of war could be justified as a moral obligation
against terrorism(Dexter, 2008: 68). Colombia and Sri Lanka were no exception.
2.5.1 Colombia: from guerrilla to narco-terrorist
In Colombia, the media played a strong rolein supporting the Uribe government. A perfect
example of this is the change in the discourse characterizing guerrilla groups. Until 2002, they
16Also
known as theIndian Ocean earthquake of 2004.
14
were referred to as guerrillas and insurgents,after the failure of the peace process and the arrival
of the Uribe administrationthey were considered terrorists or drug dealers. Looking for external
support in their fight against the guerrillas, the government started to depict members of armed
groups as terrorists, assuming a militaristic perspective on the armed conflict.
This depiction ignores elements such as the favourableopinion the FARC enjoyed among
3 to 5% of the population (Pecaut, 2008). Otheringa group as merely terrorists or drug dealersnot only simplifies its nature, but is incorrect, as the group’s nature transcends its tactics. It
would also be improper to depict the FARC as a classic Marxistgroup, as the limits between
wardlorism, mafia and revolution are indistinct(Sweig, 2002, 123).
The government created a discourse whereby the members of armed groups where referred as terrorists. Simultaneously, a nationalist discourse emerged, and the glorification of
armed forces was common.Advertisementsappeared whichtouted the victories of war and military operations, and the possibility of winning the war was latent in media coverage. This nationalist, almostchauvinist environment made critique of and opposition towards government policies difficult (Tokatlian, 2004, 652).
The Uribe administration finally managed to create a securitization of the discourse(Posada-Carbo, 2011), and created a propaganda campaign that exaggerated results against
the guerrillas (see chapter three for further discussion). It may be said that under the Uribe administration, the ‘real’ war was fought in the media.
2.5.2 Sri Lanka: from Tamil tiger to Tamil terrorist
In Sri Lanka, as in Colombia, the title of ‘terrorist’ was used with more frequency after the
9/11 attack. However, the term had a longer, more loaded linguistic history because of the tactics used by the LTTE in their attacks. The E.U.’s decision to include the LTTE in 2006’s list of
terrorist organizationscontributed to the change in discourse (Jayasundra, 2011).
Similarly to Colombia, the military campaign was accompanied by a media campaign
which glorified the army soldiers, and their‘achievements’in battle.Confidence in the military was
as high as 92% (Ibid: 31), and military battles were equated to the struggle for independence.
This process was accompanied by the accumulation of power by some political parties, and the
establishment of an autocratic political regime (Shastri, 2009).
In this context, the media took on an interesting role in legitimizing military actions and
government decisions. In what could be defined as the mediatization of victory, the media ignored
changes in the political establishment and accusations of human rights violations. In this environment, moderate voices ended up in the crossfire, with the government accusing them of supporting the ‘Tamil Terrorists’ and the LTTE labelling them as traitors. This created a context in
which only armed voices where heard.
Finally, in spite of the historic discourse defending the rights of the Tamil people,, it wasa
mostly Sinhalese discourse that emerged parallel to the rise of chauvinist parties such as the JPV
and the JHU. Post-bellum Sri Lanka can be described as a country in which language and ethnicity were politicized (Arambewela, 2010, 371) and autocracy was consolidated.
2.6 Conclusion: new wars and the offspring of 9/11
It is important to highlight those differences in context and particularity that make the
conflicts in Sri Lanka and Colombia unique despite the fact that their governments looked for
similar variables to achieve peace (chapter three). These policies emerged as an internal consequence of historical processes, where the failure of previous peace processes supported the military option.
15
In both countries, conflict emerged because the state was unable to settle disputes among
ethnic groups or classes.This was the consequence of a disconnect between the political system
and the needs of society.
The fighters in both Sri Lanka and Colombia can be, and have been, characterized simultaneously within several different categories; as legitimate rebels, terrorists and opportunists. An
historical comprehension of the nature and history of conflict is thus imperative to comprehend
the evolution of war in each context. In both conflicts, decisions made regarding war and peace
after 9/11 were driven by internal processes and failed peace processes, rather than by external
elements.
Finally, what is seen in the Colombian and Sri Lankan conflicts is the result of different
discourses and practices, where the category of ‘new wars’, merged with post 9/11 securitization
discourse, gave birth to a new breed of conflict practice, a ‘new peace’ that can be defined as nationalist (Höglund, 2010), militarist (De Mel, 2009) and chauvinist (Devotta, 2009; Lewis, 2010).
The case of Sri Lanka can be viewed as a consolidated discourse model, while Colombia can be
considered to be a work in progress, moving towards this ‘new peace’.
The ‘new peace’ concept was influenced by the new wars discourse that demonized conflicts as savage and non-ideological (Kaldor, 1999; Kalivas, 2001). This, in conjunction with the
war on terror, reduced the possibility of willing mediators and interested actors that could promote peace settlements (Höglund, 2011), thus eliminating the restrictions on trying to solve war
through war, and actually creating discursive justifications for it.
16
Chapter 3. A theory of practice? The All-out war solution.
3.1 Introduction
An understanding of the all-out war approach requires a ‘theory’that combines elements
of conflict management, conflict resolution, international relations, conflict studies and sociology. In the literature reviewed,Lutwakk (1999) is the only author to blatantly defend the idea that
war is good. For example, whenresearching peace settlements on academic journal search engines17, 85% of results referred to peace settlements achieved through negotiation, while peace
settlements achieved through military defeat comprised only 15% of the results. The area appears to be under-researched, as it is politically incorrect to assert that war might be good; despite the fact that in practice war, or military means may be preferred.
This chapter tries to present all-out war within the context of the conflict studies area.
First, this chapter suggests that within the choice to adopt an all-out war approach, there often(including in the cases under consideration) lies the possibility of addressing the failures of
prior peace processes in civil wars. The chapter then identifies the key elements that underpin allout war theory, presenting this approach more fully so as to interpret the ‘all-out war’ as a theory
of practice. Finally, the chapter presents the all-out war theory as it was operationalized in Colombia and Sri Lanka. In terms of underlying ontology, this chapter deals with agency-based approaches to conflicts18.
3.2 All-out war: a solution for failed peace processes?
Peace settlements can be defined as processes that have a clear start but an uncertain outcome. In 50% of civil wars, parties engage in negotiations, but less than 20% achieve settlements
(Wood, 2003:2). The existence of a peace treaty does not guarantee its durability, as 37% of
peace agreements fail within 5 years (Hartzel et al, 2001: 195). This can be the consequence of a
number of reasons, where small failures within the negotiations, or slight changes in the post settlement environment can entail dramatic results (for examples, see Walter, 1999;Toft, 2010;
Murshed, 2008a). This section presents some elements that are thought to contribute to the failure of peace processes (according to an agent based approach):
1. Misperceptions about the costs and benefits of reneging on an agreement.
2. The ‘discount rate’ at which settlements are valued by key actors.
3. The inability of parties involved in the negotiations to commit.
4. The presence of indivisibility issues within the claims of the warring parties.
5. The credibility of the agreements within the wider peace processes.
6. The difficulty of enforcing commitments.
3.2.1 Benefits and costs of reneging
If those involved in anagreement believe (perceive) that there is a good chance of gaining a
profit/benefit from reneging on the treaty, such as the renegotiation of better terms, they will
reignite war (Murshed, 2008b: 371).This can account for a number of the failures and paradox17A
search on the scholar.google.com search engine was conducted, using as keywords “‘peace settlement’and ‘negotiation’” and “‘peace settlement’ and ‘military defeat’”.
18To see a discussion of possible ontologies related to the study of civil war and war origins, see Jacoby
(2008)
17
esexisting in peace processes. When renegotiated agreements offer higher incentives for peace,
an unexpected incentive is created to renege in search for greater benefits(Fearon,
2004).Additionally, the threat of reneging serves strategic and tactical purposes (within and outside of the negotiation); fighting rather than settling at the negotiation table is a means for the
combatants to show their strength.
3.2.2 The discount rate
In war torn areas the discount rate19 is high (Murshed, 2008b: 372), and decisions are
therefore more risk prone (Kanheman, 2003).This, combined with the common overestimation
of the chances of a potential military victory (Murshed, 2008b: 371) reduce the possible benefits
of a long lasting peace. As theserisk-prone behaviours are related tothe evaluation of the benefits
of reneging, measures to reduce the perceived risks will accordingly reduce the incentive to go
back to war; especially in those situations of poverty and high uncertainty that are common to
post-conflict countries (Ibid; Wood, 2003).
3.2.3 The inability to commit
The inability to commit is the main challenge for peace deals(Mattes, 2009: 739). This is
because in many post conflict environments, there is an absence of institutions that would guaranteecommitment to the agreements signed (Murshed, 2008b: 371).
The presence of third party actors could discourage reneging (Collier, 2008; 464) andenforce the commitment technologies necessary to restoring peace (Hoeffler, 2009: 9;Murshed,
2008b). When a third party enforces a settlement, the probability of failure decreases by
98%20(Pearson et al, 2006: 115; Walter, 1997). This set upcan create artificial incentives for settling differences (Beardsley, 2008), or a ‘big brother’ dependent peace (Collier, 2008).
3.2.4 Indivisibility issues
What if what is at stake is impossible to negotiate? Indivisibility refers to negotiation ‘anchors’ that impede the bargaining itselfbecause limits and thresholds are set which cannot be discussed at the table. The feasibility of agreements to share power and resources are difficult to
achieve under these circumstances, and leads to less stable peacesettlements (Toft, 2010: 17;
Wood, 2003: 3; Walter, 1997).Indivisibility usually emerges when the future is heavily discounted
(Murshed, 2006: 2).Some authors go as far to suggest that if stakes are completely indivisible, the
only option for real peace is a military victory for one of the parties involved in the conflict
(Wood, 2003: 14).
