Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?

’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

BEYOND GREED OR GRIEVANCE THEORY – WHAT EXPLAINS CIVIL WAR?

Mohammad Zahidul Islam Khan*

Abstract
The near absence of inter-state conflicts and a parallel increase in the incidence of intra-state
conflicts/civil wars reflects that ‘‘orange is the new black’’ in the context of 21st century history of
warfare. Amidst such fundamental changes, Paul Collier and his team offer the greed or grievance
theory – a pioneering quantitative research to explain the Byzantine complexities of the risks and
processes of civil war onset. The most digested model of the theory examines global data on civil
war against three empirical proxies for greed and four for grievance claiming that the material
motivation (i.e. greed) holds more explanatory power than ideational motivation (i.e. grievance)
in the context of civil war onset. This paper critically examines their claim, comparing and
contrasting it with other relevant theories of civil war. It argues that the primacy of economic
motivation in civil war does not necessarily imply that the notion of greed and grievance to be
juxtaposed; instead, both greed and grievance remain inherently indivisible in civil war. The
interplay of greed and grievance in civil war is dynamic and reflects a symbiotic relationship.
They are often the shades of same problem and can mutate into one another such as into political
greed and economic grievance. The paper substantiates its arguments by highlighting the
increasing trend of internationalized civil conflicts where various external actors, exploiting the
regional conflict complex and the opportunity structures, can contribute to trigger and or prolong
civil wars. Finally, the paper highlights the issues of power and wealth distribution in society and
argues that inequality plays a central role in conflict as postulated in the horizontal inequalities
(HIs) theory. It contends that the inclusiveness offered in the HIs model to account for group
inequalities in economic, political, cultural and social dimensions incorporating both greed and
grievance makes this model better poised to explain the incidence of civil war onset.

Key words: civil war, intra-state war, greed and grievance, horizontal inequality

Introduction
The near absence of inter-state wars and a parallel increase in the incidence of intra-state civil wars
nearly implies that ‘‘orange is the new black’’ in the context of 21th century history of warfare.
Despite an overall decline, recent trend in civil wars around the world indicates a 21 per cent
increase compared to the beginning of this decade. According to the Uppsala Conflict Data
Program (UCDP), out of 40 active conflicts recoded in the year 2014 in 27 locations worldwide,
only one was intra-state 1 while the remaining were within states and 13 of them were

*
1
The only one active inter-state conflict in 2014 was between India and Pakistan. It led to fewer than 50
fatalities hence not meeting the 1000 battle deaths criteria of ‘war’. See There´se Pettersson and Peter
Wallensteen. ‘‘Armed Conflicts, 1946–2014’’. Journal of Peace Research 52(4) (2015):536–550.

1
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

internationalized. 2 The Human Security Report (2012) shows that central and south Asia is
currently world's deadliest region and death tolls from civil conflict have escalated more than
quintupling between 2005-2009. 3 Amidst such alarming trend of civil war occurrences,
researchers have focused on explaining the phenomenon using various theories, data and model.
The ‘‘greed or grievance’’ theory is one such pioneering attempt to achieve a level of
generalization about the civil war onset. Centred around the primacy of economic rationality, this
theory, introduced by Paul Collier and his team is one of the most cited and debated interpretive
framework in contemporary civil war literature not least because it presupposes greed (i.e. material
motivations) having more explanatory power than grievance (i.e. ideational motivation). Stressing
such primacy, the theory argues that ‘‘where rebellion is feasible, it will occur: motivation is
indeterminate, being supplied by whatever agenda happens to be adopted by the first social
entrepreneur to occupy the viable niche.’’4 Through a series of intriguing quantitative assessment,
Collier’s theory gained much prominence in the midst of ever-increasing trend of civil war and the
urgency of understanding its risks and processes.

Against such compelling empirical arguments, this paper critically examines the greed or
grievance theory probing beyond the juxtaposed proposition. To set the context, the paper first
outline the importance of examining civil war amidst its increasing trend. Contrasting the
conventional view that world is getting more peaceful, it contends that the decline in inter-state
wars is no panacea for a world peace, hence knowing what causes civil conflict remains important.
Second, it, outlines Collier's successive versions of the theory charting its variations and critical
claims to contrast it with different schools of thoughts involving socio-economic, political and
resource predation literature of civil war. It demonstrates that explaining the 'Byzantine

2
The internationalized conflicts include: Afghanistan, Azerbaijan (Nagorno Karabakh), Iraq, Mali, Nigeria,
Somalia, South Sudan, Uganda, Ukraine (Donetsk),Ukraine (Lugansk), Ukraine (Novorossiya), USA (the
conflict with Al-Qaeda), and Yemen. See Pettersson and Wallensteen, ‘‘Armed Conflicts’’, 537.
3
See. Simon Fraser University, Sexual Violence, Education, and War: Beyond the Mainstream Narrative.
Vancouver: Human Security Press, 2012. Print. Human Security Report p.156.
4
See the third and final version of the theory, Paul Collier, Anke Hoeffler, and Dominic Rohner.
2006. ‘‘Beyond Greed and Grievance: Feasibility and Civil War’’. CSAE WPS/2006-10.
https://www.csae.ox.ac.uk/workingpapers/pdfs/2006-10text.pdf.