3.2.5 Credibility problems and trust
Democratization and power sharing measures can be necessary, but not always sufficient
to consolidate peace and credibility (Mattes, 2009). Therefore,guarantees improve the chances of
a lasting peace and the credibility of the agreements (Collier, 2008: 464; DeRouen, 2010). Additionally, the element of trust is vital, not only in the aftermath of the war, but during the peace
19
The discount rate refers to the rate at which money is valued in the long term.
Collier (2008) says that the risk is reduced from 40% to 31% and Mattes says that the likelihood of a renewed conflict is reduced by 69% (Mattes; 2009: 752).
20
18
process itself and the credibility of the peace treaty (Addison, 2002). There is a strong incentive
to distrust the other party, as the relationship between both actors has been mediated by violence, treachery, treason and war. It is thereforedifficult to expectthese embodied beliefs to
change within a single peace process.
3.2.6 Difficulty of enforcing commitments
Sanctions, arms control, trade restrictions, foreign aid (Page, 2003), power sharing mechanisms (Hartzel, 2001) and strong DDRprograms,etc.can clearly contribute to the peace building
process. The problem is how to go about enforcing these commitments (Fortna, 2003; Hoeffler,
2009:9) and to discourage the possibilities of reneging (Collier, 2008: 464). This is particularly
important inan environment in whichthere are no strong institutions upon which to anchor the
promises of a peace deal (Murshed, 2008a: 371).
As a solution, the presence of commitment technologies that require institution building
present a paradox (Addison, 2002); as strong institutions require a peaceful environment, but this
is hard to get specially in a country with a civil war. As a solution for this problem the presence
of external interventions21 can serve the purpose of ending war, as they provide credible commitment technologies(Regan, 2002; Regan, 1996).
3.3 All-out war –a conceptual primer?
Within the conflict studies area, the negotiated settlement approach is far more pervasive,
appealing and influential than what this study calls the ‘all-out war theory’, or what Luttwak
terms ‘give war a chance’ (Luttwak, 1999; Toft, 2010: 7). The difficulty is that uncomfortable
evidence exists that appears to show that decisive victory, military victory, victor’s peace, and
military imposed political settlements tend to last longer than negotiated peace (Ibid).
War has had an historical importance in the constitution of the modern state, having to
some extent defined the nature of nations.Certain authors, such as Schumpeter (Nakamura,
2000; Giersch,1984) and Sanin (2009) present some of the possible positive externalities of war
(see table 3.1)within this process.
It is also interesting to note that after the end of the Cold War, peace settlements became
more frequent (Mattes, 2009: 739), despite the fact that military victories produce more stable
forms of peace than military stalemates and negotiated settlements(Page, 2003; Quinn, 2007). In
presenting the all-out war theory, this section will first address some of the empirical evidence
that supports it, and will then relate theall-out war theory with a number of theories linking war
tostate-making.
Table 3.1
Possible positive externalities of all-out War
Externality
Example
Technical change (creative destruction)
Military industry, technological changes
Achievement of egalitarian political objectives
South Africa, de-colonization
Political inclusion
Female enfranchisement
Interventions can be diverse; also their effects on the settlement of a conflict can vary (Regan, 2002).
To see a discussion of the different kinds of external interventions as tool for civil war ending see
(Murhsed, 2010).
21
19
Change in the political structure
Strengthening of the state
Change in leadership
Strengthened armies and police, more taxation
Keynesian effects of military build-up
State investment in public works through reconstruction programs
Source: Adapted from Sanin (2009: 7)
3.3.1 The empirical evidence
According to Page (2003: 111), 52% of peace agreements went back to war within a period
of 5 years; more favourable estimates put the risk of recurrence between 46% (Collier, 2008;
465) and 37%, with peace having an average duration of42 months (Hartzel, 2001: 195). One
possible explanation for the scarcity of successful peace settlements is that the benefits of winning a war can be so high (Walter, 1997: 342).
In contrast, wars that end in military victory have a recurrence rate of between 6% and
17% (Toft, 2010: 20). This may be a consequence of decisive military victories allowing for the
institutionalization of large powerasymmetries, thus reducing the chance of re-emergent conflicts
(Walter, 2004: 374).
Based on this evidence, it appears plausible to support the idea that war can save lives in
the long term (Toft, 2010: 20). Even pacifists must acknowledge that the benefits of lasting
peace in terms of saved lives could be substantial enough to justify war as a means to achieving
peace(Dower, 2009).
3.3.2 Back to basics: ‘old theories’ revisited
It is interesting to note that many sociologists view the nation building process either as a
violent process (Tilly, 1991), or as a process of checking violence (Cohen, 1981: 904).This argument can be extrapolated to the notion that war makes states or that war is inherent to the process of state making (Sørensen, 2001: 1).
The underlying argument is that if countries suffer internal violence, it is becausethey are
not a consolidated state; they have a weak government and problems raising resources to fund
their state apparatus (Beardsley, 2009;Bates, 2008; Cohen, 1981). War presents a response to
these problems (Sørensen, 2001).Providing territorialsecurityguarantees the presence of other
state institutions (Sweig, 2002);no insurgent group can contest government authority if it faces a
strong, well-financed, organizationally secure military structure (Fearon, 2003: 80). Additionally,
the business of war necessitates two strong institutions that are the base of modern state making:
a strongbureaucracy for taxing and a strong military presence across the territory (Mann, 2002).
Other authors highlight that internal wars (as opposed to external wars) are bad for state
capacity (understood as the existence of property rights and the capacity to levy taxes), as the
conflict erodes the existing public institutions, and therefore it constitutes in an obstacle to development and growth (Besley, 2010).
Empirical evidence shows that state weakness is a better indicator of the proclivity for
civil war than grievances (Fearon, 2003: 88). Colombia and Sri Lanka both conform to the theory
that war is the consequence of a weak state (Richani, 2007: 405; Frerks, 2005: 12).
3.4 Achievement of ‘peace’through war, Sri Lanka and Colombia
It appears that Colombia is currently in the process of building up a stronger military,
and is thus a clear example of the all-out war in theory. Both Colombia and Sri Lanka have increased their military expenditure in the last years, although Sri Lanka has not increased expenditure to the same extent as Colombia has (see figure 3.2). In terms of the proportion of the
GDP being invested in military spending, both countries appear to show a slight increase pat20
tern. However,when comparing expenditure to previous years, it becomes clear that Sri Lanka is
actually reducing the percentage of the GDP being spent on the (see figure 3.1).
The manner of increase in the military forces is quite contrasting, as Sri Lanka appears to
have the same number of soldiers as in the mid-1990s (see figure 3. 4), while Colombia has
quadrupled the size of its military forces.
However, the validity of this analysis can be misleading, as the different databases used
can present contradictory arguments. For example, the SIPRI databases do not agree with the
Sipri-Prio or UPPSALA databases on the number of battle deaths. Contradictions exist even
within the same database;on the SIPRI dataset, the total expenditure in the military appears to be
inconsistent with the size of the military forces. Therefore, datasets are not fully reliable, as they
can be instrumentalized for any argument, problematizing the nature of our analysis, and compromising the validity of our conclusions. In spite of these considerations, the following sections
will attempt to present a number of observations on the cases of Colombia and Sri Lanka.
Figure2.2
Military expenditure of Colombia and Sri Lanka
10000
8000
Source: Stockholm International Peace Research Institute, 2011.
SIPRI database. Retrieved from
<http://www.sipri.org/databases/armstransfers>
Figure3.3
Size of the insurgent forces
2010
2008
2006
2004
2002
2000
1988
2008
2006
2004
0
2002
1000
1
2000
2000
1.5
1998
3000
2
1996
4000
2.5
1994
5000
3
1992
6000
1990
4
3.5
1988
COLOMBIA MILITARY
EXPENDITURE IN US$ m
7000
1998
COLOMBIA
4.5
1996
5
SRI LANKA MILITARY
EXPENDITURE IN US$ m
9000
1990
SRI LANKA
1994
6
5.5
1992
Figure3.1
Military spending as proportion of the GDP
Source: Stockholm International Peace Research Institute, 2011. SIPRI database. Retrieved from
<http://www.sipri.org/databases/armstransfers>
Figure3.4
Size of military forces in Colombia and Sri Lanka
13000
12000
11000
10000
FARC
9000
8000
LTTE
7000
SIZE OF
MILITARY
FORCES
COLOMBIA
SIZE OF
MILITARY
FORCES SRI
LANKA
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
6000
500000
450000
400000
350000
300000
250000
200000
150000
100000
50000
0
Source: Worldbank 2011, data retrieved from
Source:Country reports on Terrorism.U.S. State Department.
<http://data.worldbank.org/country/>
As an estimate for the number of insurgents, I took the averageof the report of the estimates.
21
3.4.1 In Colombia: Continuity despite military build–up?
From the beginning of the presidential campaign, the candidate Uribe claimed that the
guerrillas had to be obliged, by force, to negotiate seriously (Tokatlian, 2004). Once established
in the presidency, Uribe created the link between the ‘war on terror’ and the Colombian conflict.
The government thus tried to use the new international framework in order to gain access to
more funding and to extend the reach of ‘Plan Colombia’ resources to fight the ‘terrorists’.
Uribe’s approach to solving the conflict differed greatly from the unpopular pace process
of the previous government. At that time, the government became involved in another peace
process, not with left wing forces, but with the paramilitaries (Diaz, 2009). This peace process
had a DMZ and an unclear ceasefire, and managed to garner the support of the OAS.
Security indicators improved, but the new ‘skills’ of the Colombian armed forces were,in
part, the outcome of the restructuration process of the military by the Pastrana government.
This can be seen in the growth in the size of military forces (see image 3.4 and 3.3), and in investment in strategic assets (airplanes, helicopters, communications systems, etc.) (see image 3.2).
Figure3.5
GDP growth in Colombia and Sri Lanka
10
8
6
GDP GROWTH
COLOMBIA (%)
4
2
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
-4
1999
-2
1998
0
GDP GROWTH
SRI LANKA (%)
-6
Source: Worldbank 2011, data retrieved from <http://data.worldbank.org/country/>
That is why, to some extent, the results of this government seem to be quite contradictory.During the government’s first four year term, the security situation and the economy showed
improvements (see image 3.5). During the period of 2006-2010, however, the government had to
face the political and economic consequences of their policy decisions.The failures of the peace
process with the paramilitaries became evident (Pardo, 2007), the number of guerrilla attacks had
not appeared to change dramatically and the alliances between paramilitaries and politicians had
started to become more visible22 (Lopez, 2010). These failings became evident in a context in
which the international community had started to criticize the human rights violations of the
government (HRW 2010).New ‘paramilitary’ groups, labelled ‘emergent bands’, developed,
anddespite being weakened by successful military attacks targeting several of their leaders, the
guerrillas kept fighting and refusing to negotiate,
At this time, the authoritarian government started to compromise on the separation of
powers between the legislative, the executive and the judiciary, with mixed results. According to
the government and the media, matters had improved (Tokatlian, 2004; Mejia, 2008), but a series
of paradoxical outcomes appeared; for example, the number of IDP’s rose dramatically, and Colombia became the country with the highest displaced population in the world (see image 4.7).
22This
process in Colombia is called Para-politics, and is the prosecution of several politicians at regional
and national levels for their coexistence with paramilitary forces.
22
Other outcomes, such as the number of deaths in battle, are contested (see figure 3.6), but human rights violations on the part of the military forces and the irregular forces seem not to have
improved significantly (Amnesty International, 2011). Interestingly, indicators such as the as the
political freedom index and the political terror scale showed no significant worsening, and in
some cases even showed slight improvements (see chapter 4).
In recent developments, the FARC has declared the group’s interest in embracing a peace
process with the new government, a move which the government considers to be result of their
efforts to force a defeated enemy into negotiation (Semana, 2011), as the government has managed to kill the leader of the FARC.
Source: UCDP Conflict Encyclopedia Uppsala University data retrieved from
<http://www.ucdp.uu.se/gpdatabase/search.php>
3.4.2 In Sri Lanka: Blitzkrieg and scorched earth
By the end of 2006, senior members of the government expressed their intentions to use
the full military might of the armed forces in order toexpel the LTTE from several areas of the
country, and then defeat them in the north of the country. Initial operations were slow but over
time, military successes began to occur. The LTTE was an irregular group that based their actions in particular areas, having low mobility and insufficient strength to stop a full scale army
assault in a terrain of predominantly flat jungle without any geographical opportunities for retreat
(the area was surrounded by sea). Additionally, the absence of the 5.000 soldiers of the Karuna
group impeded the LTTE commanders’ flexibility in diverting the attack.By 2007, the LTTE
presence was diminished, having a total force of only 5.000 soldiers (Shastri, 2009; Goodhand,
2009).
In spite of this military escalation, the ceasefire agreement was technically in place while
the government launched its military attack. By the end of 2007,sometime after the military escalation had been initiated, the government officially abandoned the agreement. It is interesting to
note that the number of soldiers in Sri Lanka did not increase, but actually decreased in comparison to 2000’s figures (see figure 3.4).
23
2010
2008
2006
2004
1990
Source: PRIO dataset 2011, data retrieved from
<http://www.prio.no/CSCW/Datasets/Armed-Conflict/BattleDeaths/The-Battle-Deaths-Dataset-version-30/>
2002
0
2000
0
1998
200
1996
500
1994
400
2010
1000
2008
600
2006
1500
2004
800
2002
2000
2000
1000
1998
2500
1996
1200
1994
3000
1992
1400
1990
3500
1992
Figure3.6
Battle deaths related to the Colombian conflict (according to two different sources)
In that time, the number of casualties increased dramatically as the combat continued
(see figure 3.7).By 2008, the LTTE announced that it would declare a unilateral ceasefire, but
unlike previous such agreements, the proposal was dismissed by the government as a trick and a
treacherous tactic (Höglund, 2010).
As the government kept the pace, gaining terrain on theground, the magnitude of the
humanitarian crisis facing the country became more evident. With more than 200,000 civilians
having been displaced in the latest round of fighting, the situation was turning into a humanitarian disaster (Human Rights Watch, 2009; Amnesty International, 2009).Questions were also
raised concerning whether the number of battle deaths included civilians. Interestingly, neither
India nor other members of the international community interfered, in spite of allegations of
abuses leveled by human rights advocates.While some pressure was exerted on the government,there was no talk of military intervention, as there had been in other times. At this stage of
the conflict, the voices of the legitimate representatives of the Tamil community abroad and inside the country went unheard.
Figure3.7
Battle deaths related to the conflict in Sri Lanka
12000
Source: PRIO dataset 2011, data retrieved from
<http://www.prio.no/CSCW/Datasets/Armed-Conflict/BattleDeaths/The-Battle-Deaths-Dataset-version-30/>
2010
2008
2006
2004
1990
2010
2008
2006
2003
2000
1998
1996
1994
1992
1990
0
2002
2000
2000
4000
1998
6000
1996
8000
1994
10000
1992
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
Source: UCDP Conflict Encyclopedia Uppsala University data retrieved from
<http://www.ucdp.uu.se/gpdatabase/search.php>
By April, the area under the control of the LTTE had been reduced to 10 km2. Meanwhile, the Tamil displaced population continued to move out of war affected areas. When the
international community responded to the situation, that response was delayed and contradictory. By the time a group of UN experts had called for a preliminary investigation to address the
situation in Sri Lanka, as many as 20,000 civilians may have been killed in the no-fire zoneamid
fighting between the army and Tamil rebels (ICG, 2010). In 2009, the UNHR actually congratulated the government; the ICRC and UNHCR closed several of their offices after the ‘end of the
war’, thus legitimizing the Sri Lankan government’s actions and supporting the impression that
the conflict was over.
The LTTE finally admitted defeat on May 17, 2009, after the loss of all of the main territories under their control, and the death of some of their top leaders. With the defeat of the
LTTE, the political landscape in Sri Lanka could be reframed, and a victorious discourse, centered on the Sinhalese ethos, emerged. Post war policies focused on the physical reconstruction
of war torn areas, where Rajapaksa became a national hero in spite of allegations of authoritarianism (Ibid), and violations of human rights during the offensive.
24
3.5 Conclusion
It has been established, on the basis of the data and literature, that the ‘peace’ which follows civil wars ending in military victory tends to last longer. In the cases of both Colombia and
Sri Lanka, the logic that prevailed was the necessity of achieving the monopoly of violence.The
logic of a ‘Tillian’ definition of the birth of the nation-state, where the legitimate use of force is
central to the process of state building, also applies in both cases (Binningsbø, 2005).
These processes of ‘creative destruction’ have several positive externalities that couldexplain the potential for war to be functional and instrumental to state building. Particularly in the
cases of Colombia and Sri Lanka, the results are not clear–cut, as variables such as taxation (see
figure 4.4), number of soldiers, number of rebels and number of battle deaths do not follow a
clear pattern that could support the ‘Tillian’idea of ‘all-out war’. It maybe that we are in the presence of a new breed of ‘new peace’ born in the womb of the ‘new wars’.
Questions arise regarding what imbalances this ‘new peace’ creates, and how they affect
other variables such as GINI, HDI, etc. It is important to remember that the logic of all-out war
obeys a particular series of beliefs and approaches, and it would be a flawed rationale to attempt
to frame a problem within a logic to which it does not fully correspond (see chapter four).
If this model produces a long term stability and results beneficial to the government, it
could become a new model for conflict management (Lewis, 2010: 667)in other countries that
have ‘weak states’; in fact, some adapted versions of these policies are already being implemented
(as plan Merida in Mexico, and in other Asian countries),
Finally, in the case of those insurgencies that aremotivated more by greed than political
or societal grievances, consideringpolitical solutions could be futile (Metz, 2007: 28), and the
need for a full scale military intervention could be more evident. But the difficulty remains; what
if the other armed group within the country has valid and legitimate grievances, and we are confusing means and ends, tactics and strategies?
25
Chapter 4. Theory,contradictions and practice: ontologies,
‘state’ making, and development
“Science depends on its concepts. These are the ideas which receive names. They determine the questions
one asks and the answers one gets. They are more fundamental than the theories which are stated in
terms of them.”(Sir George Thompson in Rose, 1991: 448)
This chapter will address some of the critiques and possible failures (conceptual and practical) of
the all-out war theory, attempting to ground the theory to its conceptual pillars and to practice,
in terms of the two case studies investigated in this document. I will try to assess and evaluate
the validity of the all-out war. In doing so, the chapter will first analyse some potential theoretical
gaps; this analysis contains discussions of the theory’s methodological elections and epistemological biases. The chapter then evaluates and contrasts the theory with the reality of both the Colombian and Sri Lankan cases, assessing whether some elements are informed by the liberal
peace approach that drove the all-out approach in post conflict practice. Finally, the chapter debates whether or not the all-out war is a ‘development model’.
4.1 Collateral damage: weak foundations of the all-out war theory
The use of econometrics is widespread in the literature of conflict studies. But quantity
does not necessarily mean quality; neither does it imply truth. Therefore, the fact that a theory or
a theoretical approach is widely used does not imply that is correct. As Medina puts it:
“The supply side theory is acquiring the undeserved status of a fact, putting down other
perspectives that were encompassed in the objective-causes school. Left unchecked, the
supply side theory is likely to cross the line that separates analysis from dogma, theory
from political weapon.” ( Medina, 2008:702-703)
The limitations of econometric studies in embracing an analysis of war can be characterized
into three areas; the method itself, the decisions related to the mathematical modelling of social
problems and the quality of data for the measurement of the variables.