2
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

complexities'5 of civil war is not limited to the putative economic motivation derived through a
number crunching exercises of tenuously linked proxy variables as postulated by Paul Collier and
his team. Third, the proxy variables and statistical method used in the model are examined and
compared with similar quantitative studies on civil war to identify their inadequacies. Fourth,
different leading schools of thoughts is discussed leading to the discussion of Frances Stewart’s
Horizontal Inequalities (HI) theory as a more comprehensive model to explain civil war. It argues
that the primacy of economic motivation in civil war does not necessarily impel the notion of greed
or grievance to be juxtaposed, instead, both greed and grievance bears a symbiotic relationship
and remains inherently indivisible in civil war onset.

Civil wars: ‘orange is the new black’?

Warfare in the 21st century has fundamentally changed. One visible trend is the overall decline of
the war – particularly interstate war, convincing many that the world is becoming more peaceful.
The most cited (and debated) contemporary work depicting this trend comes from Steven Pinker’s
acclaimed book Better Angles of Our Nature. Taking a historical sweep of 12 plus millennia,
Pinker informs us of six long-term trend of declining warfare, homicides and a wide variety of
forms of violence. 6 According to Pinker, in our quest to retreat from violence, we are now in the era of
‘‘New Peace’’ the fifth trend, that started since the end of cold war in 1989. We arrived at this era following
the ‘‘Long Peace’’ that ushered after the end of World War II, when the great powers have stopped
fighting and there was an overall decline of inter-state wars. Overlapping these two trends is the
‘‘Rights Revolution’’ starting with the Universal Declaration of Human Rights in 1948 and more recent
acknowledgements of gay, lesbian, animal, child and human rights issue.

5
Cited in Stathis N. Kalyvas, The Ontology of “Political Violence”: Action and Identity in Civil Wars.
Perspectives on Politics, 1(2003): 479.
6
The six long term trend outlined by Steven Pinker are: 1. “Pacification Process”—the uneven transitions
from anarchic hunter-gatherer, horticultural and other early human societies to the first agricultural
civilizations and then nation-states. 2. “Civilizing Process” that runs from the late Middle Ages to the
twentieth century and was accompanied by the growth and consolidation of the nation-state system in
Europe. 3. “Humanitarian Revolution” anchored in European enlightenment of the seventeenth and
eighteenth century resulting in the decline and abolition of slavery, elimination of judicial and other forms
of torture and a long-term reduction in all manner of other cruel and inhumane practices. 4. ‘‘Long Peace’’
started since the end of World War II in which the great powers have stopped fighting each other. 5. ‘‘New
Peace’’, started since the end of cold war in 1989 because of its tentative nature. Pinker’s sixth trend,
‘‘Rights Revolution’’ overlaps the era of ‘long peace’ and ‘new peace’. See Steven Pinker, Better Angle of
our Nature, A history of Violence and Humanity, (U.S.A.: Penguin Books, 2011) xxiii.

3
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

According to Pinker, the current ear of ‘new peace’ is more tenuous, yet, contrary to popular believe,
armed violence such as civil wars, genocide, repression, terrorism has declined – a claim contested
by many. Critics have argued against the measures used by Pinker to arrive at his conclusions.
Pinker asserts that the most appropriate metric for estimating the deadliness of wars is not the
absolute number of fatalities but the number of war deaths relative to the size of the population.
From this perspective, a conflict that kills 10,000 people in a society with a population of 100,000
is 10 times deadlier than one that kills 10,000 people in a society of a million people even though
the numbers killed are identical. Although, Pinker does not dispute the fact that World War II
almost certainly killed more people than any other war in history, the methods used by him makes
it only the ninth-deadliest episode of sustained violence in human history. 7