Econometrical models usually entail an analytical approach to reality(Ackoff, 2001), where
the problem under scrutiny is dissected into parts, and the properties of the parts are studied, so
that the comprehension of different elements are aggregated to explain the whole. One of the
biggest weaknesses of this approach is that problems are addressed under a mechanistic logic
where systems operate under linear logics (Forrester, 2003), thus denying their endogenous nature and the importance of historical processes.Agency is conceptually limited to reactions to
changes in the environment (Cramer, 2011) which does not allow for the fact that war is a social
institution (Hoffman, 1963: 321). The reality is that social systems, particularly wars, are complex
(Sanin, 2009; Kaldor, 1999), and it is therefore necessary to focus on the comprehension of the
system as a whole.
These models also create a separation between politics and conflict, describing conflict as a
merely economic, rational decision. This disregards the political nature of war, and the fact that
violence becomes political when it alters the distribution of power within a nation (Steenkamp,
2011: 366, Medina, 2008:707). In the cases of both Colombia and Sri Lanka, a reconfiguration of
politics as a result of war can be observed.
Methodologically, models have been built under the assumption that each country is similar
because of the mere fact that they are experiencing war. The reality, however, is that each dynamic might display particular patterns, and therefore an analysis of each country can shed more
light on a particular conflict.It makes little sense to try to prove or disprove highly general theo-
26
ries using as evidence inherently different cases in different contexts (Medina, 2008: 705; Cramer,
2003; Murshed,2011).
Additionally, models that are used to analyse agent based theories, such as the greed hypotheses,are theoretical, because they do not have a model based on the idea of maximizing profitsfor the agents waging war.Their understanding of the reasons why actors go to war could
therefore be spurious (Murshed, 2009: 90).Models used to analyse these hypotheses are based on
the assumption of a single equilibrium, where the truth is that models can have multiple equilibriums, where war itself represents a particular equilibrium as opposed to peace (Medina, 2008:
704).
Furthermore,any kind of mathematical modelling (including econometrics) faces the challenge of trying to approach the measurement of variables that are hardly quantifiable. Given this
difficulty, modellers are left with two possible options: to ignore these variables, or to try to
measure them through proxies.
A proxy variable is the selection of a variable as an heuristic to the measurement of a certain
parameter. In econometrical studies related to conflict studies, proxy variables are used on a
common basis (almost indiscriminately and without any discussion on their selection). For example, studies that determine the role of inequality in conflict (Collier, 2004) approach inequality
and its measurement through the measurement of taxation.
As agent based theories try to seek evidence of the homo-economicus nature of conflict
(Cramer, 2002) they forfeit the role of their biases. For example, they approach the lootability of
resources through the presence of primary commodity exports, ignoring the role of shadow
economies and illicit trafficking (Brauer, 2010; Cramer, 2002: 1849), that, as in Colombia, can
account for as much 3% to 5% of the GDP (Sanin, 2009). These theories seem to prioritize an
explanation of the financiation of war, rather than its origins.
This bias is reinforced by the accessibility of data and more ‘reliable’ proxies for variables related to greed motivations; for example, GDP can be ascertained in more datasources, andis
therefore easier to measure as an indicator of the costs of war (Brauer, 2010) than ethnicity, religion, class and inequality.
The expression “trash in – trash out” is a term coined in mathematical modelling that argues
that the quality of the results of a model will be related with the quality of the model itself and
the quality of the data. Good models cannot correct bad data. In conflict situations, one of the
first things to fails is reliable information. Distributional data is notoriously poor and is even
worse in cases of inter-group conflict (Cramer, 2003: 400). In this context, if one wishes to study
a subject through ‘reliable’ data sources, there are few options, and there is no guarantee that data alone will be context insensitive.
Another problem is the reach and credibility of the conclusions at which these models arrive, as the same variables can support completely contrasting hypotheses. It is possible to find
models that recognize the role of ethnicity (Murshed, 2008: 368) or thatrefute its importance
(Fearon, 2003).
It would be unfair to say that econometrical models are bad per se, as almost any methodology that embraces data management will share similar problems. Therefore, it is necessary to
recognize that econometrical models are a valid tool for analysis, and work quite well in situations where linear non-complex relations exists (Forrester, 2003).However, their use is not a substitute for a conceptual analysis (Medina, 2008: 706), as economic factors do not declare war by
themselves (Offer, 1989 in Cramer, 2011:283).
4.2 Big Bang: ‘‘Tillian’’ Wars
27
An assessment of the elements of the ‘give war a chance’ theory in practice requires two
things; first, a summary of the constitutive elements of the theory, and second, a definition of
possible variables to measure these elements to analyse the theory. As was mentioned in chapter
three, the ‘give war a chance’ theory can be described by the four attributes presented in figure
4.1. Elements of this theory related to the Schumpeterian theory of creative destruction will not
be analysed in this document, given the difficulty of approaching the measurement of these variables, and the limitations on space.
Figure4.1
Quintessential elements of the ‘give war a chance’ theory
Source: Chapter three
4.2.1 Colombia: The good ‘‘Tillian’’ student with bad grades
In theory, Colombia could be analysed as a perfect example of a ‘‘Tillian’’ war, yet with
‘bad grades’. As can be seen in figure 4.2, the approach to the measurement of the variables of
state presence, strength of military forces and strength of the taxing apparatus, will be made
through the variables of government expenditure as a share of the GDP23, size of the military
forces, and revenue as a share of the GDP24.
It can be observed that, in the period between2002 and 2010, the country has almost
doubled the size of the armed forces within the country. This increase does not only account for
the number of soldiers, but also for the amount of weaponry and technology involved.The
SIPRI Arms Transfers Database shows that Colombia bought almost 100 aircraft in the period
of 2000-2010. This increase in weaponry, equipment and soldiers was accompanied by training
and capacitation in a reengineering process that started with the introduction of Plan Colombia
by the Pastrana Government.
23This
variable is the general government’s final consumption expenditure (formerly general government
consumption) and includes all government expenditures for purchases of goods and services (including
compensation of employees). It also includes most expenditures on national defence and security, but
excludes government military expenditures that are part of government capital formation (World Bank,
2011).
24Revenue is cash receipts from taxes, social contributions, and other revenues such as fines, fees, rent,
and income from property or sales. Grants are also considered as revenue but are excluded here (Ibid).
28
Parallel to this process we can observe that the increase of revenue as a share of the
GDP, increased in net terms in comparison with early 2000, and in relative terms to 2002, increased slightly after a dramatic reduction between 2007 and 2008.
Figure4.2
Size of the military forces, government expenditure and revenue in Colombia
450000
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
2008
2006
50000
2004
1990
100000
2002
150000
2000
200000
1998
250000
1996
300000
1994
350000
1992
24
22
20
18
16
14
12
10
8
400000
COLOMBIA GOVERNMENT EXPENDITURE AS
SHARE OF THE GDP
COLOMBIA REVENUE AS A SHARE OF THE GDP
SIZE OF MILITARY FORCES COLOMBIA
Source: Worldbank 2011, data retrieved from<http://data.worldbank.org/country/>
The state presence, understood as government expenditure, declines slightly, which could
refute the idea of a ‘‘Tillian’’ war, but in terms of figures, more or less maintains itself. The definition of state presence can alter the analysis, as Sanin(2009) defines state presence as strength of
armed forces and more/higher taxation. Another approach to the measurement of state presence
is that of Fearon(2003), who conceives of state presence through the variable GDP per capita,
which in the case of Colombia is increasing in net terms. Finally, the interpretation of the variable government expenditure must take into account the ideological bias of the government and
its budget management.In the case of Colombia, this bias is defined by some as neoliberal
(Robledo, 2007), while other authors describe it as neo-populist (Galindo, 2007).
It is interesting to note here that in spite of government efforts to weaken and destroy the
guerrillas, there have been a series of attacks from these groupsdemonstrating that, tactically and
strategically, they maintain their operational capabilities, and a possible military defeat is not yet
in sight.
4.2.2 Sri Lanka: The bad ‘‘Tillian’’ student with good grades
Sri Lanka appears to be less a ‘‘Tillian’’ war than Colombia,but can also be construed as a
successful ‘‘Tillian’’ war. For example, the size of the military forces is not as substantial as in the
Colombian case, but the size of the military increased during the years of the offensive (20072009). Interestingly, the size of military forces is smaller than in themid-1990s or early 2000s (see
figure 4.3). This statistic demonstrates a qualitative change in the capabilities of the Sri Lankan
forces. As the SIPRI Arms Transfers Database shows, the Sri Lankan army improved their weaponry, buying, over the course of the decade as many as 65 aircraft and 143 armoured vehicles.
At last the military strength of the Sri Lankan forces became incontestable, and they destroyed,
killed or captured most of the LTTE forces.The strengthening of the Sri Lankan armed forces
isthus clearly indisputable.
29
Figure4.3
Size of the military forces and revenue in Sri Lanka
19
17
250000
15
200000
13
150000
11
100000
9
50000
7
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1990
5
0
SRI LANKA GOVENRMENT EXPENDITURE AS
% OF THE GDP
SIZE OF MILITARY FORCES SRI LANKA
Source: Worldbank2011, data retrieved from <http://data.worldbank.org/country/>
Figure4.4
Revenues25and taxes on income in Sri Lanka
18
19
17.5
18
17
17
16.5
16
16
15
15.5
14
15
13
14.5
2008
2007
2006
2005
2004
2003
2002
2009
2008
2007
2006
2005
2004
2003
2002
2001
2001
12
14
TAXES ON INCOME PROFITS AND CAPITAL
GAINS SRI LANKA % OF REVENUE
SRI LANKA REVENUE
Source: Worldbank 2011, data retrieved from <http://data.worldbank.org/country/>
25Revenue
is cash receipts from taxes, social contributions, and other revenues such as fines, fees, rent, and income
from property or sales. Grants are also considered as revenue but are excluded here.
30
In this particular case, an understanding of state presence in terms of government expenditure or strength of military forces seems to best fit the definition of a ‘‘Tillian’’ war. Having
dramatically increased in the years in which the government carried out military offensives, government expenditure decreased sharplyafter the defeat of the LTTE.