Taking a departure from the conventional criticism of the Better Angles, I posit a different
argument as to why the view that ‘‘world is more peaceful today then before’’ is fallacious in
general requiring us to focus more vigorously on incidences of civil conflicts/war in particular.
Almost all accounts of projecting the world as a more peaceful one relies on accounting the number
of battle related deaths, incidence of war/civil wars/conflicts etc. Such measures account for the
manifested ‘static’ property of violence – one that resides on ‘actuality’ and disregards the
‘potentiality’ that resides on the dispositional properties of violence.8 Dispositions are intrinsic.
They are independent of the instantiation of other properties, especially of their manifestations.
Dispositions embodies two fundamental characteristics. First, an entity/system can hold both static
and dispositional properties at the same time. There is nothing about the actual behaviour of the
object that is ever necessary to have a dispositional property. Second, dispositional ascriptions
exhibits its characteristic manifestations under some ‘stimulus conditions’: ‘x is violent at t’ is
associated with ‘x would wage war/use weapon of mass destruction/chemical weapons, when faced
with certain stimulus conditions at t’. Such a view point of violence, and particularly organized
violence such as wars require us to probe into more inquisitive comparison of not just the

7
For an enlightened discussion on Pinker’s argument and critics, see Human Security Report 2013,
Decline in the Global Violence, Evidence, Explanation and Contestation, Simon Fraser University
(Vancouver, Canada : Human Security Press, 2013), 17-49.
8
See Sungho Choi, and Michael Fara, "Dispositions", The Stanford Encyclopedia of Philosophy (Spring
2016 Edition), available at, https://plato.stanford.edu/archives/spr2016/entries/dispositions// accessed on
12 December 2016. Also see Micheal Esfield, ed., John Heil Symposium on his Ontological Point of
View, (U.S.A.: Transaction Books, 2006), 79-81.

4
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

‘actuality’ (manifested battle-death) but also the ‘potentially’ (potential battle-deaths) of war
making – that has increased manifold with the advent of nuclear weapons, weapons of mass
destruction (WMD) etc. in this era of ‘‘New Peace’’ (see table 1 below).

Table 1: Estimated Global Nuclear Warhead Inventories:2016

10000
9000
1796
Number of nuclear weapons

8000 1367 Legend

7000 Warehead not in stockpile, awaits dsiarmament.
6000 Warehead assigned to potential use.
Warehead fitted in ballistic missiles and aircraft bases.
5000
4000
4500
3000 4571
2800
2000 2500

1000
300 260 215 140 110 80 8
0
Russia USA France China UK Pakistan India Isreal DPRK
Nuclear Tipped Countries

Source: Arms Control Association, available at,

https://www.armscontrol.org/factsheets/Nuclearweaponswhohaswhat/ accessed on 18 December 2016.

The world today is armed with too many earth-shattering nuclear weapons and all other sorts of
WMD. Thus, while the relative number of actual battle related deaths have declined, the potential
number of battle/conflict related death involving WMD that can happen under a range of ‘stimulus
conditions’ in the era of ‘New Peace’ can be far greater. According to one report, even a regional
war with an exchange of 100 nuclear weapons, apart from the death and destructions, can produce
5,000,000,000 kg of black soot that can block the sunlight lasting for 25 years, temporarily
destroying much of earth’s protective ozone layer and ushering a global nuclear famine. 9 Viewed
this way the happy proposition of ‘‘world becoming much more peaceful’’ loses much of its
ground to be confirmed. The civilizational process that has landed us at the doors of the ‘‘New
Peace’’ is accompanied by humanity's most earth-shattering weapons having more than enough
destructive power to wipe out the human race for good.

9
Michael J. Mills, Owen B. Toon, Julia Lee-Taylor, Alan Robock, Multidecadal global cooling and
unprecedented ozone loss following a regional nuclear conflict, Earth’s Future, 2,(2014), 161–176.

5
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

Second, the tenuous nature of the ‘new peace’ and the blurring distinctions amongst different types
of war informs us that the decline in inter-state wars is no panacea for a world peace. In the context
of civil war, the regional and international players are not always a passive actors resulting in ‘spill
over’ effects but also an ‘active contributors’ and ‘opportunistic interventionists’. Indeed as
conflicts data between 2006-2015 shows, there is an increase in the internationalized civil conflicts
– defined as those armed conflicts between a government and a non-government entities where the
government side, the opposing side, or both sides, received support from other governments that
actively participate in the conflict, are on the rise (see table 2 below).

Table 2: Trend in Type of Conflicts between 2006-2016

120

100
Number of conflicts

80
50

60 41
38 37 38
35 35
33 33
31
40 20
4 7 7 7 13
5 9 9
8
20
28 31 30 30 30 27 29
23 23 26

0 00 0
0 1
0 0
0 0
0 1
0 1
0 0
0 1
0 10
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Year
Extrasystemic armed conflict Inter-state armed conflict
Internal armed conflict Internationalized Intra-state conflicts
Total Conflicts 2006-2015

Source: Author’s compilation from UCDP data sources.