The role of taxation seems to oppose ‘‘Tillian’’ logic. For example, when analysed, the revenues of the country appear to have a decreasing tendency (Figure4.4). However, when we compare this with the percentage of the revenues that come from taxes on income and capital gains,
revenues have an increasing tendency, showing higher taxation (in proportion and in amount)
which could illustrate a stronger taxing bureaucracy.
It would be premature to analyse or predict the longevity of peace in Sri Lanka, as only 2
years have passed since the ‘end’ of the war.Studies usuallyanalyse the recurrence of war over a
period of five yearsafter the end of hostilities, and this variable can therefore not be assessed. It
is necessary to point out that the peace achieved in Sri Lanka can be defined as a negative peace
(see section 4.3) (Galtung, 1967).
4.2.3 In summary: can Colombia and Sri Lanka be described as ‘successful’
‘Tillian’ wars?
Contrary to the arguments of some authors (Kaldor, 1999 on Sanin, 2009), ‘‘Tillian’’ wars
are possible, and are a reality in a globalized world.The cases of Sri Lanka and Colombia are, to
some extent, particular cases that obey Tilly’s theory.
In either case it is noteworthy that the decision to choose the military option was marked
by uncertainty and unclear results (Kanheman, 2003). In terms of scale, Colombia has exerted a
greater effort, but has not achieved the same results as Sri Lanka. Therefore, there are elements
beyond the three described by Tilly that must be incorporated intoour comprehension of the
effectiveness of the all-out war theory. Such elements, among others, are; the nature of the insurgency, thetactics used by the insurgents, their structure,their adaptability and the terrain on
which the war is fought. In the case of Colombia, dramatic changes in the nature of the tactics
used ‘rebalanced’ the stakes against a possible military victory.
In the case of Sri Lanka, where the government managed to achieve victory, as the LTTE never
changed their strategies, and the military might of the army was sufficient to defeat the LTTE
forces in a smaller area (see table 2.1).
In terms of other development variables, we can observe results that raise questions
about the success of the outcomes of the war. Independent of the outcome, any analysis must
not forget that the definition of success for a war must be political, not economic (Medina, 2008:
709). For example, if we analyse the GDP growth, both countries are atypical. Sri Lanka’s economy showed an average growth of 4.5% prior to war, but a wartime growth of 4.8%
(Worldbank, 2011). In the case of Colombia, growth was at4,23% before the military escalation,
compared to4,37% in war times. In the case of Sri Lanka, the peace dividend appears to be higher than predicted- the expected peace dividend after peace has been achieved is 1% (Hoeffler,
2009). In 2010, Sri Lanka’s GDP growth was higher than 8%.Therefore, on a descriptive analysis
basis, it appears that the military escalation positively affected the GDP.
In practice, it is difficult to separate security and development agendas (Murshed, 2011),
and military victories or military campaigns are related to particular development projects. As
development (or its failure) may itself be a cause of violence (Abeyrane, 2004; Murshed, 2011) it
is necessary to assess the development models embraced in both countries.
31
4.3 Death (practice): Is war producing a particular model of development
and state making?
In both Sri Lanka and Colombia, during and after the military escalation, a new development model was established. The model in both cases is a top-down take on development
(Höglund, 2010: 371), that seeseconomic growth as central for development.
This is a by-product of the way the all-out war conceives of victory and frames development. It could be defined as a ‘liberal peace’ model, in which governments focus on security as a
precondition for economic growth and development (Colombia), where economic growth and
stabilization are sought as a consolidation of victory (Goodhand, 2010:346), and where grievances are reduced to an economic element (Sri Lanka).
As a military victory is not yet foreseeable for Colombia, it is difficult to contrast pre-war
and the post-warenvironments. Part of the strategy to weaken the insurgent forces, besides the
strengthening of the military, is to provide education, electricity,shelterand water in conflict areas
in means similar way to Sri Lankan post-conflict policies (Goodhand, 2010: 346).
This approach to post conflict development disregards the facts that inequality could have
played a role in the outburst of conflict itself (Cramer, 2003), and that the possibility of increased
inequalities can refuel or create new grievances, thus fostering new conflicts or reigniting existing
ones (Steenkamp, 2011: 636; Murshed, 2011).
Some authors refer to such exercises, which forget that the involvement of the parties affected by new policies is necessary to achieve a sustainable arrangement(Ahmed, 2005: 169), as
‘power building’ (Goodhand, 2010).These exercises of peace building can be framed as examples
of a liberal peace model.
4.3.1 The ‘liberal peace’
Both Colombia and Sri Lanka can be analysed as different examples of an implemented‘liberal
peace’model. However, according to the Kantian definition of a liberal model, I would argue that
they are non-liberal development models.
For Doyle (1983),the model can be understood as a process in which a post–conflict country
seeks to promote a market economy, strong political rights, political liberties and polities that are
externally sovereign. Doyle later presents an alternative definition, where human security replaces
sovereignty. The idea is simple; if a society wants to avoid conflict, it should embrace the ‘liberal’
principles(Doyle, 2005).
Despite the fact that liberal peace was initially used to explain peace among nations, the concept
has expanded, and been applied in an intra-state framework (Murshed, 2009)with widespread use
(Salih, 2009). Opinions regarding the role of the liberal peace model and conflict are mixed, and
some authors (Salih, 2008) argue that it is a process of economic and political liberalization, not
focused on the social conditions of the population. Also, in some cases, the model has produced
destabilizing effects that undermined the consolidation of peace. Other authors(Paris, 2010: 341)
argue that the liberal peace has worked in some particular contexts, such as Salvador.
It can be asserted that ‘liberal peace’alone is not sufficient to achieve positive peace (Galtung,
1967) in a post-conflict society, where democratization and marketization were believed to be
magical formulae (Paris, 2010: 338).In practice, the all-out war can be framed as a deliberate design for theconsolidation of peace, which appears to be informed by some of the elements of
Doyle’s definition (Murshed, 2009).
The importance of liberal peace in understanding the all-out war practice is a by-product of the
common practices shared by both approaches. Both prioritize economic growth and the presence of political rights. In Colombia and Sri Lanka, liberal peace models had previously been im32
plemented, in Sri Lanka as part of the peace process in 2002 and in Colombia in relation to the
1991 constitution. In Colombia the model was implemented as a means to opening the economy
and political system, and increasing public investment. In Sri Lanka its implementation during
the peace process of 2002 received strong opposition, as it was opposed to the statism and welfarismof the country’s reforms (Stokke, 2009: 936).
4.3.2 The illiberal peace?
I argue that both countries embraced models that can be defined as ‘illiberal peace development models’; they are ill-liberal in the sense that they do not fulfil the requisites of Doyle’s
(2005) approach. In order to asses this, economic openness, the political terror scale and the percentage of debt to GDP will be assessed as proxy variables (see section 4.1) for analysing the
promotion of a market economy, political rights, civil liberties and sovereignty (see figures 4.5,
4.6 and 4.7).
Figure4.6
Economic openess26
Figure4.5
Government Debt
70
68
66
64
62
60
58
56
54
52
60.00
50.00
2009
2007
2005
2003
2001
40.00
Colombia
Central
government
debt, total
(% of GDP)
SRI LANKA
2010
70.00
COLOMBIA
2008
80.00
2000
90.00
2006
Sri Lanka
Central
government
debt, total
(% of GDP)
2004
100.00
2002
110.00
Source: The Heritage Foundation and The Wall Street Journal Index of
Economic Openness. Data Retrieved from
<http://www.heritage.org/index/Ranking.aspx>
Source: Worldbank 2011, data retrieved from
<http://data.worldbank.org/country/>
As a share of its GDP, Sri Lanka has reduced its debt,thereby increasing its sovereignty.
However, the country has embraced a different model of liberal peace, as their economic openness is reduced and the terror has increased. This arrangement shows signs of being a nationalistic modelthat could be described as an authoritarian regime with some good ‘development’ indicators.
Both countries are alsoilliberal democracies according to Doyle (1998;2005), and in spite of
improvements in some areas, certain negative features outweigh the gains of democracy, sovereignty, or civil rights. Regardless, it can be asserted that in both cases these models share a notion of development that centres on and is informed mostly by economic improvement as a
common pattern, that relies on a top-down definition of development and is a by-product of reducing grievances to a mere economic proxy.
26The
higher the score the more open is an economy. The variable is on a scale between 0 and 100.
33
Figure4.6
Political terror scale for Colombia and Sri Lanka27
5.5
5.0
4.5
4.0
3.5
COLOMBIA
3.0
SRI LANKA
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
2.5
Source: Political Terror 2011Retrieved from <http://www.politicalterrorscale.org/>
Figure4.7
Displaced population of Colombia
4500000
4000000
3500000
3000000
2500000
2000000
IDP's
COLOMBIA
1500000
1000000
500000
2010
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
0
Source: UNCHR statistical population database 2011
This raises the question of the real chances of achieving success without the support of the
population (Richmond on Paris, 2010). Sri Lanka and Colombia appear to be emerging as governments without a strong social contract, where economics seems to outweigh other notions of
development, an outlook which forgets that real peace requires reconciliation. As a bridge between these gaps, I propose that these notions of development (informed by agent based theories and the liberal peace practice) should also embrace Doyle’s interest in involving human security as a variable (2005).
This is because human security argues that security (central to the all-out war perspective)
should emerge and refer to the individual. This concern is therefore linked with agent based the27A
yearly report measuring physical integrity rights violations worldwide. The PTS measures levels of
political violence and terror that a country experiences in a particular year based on a 5-level ‘terror scale.’
The data used in compiling this index comes from two different sources: the yearly country reports of
Amnesty International and the U.S. State Department Country Reports on Human Rights Practices.
Thehigher the value, the higher the terror and political violence.
34
ories, and allows a broader perspective on the possible consequences of all-out war. It can also
incorporate elements that tackle the consequences of the military campaign, such as the decrease
in political freedom (figure 4.6) IDP’s (see figure 4. 7), or some the negative externalities of war
and their costs (Sanin, 2009).
4.4 Conclusion: Biases and implications
“Men are disturbed not by things, but by the view which they take of them.”