Internationalized intra-state conflict has increased from just 15 percent in 2006 to 40 percent in
2015. At the same time, the self-sustaining internal civil wars have declined from 84 percent in
2006 to 58 percent. This clearly indicates that rebels/insurgents are not necessarily the only conflict
entrepreneurs. A mix of external and internal war entrepreneurs, together with the transnational
terrorist organizations, the political economy of intra-state conflicts, and its regional and global
links to licit and illicit trade and financing structure hints at the real possibilities of inadvertent

6
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

possession and subsequent use of WMD or so called ‘dirty bombs’ in such conflicts. Such a reality
and the potential deadliness of civil wars indeed reflects that ‘‘orange is the new black’’ in the
context of modern warfare. Today, more than ever, the ability of destroy, disrupt or paralyzing a
city, village or vital infrastructure is not limited to a formal state actors but can be carried out by
ideologically motivated and materially empowered non-state actors. Such reality informs us to
take critical look at the civil wars going beyond the polarized view along the greed or grievance
theory juxtaposing the ‘‘loot –seeking’’ and ‘‘justice-seeking’’ motives of rebellion.

Greed or grievance theory of civil war: the primacy of economic motivations

Contrasting the conventional qualitative analysis on civil war by the political scientists, Paul
Collier and Anke Hoeffler first introduced the greed or grievance theory of civil war onset in 1998
using the global data on civil war from the Michigan University, USA. It is the pioneering
quantitative research that analyses the factors that might account for the onset of conflict. Their
initial effort was revised in 2004 and again in 2006 to include omitted data and to make the results
more definitive. The first study of 1998 included 78 large civil conflicts between 1960-99, the
second study involved 54 wars and 688 observations while the core regression of their latest 2006
version is based on 71 wars and over one thousand observations involving 172 countries covering
the period from 1965-2004. 10 Prior to such quantitative research, academics emphasized the
relevance of ‘Tolstoy effect' — i.e. ‘all happy families are alike while every unhappy family is
unhappy in its own way’ in the context of civil war analysis. Hence, fieldwork-driven case studies
on war civil were preferred over quantitative analysis asserting that the later method would lack
capturing ‘‘the kind of nuance and attention to detail’’ 11 that are common in ‘unhappy families’
(i.e. civil wars).

Breaking the barrier, Collier and his team focused on quantitative global data to explain the risks
and processes of civil war. Their most digested model, ‘greed versus grievance’, examines global

10
See Paul Collier and Anke Hoeffler, "On Economic Causes of Civil War", Oxford Economic Papers, 50
(1998): 563-73. Paul Collier and Anke Hoeffler. "Greed and Grievance in Civil War", Oxford Economic
Papers, 56 (2004): 563-95. And Collier, Hoeffler and Rohner, ‘‘Beyond Greed and Grievance’’, 8.
11
See Nicholas Sambanis, “Conclusion: Using Case Studies to Refine and Expand the Theory of Civil
War,” in Understanding Civil War: Evidence and Analysis, eds., Paul Collier and Nicholas Sambanis,
(Washington DC: The World Bank, 2005), 303-329.

7
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

data against three empirical proxies for greed and four for grievance. The proxies for measuring
greed are the level, growth and structure of income and all proved statistically significant in
conflict risk. According to the theory, faster growth reduces conflict risk as it raises the opportunity
cost of joining rebellion. Connecting the low opportunity costs for recruitment in poverty stricken
countries, they assert that poorer counties with low per capita income (proxy for level of income)
are more likely to experience civil war. Although the relationship is non-linear, they find countries
with abundant natural resources (proxy for economic structure) have a higher risk of conflict. Most
controversially their initial findings indicate that social fractionalization measured in terms of
religious and ethnic diversity lowers the risk of conflict. However, in their later version, they
distinguish between 'greed rebellion' from 'grievance rebellion', and suggest that 'ethnic
dominance’ -- i.e. where one ethnic group makes up 45-90 percent of the population, poses higher
risks for civil war. It also contends that the risk of civil war recurrence decline as the duration of
peace is lengthened. However, as highlighted before, the theory has been revised three times. The
initial ‘greed versus grievance’ model was renamed as ‘motive versus opportunity’ (2004) and
then to ‘feasibility theory versus motivational theory’ in 2006 with an amended title of the paper
as 'Beyond Greed and Grievance: Feasibility of Civil War' (italic added).

The 2006 version of the model speaks of feasibility theory and motivational theory, the latter being
subdivided into greed and grievance. Subsequently they reject the motivational theory and
concludes, ‘‘where rebellion is feasible it will occur: motivation is indeterminate, being supplied
by whatever agenda happens to be adopted by the first social entrepreneur to occupy the viable
niche.’’ 12
The primacy of the material dimension as opposed to any motivational factors is
premised on their argument that regardless of its agenda or appeal, the establishment of a rebel
army is both prohibitively expensive and extremely dangerous in civil wars. Thus, the relatively
rare circumstances in which rebellion is materially feasible constitute an important part of any
explanation of civil war. Regarding the other variables, they now conclude that linguistic and
religious fractionalization significantly increases the risks of civil war and adds that mountain
terrain and proportion of males in 15-29 age groups adds significant risks while large diasporas
contributes to reduced risks of civil war . Regarding their earlier assertion on civil war recurrence

12
Collier, Hoeffler and Rohner, ‘‘Beyond Greed and Grievance,’’20

8
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

they now predict a much lower 23 percent in first four years and 40 percent in ten years’ time
frame. However, regardless of the changes, their claim that their ‘‘core findings have survived’’13
that keeps the debate on.

Can ‘greed’ alone explain civil war?

Despite several refinements, the theory remains highly contested if not problematic on various
accounts. First, the theory models rebellion as ‘an industry that generates profit from looting’ and
focuses on the predatory activities of the rebels, typifying the large scale organized violence by
the rebel army as the ‘defining feature’ of civil war. 14 Such focus may yield partial view
categorizing all insurgencies as an extreme form of common criminality. It also restrict the analysis
of several conflict-promoting acts by the internal or external actors, political economy and most
importantly the opportunity structures that may prevail in a particular context. 15 Arguably, in
Kosovo, Nepal, the Democratic Republic of Congo (DRC) and Sierra Leone, the severely
diminished military capacity and legitimacy of the state, internal decay and geopolitical shift
hastened the civil war.16 The creditors, donors and contractors to the weak states could also trigger
and contribute in creating shadow states. For example, the creation of sobel (soldiers by day and
rebels by night), in Sierra Leone was triggered by the financially pressed government’s decision
to trim one third of state employees allowing the armed soldiers to engage in "Operation Pay
Yourself".17 Such a project resulted in over 20 percent soldiers of the woefully under resourced

13
Ibid, 27.
14
Collier and Hoeffler, ‘‘Greed and Grievance’’, 3 and Ibid, 4.
15
For the economic dimensions of civil was that goes beyond ‘greed’, see Karen Ballentine, ‘‘Beyond
Greed and Grievance: Reconsidering the Economic Dynamics of Armed Conflict,’’ in The Political
Economy of Armed Conflict - Beyond Greed and Grievance, eds. Karen Ballentine and Jake Sherman,
(London: Lynne Rienner, 2003), 259-285.
16
. For country specific analysis on Nepal, see John Bray, Leiv Lunde and Mansoob S Murshed, ‘‘Nepal:
Economic Drivers of the Maoist Insurgency’’ in Ballentine and Jake, eds., ‘‘Political Economy of Armed
Conflict’’107-132. For Kosovo, see Yannis Alexandros ‘‘Kosovo: The Political Economy of Conflict and
Peacebuilding’’, in Ballentine and Sherman, eds., ‘‘Political Economy of Armed Conflict’’, 167-196. For
DRC and Sierra Leone see Michael L. Ross, (2003) ‘‘Oil, Drug and Diamonds: The Varying Roles of
Natural Resources in Civil War’’, in Ballentine and Sherman, eds., ‘‘Political Economy of Armed
Conflict’’, 47-72.
17
Ballentine and Sherman, eds., ‘‘Political Economy of Armed Conflict’’, 90. Also see Michael Pugh, Neil
Cooper and Jonathan Goodhand, War Economies in Regional Context Challenges of
Transformation,(London: Lynne Rienner, 2004), 17-45.