Epictetus
Our epistemological biases determinewhat we see and how we analyse it; all-out war is not
an exception, and, as this chapter has shown, there are assumptions and pitfalls linked with the
nature of the methodologies and biases implicit in the approach to the study of civil wars. Given
that no theory is perfect, it is unfair to dismiss a theory as bad per-se. However, the real problem
emerges when we forget the nature of the methodologies we use and the biases/problems these
entail.
The Colombian and Sri Lankan conflicts have embracedall-out war with different results. As
mentioned in chapter three, Sri Lanka seems to be a more successful case, but their success does
not fit a ‘‘Tillian’’ process of state construction. On the other hand, Colombia, which can be seen
as a perfect ‘‘Tillian’’ example,albeit with many negatives and difficulties, is being praised as a
possible solution to face the challenges of the war in Afghanistan (Wolfovitz, 2011).
‘Tillian’ war exists, in spite of the hypothesis of Kaldor (1999). However, it is interesting to
note that in both case countries, the conflict is dominated by intra-national dynamics. Therefore,
in these particular conflicts, Kaldor’sargument might still be said to apply.
Particular illiberal peace development models with different policies have been implemented
in both countries, but in both the market has been seen as an element central to development.
These models have focussed on the improvement of economic conditions, and neglected other
possible approaches to development. Regardless of the approach employed towards state making, it is necessary to understand the role of development in state making and war making.
These approaches haveseveral consequences for development which can be difficult to
trace. It is therefore necessary to use an ‘agent-based development’ system for an agent based
theory. The notion of ‘agent-based development’ should embrace a human development/human
security perspective, because this approach could prove useful in guaranteeing development that
can support and consolidate peace, at the same time creating legitimacy for the government.
35
Chapter 5. Conclusions and reflections
This research approached the study of the all-out war policy and its operation in two countries that could be described as classical protracted conflicts. As war is part of a policy making
process in several countries in the world (especially in those with civil wars), understanding these
two cases can provide information on a particular means of ensuring state presence.
Nocivil war analysis can be separated from a detailed analysis of its environment, which allows for the comprehension of particular dynamics affecting decision making processes. As documented in chapter two, the call for a military option to achieve peace emerged within the
boundaries of the nations, as a different means to approach for the quest for ‘peace’.
War itself is unpopular, and can only be justified after failed attempts at conducting a peace
settlement. In both case countries, peace settlements failed for myriad reasons, such as poor negotiation skills, instrumentalization of the peace process itself, lack of compromise and commitment from the parties. Most importantly, there was a lack of real involvement from civil society
and other political parties, which kept the peace processes captive from particular agendas, and
eroded the possibility of broad public support and commitment for peace.
It can be said that war emerged as the governments were unable to settle internal political
and societal disputes, and not simply because there was a chance of profiting from war. For the
irregular groups waging war, their ability to find funding has been an important element explaining their continued existence despite changes inthe international landscape. While the interest in
profit certainly played a role in the emergence of the war, it was not by itself sufficient cause for
conflict.
Discourses have been instrumentalized for consolidating or justifying war in both countries.
Interestingly, the demonization of the armed groups has been instrumental in building up this
public support. The post 9/11 environment created a perfect framework in which justify the war
as legitimate in an international and intra-national context.
I argue that the Colombian and Sri Lankan cases present a new breed of peace, as seen in
the work of Kaldor (1999). I also argue thatpost 9/11 discourse has constructed a series of
strong conceptual blinders. The truth is that war, by definition, has always been savage, and its
idealization in discourse and the media will notassist conflict resolution and conflict studies.
Both countries have followed their own versions of what could be called a ‘‘Tillian’’ war, arguing the necessity of strengthening the state and achieving a monopoly of force. In both cases,
the countries have displayed elements consistent with the theory, but also adaptations of the theory particular to their context.
If the Colombian and Sri Lankan models achieve a stability that appeases their population as
well as their political elites, we may be entering into the age of a “new peace” model which bases
its peace seeking operations on the idea of force and state strengthening.
These‘‘Tillian’’ wars emerged after failed peace processes, but the particular reasons for the
failure of peace processes in each case can shed light on the conflict resolution area, and can
prove useful to future peace processes.
A ‘Tillian’ war is easily justified in the framework of a purely greed-driven enemy, but when
other motivations exist, we face the problem of confusing means and ends, tactics and strategies,
and executing policies informed by this confusion.
The investigation found that,contrary toKaldor’s theories, in a post-globalization planet,
‘‘Tillian’’ warsdo exist,and can succeed. Their existence is particularly related to illiberal peace
development models. These ‘development’ models considerthe improvement of economic con-
36
ditions of persons as central to their approachand important to the post-conflict reconstruction
effort.
War itself does not guarantee a military victory; as the Colombian case shows, the organizational nature and adaptability of the armed groups, together with strategic natural safehavens
(Wolfovitz, 2011) can make a decisive military victory difficult.
These approaches’consequences for development will be determined by the epistemological
and ontological biases which we adopt, so in order to do a proper assessment of the consequences of these policies, it is also necessary to reflect on the biases that inform our notions of
what development is. This study could not find strong evidence to support the idea of higher
structural violence. Rather,there seems to be an increase of violence accompanying the reduction
of political liberties.
‘Tillian’ wars and policies towards state strengthening have been underresearched, so that a
gap exists between the literature and the practice. There is also a gap between conflict and development, which dwells on the peace dividend, and ignores the fact that peace involves a political
component.
Finally, if war is a process that involves structure, agency and discourse, it is imperativeto
abandon the tendency to study it only from one angle; but rather to address conflict from a position which simultaneously includes structure, agency and discourse.
.
37
References.




















Abeyrane, Sirimal (2004) ‘Economic Roots of political conflict’, The World Economy
27(8):1295-1314.
Ackoff, Rusell (2001) ‘OR: after the post mortem’, System Dynamics Review 17(4): 341–346.
Addison Tony and MurshedMansoob (2002) ‘Credibility and Reputation in Peacemaking’. Journal of Peace Research39 (4): 487-501.
Andrade, Gustavo (2010) Los caminos a la violencia. Vinculación y trayectorias de los niños en los
grupos armados ilegales en Colombia.2010. Bogotá:EdicionesUniandes.
Ahmed, Salman (2005) ‘Review: No Size fits All: Lessons in Making Peace and Rebuilding States’ Foreign Affairs 84(1):162-169.
Amnesty International (2010) ‘Sri Lanka - Amnesty International Report 2010’. Accessed
20 June 2011 <http://www.amnesty.org/en/region/sri-lanka/report-2010>.
Amnesty International (2011) ‘Annual Report 2011: The state of the world's human
rights
(Colombia)’.
Accessed
15
August
2011
<http://www.amnesty.org/en/region/colombia/report-2011>.
Amnesty International (2011) ‘Annual Report 2011: The state of the world's human
rights (Sri Lanka)’. Accessed 15 August 2011 <http://www.amnesty.org/en/region/srilanka/report-2011#ai-reports>.
Arambewela,Nadeeka and Arambewela, Rodney. (2010). ‘Post-war opportunities for
peace in Sri Lanka: and ongoing challenge?’ Global Change, Peace and Security 22 (3): 365375.
Ballentine, K. and Sherman, J. (eds) (2003) The political economy of armed conflict. Beyond Grief
and grievance. The International Peace Academy. New York.
Bates, Robert (2008) ‘State Failure’ Annual Review of Political Science 11:1-12.
Beardsley. K. (2008) ‘Agreement without Peace? International Mediation and Time inconsistency problems’ American Journal of Political Science 52(4): 723-740.
Beardsley, Kyle (2009)‘Rebel Groups as Predatory Organizations The Political Effects of
the 2004Tsunami in Indonesia and Sri Lanka’, Journal of Conflict Resolution53 (4): 624-664.
Besley, Timothy and PerssonTorsten (2010) ‘State Capacity, Conflict, and Development’ Econometrica78 (1) :, 1–34.
Binningsbø, Helga (2005) ‘Consonciational Democracy and PostconflictPeace. Will PowerSharing InstitutionsIncrease the Probability of Lasting Peace after Civil War?’. Paper presented at the Annual Political Science Conference, Hurdalsjøen, Norway (5-7 January).
Brauer, Jurgen and Dunne, J. (2010) ‘On the cost of violence and the Benefits of Peace’.
Paper presented at the Fourteenth Annual International Conference on Economics and
Security, Izmir, Turkey (17-18 june)
Bushnell, David (1996) ‘Colombia. UnaNación a pesar de símisma’. Bogotá: Ed. Planeta.
Chernik, Marc(2005) ‘Economic resources and internal armed conflicts. Lessons from
the Colombian Case’, In Arnson Cynthia and Zartman William (eds) Rethinking the economics of War. The intersection of need, Creed, and Greed.Woodrow Wilson Center Press; Baltimore.
Cohen, Y., B. Brown, & A. Organski (1981) ‘ The paradoxical Nature of State Making:
the violent creation of order’ The American PoliticalScienceReview. 75 (4): 901-910.
38





















Coleman, P. (2000) ‘Intractable Conflict’ in Morton Deutsch and Peter Coleman (eds.)
Handbook of Conflict Resolution. San Francisco: Jossey-Bass.
Collier, Paul, HoefflerAnke and Söderbom Mans (2004) ‘On the Duration of Civil War’,
Journal of Peace Research: 41 (3): 253–273.
Collier, D. Hoeffler A. and Söderbom, M (2008) ‘Post-conflict Risks’ Journal of Peace Research. 45: 461-477.
Cramer, C. (2002) ‘Homo Economicus Goes to War: Methodological Individualism, Rational Choice and the Political Economy of War’, World Development 30 (11):1845-1864.
Cramer, C (2003) ‘Does Inequality cause conflict?’ Journal of International Development
15:397-412.