9
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

Sierra Leone military to engage in freelance armed activities like killing civilians, extorting money
and engaging in illegal diamond trading like the rebels.18 The coinciding economic interest and
incentives of the Sierra Leon's military and the rebels under a ‘regional conflict complex’ and
‘opportunity structure’ prolonged the war. 19 More so, when the state itself is criminalized,
corruption and rent seeking by the elite undermines government's legitimacy and eventually
transform the state being ‘‘above politics and into the prize of political competition.’’
Consequently the creation of ‘shadow states’ where the government's misrule combined with the
linkage of well-functioning state machineries with the international and regional clients - both licit
and illicit, makes the state more attractive for armed contestation. 20 Regrettably, the defining role
of such conflict ‘complexes’ and structures, and their corresponding impact on social capital and
breeding inequality, that may trigger and contribute to civil war onset do not feature prominently
in the greed or grievance model. 21

Second, the model views rebels as a unitary actor with an interest of predation. Such a view is
inadequate to define the distinctive nature and actions of the core rebel leadership and the
peripheral activities by the followers in civil war. As a result it undermines and/or overlooks the
possibility of capturing the interactions of the economic (or other) incentives and rationales that
are at play amongst the central leadership (i.e. master cleavage) and at the peripheral followers
(i.e. local cleavage). Indeed as Ballentine and Sherman argues, while followers on all sides may

18
The quality and discipline of the Sierra Leone army declined after 1991, when the force was dramatically
expanded, from 3,000 to 16,000, in order to intensify the war against the RUF. However, the government
was unable to pay for such a large army. See Robert L. Feldman and Michel Ben Arrous, ‘‘Fighting irregular
Fighters Confronting Africa's Sobels’’ Parameters, 43(4), Winter 2013-14, 67-75.
19
For an excellent account of the economic network of regional conflict complex with illustration of the
Colton supply chain see Pugh, Cooper and Goodhand, ‘‘War Economies’’, 26-28; while Ballentine,
‘‘Beyond Greed and Grievance’’, 264-268, illustrates opportunity structure.
20
See William Reno, ‘‘Shadow States and the Political Economy of Civil Wars’’, in Greed and Grievance
Economic Agendas in Civil War, eds. Mats Berdal and David M. Malone, (London: Lynne Rienner, 2000),
43-65. And Dietrich Jung, ed., Shadow Globalization, Ethnic Conflict and New Wars a Political Economy
of intra-state war, (London: Routledge, 2003) 9-27.
21
Jean-Francois Bayart identifies five symptoms in the context of sub-Saharan African states and also
highlights the lack of social capital. See Jean-Francois Bayart, Stephen Ellis, and Beatrice Hibou, The
Criminalization of State in Africa, (Bloomington: Indiana University Press, 1999), 4-8 & 69-114. William
and Picarelli highlights the organized crime dimension. See Phil Williams and John T. Picarelli,
‘‘Combating Organized Crime in Armed Conflict’’ in: Karen Ballentine and Nitzschke Heiko, ed., Profiting
from Peace Managing the Resource Dimension of Civil War, (London: Lynne Rienner, 2005), 127.

10
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

be attracted by the prospect of material gain, they may also be attracted by ideological zeal, ethnic
loyalty, or aspirations to power. 22 Tracing the ontology of political violence, Stathis Kalyvas
highlights this issue more vividly: 23

‘‘…civil wars are not binary conflicts, but complex and ambiguous processes that foster
the “joint” action of local and supra-local actors, civilians, and armies, whose alliance
results in violence that aggregates yet still reflects their diverse goals.’’

According to Kalyvas, ‘greed’ is anchored in Hobbesian inspiration24 (of competition, diffidence

and glory) and hence reflects the private sphere; while the ‘grievance’ dimension is best captured
from a Schmittian lens – that entails an ontology of civil wars based on abstract group loyalties
and beliefs, whereby ‘‘the political enemy becomes a private adversary only by virtue of a prior
collective and impersonal enmity’’. 25 Contrasting the Hobbesian view, the Schmittian perspective
stresses the fundamental political nature of civil war and its attendant processes offering better
explanatory leverage to interpret its different variants including ethnic civil war that stress strong
beliefs, group enmity, and cultural antipathy. Rejecting Collier and other’s juxtapose dichotomy,
Kalyvas points to the interactions of political and private identities and actions. Exposing the
disjunction between the ‘centre’ and ‘periphery, Kalyvas suggests that the actions on the ground
are often more local and private. Interestingly, Collier et al. recognizes such disjunction and yet
rejects the narratives of grievances saying ‘‘grievance is to rebel organization what image is to a
business’’. 26 Drawing rationality from the Machiavelli Theorem, ‘‘no profitable opportunity for
violence would go unused’’ they characterize the (false) sense of grievance as a tool to generate
support and facilitate recruitment to sustain rebellion. Citing a controversial parallel from the
Marxist theory, Collier and authors assert that ‘‘even where the rationale at the top of the
organization is essentially greed, the actual discourse may be entirely dominated by grievance.’’27

22
Ballentine and Sherman, ‘‘Political Economy of Armed Conflict’’, 8.
23
Kalyvas, ‘‘Ontology of “Political Violence’’, 474.
24
Hobbes outlines three principal causes of quarrel: competition, diffidence and glory. He argues that
human nature of 'competition' 'maketh men invade for gain' using violence to make themselves 'master of
other men's persons, wives, children, and cattle;' Thomas Hobbes, , ‘‘The Leviathan’’, 1660, available at .
http://www.ttu.ee/public/m/mart-murdvee/EconPsy/6/Hobbes_Thomas_1660_The_Leviathan.pdf/,
accessed on 12 November 2016.
25
Kalyvas, ‘‘Ontology of “Political Violence’’, 475.
26
Collier, Hoeffler and Rohner, ‘‘Beyond Greed and Grievance’’,92.
27
Collier et al (2000), p.92.

11
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

However, Michael Pugh contends that Collier’s greed or grievance model focuses on economic
data and not the ideational processes – that is at the heart of the Marxists theory. As such drawing
parallel with the Marxist theory and contending the primacy of material dimension amounts to
‘‘inverting the meaning of the Marxist concept’’.28 Nevertheless, Collier, responded to his critics
expanding his thoughts on three possible variants of the theory and asserted that 'doing well out of
war' is what matters irrespective of whether rebel's motivation is linked to power-seeking or
grievance and rejects the notion of objective grievance. As he succinctly puts elsewhere.29

‘‘…rebellion is motivated by greed, so that it occurs when rebels can do well out of war. On the
power seeking variant of the predation theory, rebels are motivated by a lust for power, but rebellion
occurs only when rebels can do well out of war. On the subjective grievance variant of the predation
theory, rebels are motivated by grievances, imagined or real, but rebellion occurs only when rebels
can do well out of war.’’ (Italics added)

While ‘doing well out of war’ thesis appears convincing in plain sight, it remains highly contested
when one discovers that ‘well’ is defined only in terms of material gain, rejecting the ideational
aspects of civil war. Such a fallacy encourages us to probe into the proxy variables and data used
in such models.

Limits of the proxy variables, data and model

The variations in the statistical relationship and the corresponding significance level as reported in
different studies on civil war is listed in table 1. 30 As evident, despite the obvious advantages of
quantitative methods to arrive at generalization, the clarity of the statistical methods and the use
of proxy variables in greed or grievance model is theoretically problematic and empirically
unsatisfactory. First, the empirical link of social/ethno-religious fractionalization with civil war

28
Pugh, Cooper and Goodhand, ‘‘War Economies’’, 22
29
Paul Collier, ‘‘Doing Well out of War: An Economic Perspective’’ in Berdal and Malone eds., ‘‘Greed
and Grievance’’, 91-112.
30
Compiled from Collier, Hoeffler and Rohner, ‘‘Beyond Greed and Grievance’’, James D. Fearon and
David D. Laitin, ‘‘Ethnicity, Insurgency, and Civil War’’, American Political Science Review, Volume 97,
Issue 1, February 2003, 75-90 and Charles T. Call and John Schmitt, ‘‘Why Peace Fails: Explaining Civil
War Recurrence,’’ Paper presented at the annual conference of International Studies Association, New
York, 2009.

12
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

disregards the political dynamics associated with civil wars. Collier‘s model combines the data of
ethnicity and religion aspect to proxy social fractionalization; ethnic dimension is measured by the
probability that two randomly drawn individuals from a given country do not speak the same
language. For the religious dimension, they use the data from Barrett (1982) and construct an
analogous religious fractionalization index aggregating various religious affiliations into nine
categories: Catholic, Protestant, Muslim, Jew, Hindu, Buddhist, Eastern Religions (other than
Buddhist), Indigenous Religions and no religious affiliation. The final fractionalization indices
range from zero (for completely homogenous society) to 1 (for completely heterogeneous
society).31 Notwithstanding such sophistication the proxy remains a poor indicator to capture the
political dynamics because not all ethno-linguistic/ religious groups matter for civil wars. The
linguistic, religious and cultural diversity among the population is mediated through the internal
frictions that manifest in the hierarchies in either socio-economic terms or in terms of their
visibility in the political arena.32 Data reflecting politically relevant groups is missing from the
model. As reflected in table 1, the Fearon and Laitin’s study does not report any significant relationship
while the 2006 study of Collier reports a highly significant positive relationship between the civil war onset
and social fractionalization. Again the study on civil war recurrence by Call and Schmitt reports a highly
significant but negative relationship.