Cramer, C. and Richards Paul (2011) ‘’Violence and war in Agrarian Perspective’ Journal
of Agrarian Change11(3):277-297.
De Mel, Neloufer (2009) ’Gendering a New Security Paradigm in Sri Lanka’ IDS Bulletin
40 (2). 36-40.
DeRouen, Karl; Ferguson Mark J, Norton Samuel,Hwan Park Young, Lea Jenna and
Streat-Bartlett Ashley (2010)‘Civil war peace agreement implementation and state capacity’ Journal of Peace Research47 (3): 333-346.
Devotta, Neil (2000) ‘Institutional Decay, and the Quest for Eelam: Explaining Ethnic
Conflict in Sri Lanka’ Pacific Affairs 7 (1): 55-76.
Devotta, Neil (2005) ‘From ethnic outbidding to ethnic conflict: the institutional bases
for Sri Lanka’s separatist war’ Nations and Nationalism 11(1): 141-159.
Devotta, Neil (2009) ‘The Liberation Tigers of Tamil Eelam and the Lost Quest for Separatism in Sri Lanka’ Asian Survey 49(6): 1021-1051.
Devotta, Neil (2011) ‘Sri Lanka: From turmoil to Dynasty’ Journal of Democracy 22 (2): 130144.
Dexter, Helen (2008) ‘The ‘New War’ on Terror, Cosmopolitanism and the ‘Just War’
Revival’ Government and Opposition 43 (1): 55-78.
Diaz, F. (2009) ‘Magical Realism: The evolution of paramilitarism in Colombia’. Paper
prepared for the System Dynamics World Conference 2009. Albuquerque, USA (22-26
July).
Dower, Nigel (2009) Pacifism, Non-Violence and the Way of PeaceThe Ethics of War and Peace.
Cambridge: Polity, Cambridge.
Doyle, W. (1983) ‘Kant, Liberal Legacies, and Foreign Affairs’, Philosophy & Public Affairs
12 (3): 205-235.
Doyle, W. (2005) ‘Three Pillars of the Liberal Peace’, American Political Science Review 99(3):
463-466.
Duncan, Gustavo (2004) ‘Violencia y conflicto en Colombia comounadisputapor el control del estado en lo local’. Documento CEDE 2004 -11. Bogotá: Universidad de los
Andes- CEDE.
Duncan, Gustavo (2005). ‘Del campo a la ciudad en Colombia. La infiltraciónurbana de los señores
de la Guerra’. Documento cede 2005-2. Bogotá: Universidad de los Andes /CEDE.
Duouet, Veronique (2008) ‘Negotiating conflict Settlements: Lessons learnt and challenges’. Roundtable Meeting Report. Schwanenwerder, Berlin. Berghof Project. “Resistance/Liberation Movements and Transitions to Politics: Building a Network of Experience”
Fearon, J. (1995) ‘Rationalist explanations of war’ International Organization 49 (3): 379414.
39




















Fearon, James and Laitin, David (2003) ‘Ethnicity, Insurgency, and Civil War’ American
Political Science Review. 97(1): 75-90.
Fearon, J. (2004) ‘Why do Some Civil Wars Last so Much Longer than Others?’ Journal of
Peace Research 41 (3) : 275-301.
Forrester, Jay (2003) ‘Dynamic models of economic systems and industrial organizations’, System Dynamics Review 19 (4).
Frerks, Georg and Klem, Bart (2005) ‘Sri Lankan discourses on Peace and Conflict’,InFrerks, Georg and Klem Bart (eds) Dealing with Diversity: Sri Lankan Discourses on
Peace and Conflict, pp 1-46. The Hague: Netherlands Institute of International Relations.
Galindo, C (2007) ‘Neopopulismo en Colombia: el caso del gobierno de Álvaro Uribe
Vélez’, Iconos. Revista de CienciasSociales 27: 147-162.
Galtung, J. (1969) ‘Violence, Peace, and Peace Research’ Journal of Peace Research
6(3): 167-191.
Garrison, S. (2009) ‘Competition, Greed, or Inertia: The Dynamics of Civil War Escalation’ Paper presented at the annual meeting of the ISA's 50th ANNUAL
CONVENTION "EXPLORING THE PAST, ANTICIPATING THE FUTURE",
New York 15th of February).
Goodhand, Jonathan and Walton, Oliver (2009) ‘The Limits of Liberal Peacebuilding?
International Engagement in the Sri Lankan Peace Process’ Journal of Intervention and
Stetaebuilding 3 (3): 303-323.
Goodhand, Jonathan (2010) ‘Stabilising a victor’s peace? Humanitarian actionand reconstruction in eastern Sri Lanka’, Disasters 34(3): 342-367.
Gorka, S. And Kilcullend, D. (2011) ‘An Actor Centric Theory of War’. Policy Essays.
Washington: Council for emerging national security affairs.
Giersch, Herbert (1984) ‘The age of Schumpeter’, The American Economic Review 74 (2):
103-109.
Harris, M. (2011) ‘Choosing Escalation during War: Considering the Evolutionary Dynamics in the Intra-War Phase’. Paper presented at the annual meeting of the International Studies Association Annual Conference "Global Governance: Political Authority
in Transition, Quebec (16th of March).
Hartzell, C. Hoddie, M. and Rothchild D. (2001) ‘Stabilizing the Peace After Civil War:
An Investigation of Some Key Variables’ International Organization 55(1): 183-208.
Hewitt, J. Wilkenfield, J. and Gurr, T. (2010) ‘Peace and Conflict 2010’. Center for International Development and Conflict Management. Maryland:University of Maryland.
Hoeffler, A. (2009) ‘State Failure and Conflict Recurrence’. Working Paper. Center for
the Study of African Economies. Oxford: University of Oxford.
Hoffman, Stanley (1963) ‘Russeau on War and Peace’ The American Political Science Review
57(2): 317-333.
Höglund, K. and Söderberg, M. (2010) ‘Beyond the Absence of War: the diversity of
peace in post-settlements societies’, Review of International Studies 36: 367-390.
Höglund, Kristine and Orjuela, Camilla (2011) ‘Winning the peace: conflict prevention
after a victor’s peace in Sri Lanka’ Contemporary Social Science 6(1): 19-37.
Hoole, Rajan (2009) ‘Sri Lanka: Ethnic Strife, Fratricide, and the Peace vs. Human Rights
Dilemma’ Journal of Human rights practice 1(1): 120-139.
Human Rights Watch (2009) ‘Sri Lanka: Protect Civilians in ‘Final’ Attack’. Accessed 30
june 2011 <http://www.hrw.org/news/2009/04/20/sri-lanka-protect-civilians-finalattack>.
40

























Huth, Paul and Russet, Bruce (1988) ‘Deterrence Failure and Crisis Escalation’ International Studies Quarterly 32 (1) : 29-45.
Instituto de Estudios Políticos y Relaciones Internacionales (IEPRI) (2006). Nuestra Guerra Sin Nombre. Transformaciones del Conflicto en Colombia Bogotá: Grupo Editorial Norma.
Jacoby, Tim (2008) Understanding Conflict and Violence: Theoretical And Interdisciplinary Approaches. Routledge
Jayasundara, Shyamika (2011) ‘Conflict, War and Peace in Sri Lanka- Politics by Other
means?’. Unpublished article. The Hague: Institute of Social Studies.
Kaldor, M. (1999). New and old wars: Organized Violence in a Global Era. Stanford Calif:
Stanford University Press.
Kaldor, Mary (2005) ‘Old Wars, Cold Wars, New Wars, and the War on Teror’ International Politics 42: 491-498.
Kalyvas, Stathis N. (2001) ‘ "New" and "Old" Civil Wars: A Valid Distinction?’, World
Politics, 54 (1): 99-118.
Kalyvas, Stathis and Balcells,Laia (2010) ‘International System and Technologies of Rebellion:How the End of the Cold War Shaped Internal Conflict’ American Political Science
Review (2010) 104: 415-429.
Kanheman, Daniel and Renshon, Jonathan (2003) ‘ Why Hawks Win’, Foreign Policy 158:
34-38.
Kilcullen, D. (2006) ‘Counter-insurgency Redux’, Survival 48(4):111-130.
Kinross, S. (2004).’Clausewitz and Low-Intensity conflict’, Journal of Strategic Studies. 27
(1):35-58.
Kiszely, J. (2006) ‘Learning about Counter-Insurgency’ Rusi Journal December.
Korf, Benedikt (2006) ‘Functions of Violence revisited: greed, pride and grievance in Sri
Lanka’s civil war’ Progress in Development Studies 6 (2): 109-122.
Korf, Benedikt. 2006. ‘Who is the rogue? Discourse, power and spatial politics in postwar Sri Lanka’, Political Geography 25:279-297.
Kriesberg, Louis (1998). ‘Constructive Conflicts: From Escalation to Resolution’. Lanham, Maryland: Rowman and Littlefield.
Kurtz M. (2009) ‘The Social Foundations of Institutional Order: Reconsidering War and
the “Resource Curse” in Third World State Building’Politics & Society 37 (4): 479-520.
Leal, Francisco (2006) En la encrucijada: Colombia en el siglo XXI. Bogotá: Editorial Norma.
Lewis, David (2010) ‘The failure of a liberal peace: Sri Lankas counter insurgency in
global perspective’ Conflict Security and Development 10 (5): 647-671.
López C. (ed) (2010) Y refundaron la patria… de cómomafiosos y políticosreconfiguraron el Estado
Colombiano.- Bogotá: Corporación Nuevo Arco Iris –Debate editors.
Luttwak, Edward N. (1999) ‘Give War a Chance’, Foreign Affairs 78(4): 36-44.
Mann, M. (2002) ‘The Crisis of the Latin American Nation State’. Paper presented at the
conference “The Political Crisis and Internal Conflict in Colombia”, Universidad de los
Andes, Bogota (10-13 April).
Mattes, M. and Savun, B. (2009) ‘Fostering Peace After Civil War: Commitment problems and Agreement Design’ International Studies Quarterly 53: 737-759.
Medina, Luis (2008) ‘A critique of "resource-based" theories of colombia’s civil war’, Analisis Politico 21 (62): 44-57.
Mejía Quintana Óscar and Múnera Ruiz Leopoldo (2006) ‘Constitución, democracia y
Estado autoritario en Colombia’, CienciaPolítica 6: 80-108.
Metz, Steven (2007) ‘New Challenges and Old Concepts: Understanding 21st Century Insurgency’ Parameters Winter:20-32.
41






















Mills, G. (2007) ‘Ten Counterinsurgency Commandments from Afghanistan’. Foreign
Policy Research Institute.
Murshed, M. (2006) ‘Indivisibility, Fairness, Farsightedness and their Implications for Security’. United Nations University Research paper No. 2006/28. Japan: United Nations
University.
Murshed, M. and Verwimp, P. (2008a) ‘Enforcing peace agreements through Commitment Technologies’ United Nations University Research paper No. 2008/45. Japan:
United Nations University.
Murshed M. (2008b) ‘National Business, Civil war Abatement and Peacebuilding. Contributions to Conflict Management’ Peace Economics and Development 7: 365-381.
Murshed, Mansoob (2009a) ‘Revisiting the Greed and Grievance Explanations for violent Internal Conflict’, Journal of International Development 21: 87-111.
Murshed, Mansoob (2009b) ‘The Liberal Peace and Developing Countries’. Unpublished
article. The Hague: Institute of Social Studies.MurhsedMansoob (2010) Explaining civil war
: a rational choice approach. Cheltenham, UK ; Northampton, MA : Edward Elgar.
Murshed, Mansoob (2011) ‘The economics of conflict and peace’. Unpublished paper, Institute of
Social Studies, Erasmus de Rotterdam University.
Nakamura, Leonard (2000) ‘Economics and the New Economy: The Invisible Hand
Meets Creative Destruction’ , Business Review July.
Orjuela, Camilla (2003) ‘Building Peace in Sri Lanka: A Role for Civil Society?’ Journal of
Peace Research 40(2): 195-212.
Page, V. (2003) ‘Scraps of paper? Agreements and the durability of peace’ International Organization 57: 337-372.
Palacios, Marco (1995)Entre la legitimidad y la violencia: Colombia 1875–1994. Santafé de Bogotá, Colombia: Grupo Editorial Norma.
Pardo, Rueda Rafael (2007) Fin del Paramilitarismo. ¿Esposiblesudemonte?. Bogotá: Ediciones
B - Grupo Zeta.
Paris R. (2010) ‘Saving Liberal Peace’, Review of International Studies 36: 337-365.
Pearson, F. Olson, M. Walker, S. and Mann, S. (2006) ‘Rethinking Models of Civil War
Settlement’ International Interactions 33(2): 109-128.
Pecaut, Daniel (2008) ‘Las FARC: Fuentes de suLongevidad y de la Conservacion de su
Cohesion’ Analisis Politico 21(63)
Posada-Carbó Eduardo (2001) ¿Guerra Civil? El Lenguaje del conflicto en Colombia. Bogotá:
Alfaomega.
Posada-Carbó Eduardo (2011) ‘Colombia after Uribe’, Journal of Democracy 22 (1): 137 –
151
Posada Jaime (2004) Manual de historia deColombia. Bogota Colombia: Editorial Planeta.
Quinn, J. Mason D. and Gurses M. (2007) ‘Sustaining the Peace: Determinants of Civil
War Recurrence’ International Interactions 33 (2): 167-193.
Rangel, Alfredo (2001) Guerra Insurgente. Conflictos en Malasia, Perú, Filipinas, El Salvador y Colombia. Bogota: IntermedioEditores.
Regan, Patrick M. (1996) ‘Conditions of Successful Third-Party Intervention in Intrastate
Conflicts’. Journal of Conflict Resolution40 (2): 336-359.Regan, Patrick (2002) ‘Third-party
Interventions and the Duration of Intrastate Conflicts’ Journal of Conflict Resolution 46 (1):
55-73.
Richani, N. (2002). Systems of Violence: The Political Economy of War and Peace in Colombia.
New York : Suny Press.
42





















Richani, N.(2007), ‘Caudillos and the crisis of the Colombian State: fragmented sovereignty, the war system and the privatization of counterinsurgency in Colombia’ Third
World Quarterly 28 (2): 403-417.
Robledo, Jorge (2007) ‘La globalización neoliberal niega la democracia’ . in Hoyos,
Guillermo (ed)Filosofía y teoríaspolíticas entre la crítica y la utopíapp69-82. Buenos Aires:
CLACSO, ConsejoLatinoamericano de CienciasSociales.
Rojas, Cristina (2009) ‘Securing the State and Developing Social Insecurities: the securitisation of citizenship in contemporary Colombia’ Third World Quarterly 30(1):227-245.
Romero, Mauricio (2007) Parapolìtica: La ruta de la expansiónparamilitar y los acuerdopolíticos.
Corporación Nuevo Arco Iris. Bogota: IntermedioEditores.
Ropers Norbert (2008) ‘Systemic Conflict Transformation: Reflections on the conflict
and Peace Process in Sri Lanka’. Berghof Handbook Dialogue. 6: 11-43. Berlin: Berghof
Research Center for Conflict Management.
Salih, M. (2009) ‘A critique of the political economy of the Liberal Peace: Elements of an
African Experience’. In Edward Newmann, Roland Paris and Oliver Richmond (eds.)
New perspectives on the Liberal Peace, pp 133-155.Tokyo, United Nations University.
Sambanis, Nicholas (2004) ‘What Is Civil War? Conceptual and Empirical Complexities
of an Operational Definition’ The Journal of Conflict Resolution 48(6): 814-858.
Sanín, Francisco (2004)’ Criminal Rebels? A Discussion of Civil War and Criminality
from the Colombian Experience’,Politics & Society 32 (2): 257-285.
Sanin, Francisco and Barón, Mauricio (2005) ‘Re-stating the state: paramilitary territorial
control and political order in Colombia (1978-2004)’ Crisis states research programmeWorking paper No. 66. London School of Economics Crisis States Research Center.
Semana (2011) ‘¿Dóndeestá la llave de unanegociación entre Gobierno y Farc?’. Accessed
20 October 2011 <http://www.semana.com/nacion/donde-esta-llave-negociacionentre-gobierno-farc/152024-3.aspx>.
Sepp, K. (2005) ‘Best Practices in Counterinsurgency’ Military Review May-June.
Shastri, Amita. 2009. ‘Ending Ethnic Civil War: The Peace process in Sri Lanka’ Commonwealth & Comparative Politic. 47 (1): 76-99.
Sørensen, G. (2001) ‘War and state making- why doesn´t it work in the third world?’. Paper prepared for the Failed States Conference, Florence (10-14 April).
State Department (2001) ‘2001 Report on Foreign Terrorist Organizations’, State Department Office of the Coordinator for Counterterrorism. Accessed July 24th 2004
<http://www.state.gov/s/ct/rls/rpt/fto/2001/5258.htm>
Steenkamp Christina (2011) ‘In the shadows of war and peace: making sense of violence
after peace accords’, Conflict, Security & Development 11(3): 357-383.
Stokke, Kristian (2009) ‘Crafting Liberal Peace? International Peace Promotion and the
Contextual Politics of Peace in Sri Lanka’, Annals of the Association of American Geographers
99 (5): 932-939.
Sweig, Julia (2002) ‘What Kind of War for Colombia?’ Foreign Affairs 81 (5): 122-141.
Tilly, Charles (1991) ‘ War and State Power’, Middle East Report 171:38-40.
Tilly, Charles (2004) ‘Terror, Terrorism, Terrorists’, Sociological Theory 22 (1): 5-13.
Toft, Monica (2010) ‘Ending Civil Wars: A Case for Rebel Victory?International Seurity34
(4):7-36.
Tokatlian, Juan (2004) ‘Una reflexion en torno a Colombia. 199-2002: ?negociacionpara
la paz o procesopara la guerra?’, ForoInternacional 44(4): 635-655.
43









Uyangoda, Jayadeva (2007) ‘Ethnic conflict in Sri Lanka: Changing Dynamics’. Policy
Studies 32. Washington: East-West Center.
Walter, Barbara F. (1997) ‘The critical Barrier to Civil War Settlement’, International Organization 51 (3): 335-364.
Walter, Barbara F. (2004) ‘Does Conflict Beget Conflict? Explaining Recurring Civil
War’,Journal of Peace Research 41 (3): 371–388.
Walter, Barbara (2009) ‘Bargaining Failures and Civil War’ Annual Review Polit. Sci. 12:
243-261.
Wolfovitz, Paul and O'Hanlon, Michael ( 2011) ‘Plan Afghanistan Why the Colombia
model -- even if it means drug war and armed rebellion -- is the best chance for U.S. success in Central Asia’, Foreign Policy Journal. Accessed 29th of October 2011
<http://www.foreignpolicy.com/articles/2011/10/27/plan_afghanistan_colombia>
Wood, E. (2003) ‘Modeling robust Settlements to Civil War: Indivisible Stakes and Distributional Compromises’. Working paper. Santa Fe: Santa Fe Institute.
Zartman, W. and Aurik, J. 1991 ‘Power Strategies in De-escalation’ in S.J. Thorson & S.J.
Syracuse (eds) Timing the De-Escalation of International Conflicts. Syracuse, NY: Syracuse University Press.
Zartman, William (1995) ‘Elusive Peace: Negotiating an End to Civil Wars’. Washington:
Brookings Institution Press.
Zartman, William and Arnson, Cynthia (eds)(2005) Rethinking the economics of War. The intersection of need,creed and greed. Woodrow Wilson centre press. Baltimore.
44