Second, the per capita GDP and percentage of oil export/primary commodity dependency is
considered as a proxy for limited economic opportunity in Collier’s model while the same is used
as the proxy for state weakness by Fearon and Laitin. Highlights the fallacies of using such
tenuously related proxy, Christopher Cramer (2006) points out that a decline is GDP and primary
commodity dependency could also be a reflection of many factors including failed policy, missing
economic dynamism, introducing the structural adjustment plan (SAP), a probable shortage of
consumer goods and imports and widespread grievance or dissatisfaction. More so, belligerent’s
economic agendas are hard to quantify as the volume of all primary commodity exports (i.e.
agricultural goods) may not show up in national accounts.33 Thus, such creative use of available

31
Collier, Hoeffler and Rohner, ‘’Beyond Greed and Grievance’’, 28.
32
See Andreas Wimmer, Lars-Erik Cederman and Brian Min, ‘‘Ethnic Politics and Armed Conflict: A
Configurational Analysis of a New Global Data Set’’ American Sociological Review, Volume 74, (April
2009), 316 –337.
33
Sambanis ‘‘A review of recent advances and future directions in the quantitative literature on civil
war’’, Defence and Peace Economics, Volume 13(3), (2002), 221.

13
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

cross-national data may not be perfect to confirm or reject the hypothesized causal connection.
Indeed as table 2 below reflects the variations in statistical significance and relationship, arrived
at by different studies using such thin and questionable proxy variables. Contrasting Collier’s
findings, Fearon and Laitin reports that ‘‘neither the share of primary commodity exports in GDP
nor its square is remotely significant when added to the model.’’ 34 Again, reporting on the civil war
recurrence, Call and Schmitt find that GDP (lagged) is not a significant predictor of civil war whie
confirming its negative relationship.35

Table 3: Relationship of proxy variables as reported in different studies of civil war.

Statistical relationship in quantitative studies
Collier, Hoeffler and Fearon and Laitin Call and Schmitt (2009)
Rohner (2006) (2003)
Dependent Variable Civil war onset Civil war onset Civil war Recurrence
Independent variables
Economic Growth, level and structure
GDP per capita growth / Negative and highly Negative and highly Negative but not
Per capita income significant (at 1%) significant (p < .001) significant.
Share of oil export - Positive and Positive and significant
significant (p<0.01) (at 1%)
Share of primary Positive and significant Not significant. -
commodity exporter (at 10%)
Demography
Log of total population - Positive and highly
significant (p<0.001)
Proportion of young men Positive and highly Positive but highly Positive and highly
in population significant (at 1% insignificant (p=.21) significant (at 1%
level) level)
Terrain condition
Mountainous terrain Positive and significant Positive and significant -
(at 5%) (p<0.01)
Noncontagious territory - Positive but Negative but
insignificant insignificant.
History
Prior instability/war Negative but Negative but Positive and highly
insignificant. significant (p<0.01). significant (at 1%)
Former French colony Negative but Not measured Not measured
insignificant
New state Positive and highly Positive and highly Not applicable.
significant (at 1%) significant (p<0.001)
Social

34
Fearon and Laitin, ‘Ethnicity, Insurgency, and Civil War’’, 78.
35
See table 2.1 in Call, ‘‘Why Peace Fails’’ and Charles T. Call and John Schmitt, ‘‘Why Peace Fails:
Explaining civil war recurrence’’

14
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

Ethnic/social/religious Positive, highly Positive but not Negative and highly

fractionalization significant (at 1% l) significant significant (at 1%)
Democracy Positive but Positive but Negative but
insignificant. insignificant. insignificant.
Sources: Author’s compilation from the mentioned studies.
Third, the blurring distinctions between civil war ‘onset’ and ‘recurrence’ and the coding of ‘war’
and ‘conflicts’ also results into considerable anomalies. A ‘conflict’ is defined as producing more
than 25 battle deaths in a year while the same is coded as ‘war’ when the battle deaths reaches
1000/year. As a result only a minor adjustment in the coding rules can have quite dramatic effects.
For example, the 2006 version of Collier’s model starts with those observations that has reached
the 1000 battle death ending in the year when the armed conflict did not generate any deaths. What
if a conflict has generated only 999 battle death in the first year? Such coding also poses the danger
of merging civil war ‘onset’ data with ‘recurrence’. A clear distinction between ‘onset’ and
‘recurrence’ is particularly important as each can produce divergent statistical relationship. Call
points out that Colliers’s study between 2002 -2006 reports remarkably divergent rates of armed
conflict recurrence including 50 percent, 44 percent, 23 percent and 21 percent.36 Such anomalies
argues in favour of exercising caution while accepting the claims made in these studies.

In sum, these statistical models along with their tenuously linked proxy variables not only
marginalizes objective grievance as a potential cause of civil war but also over simplifies the
political economy of civil war. Viewed from the narrow lenses of greed, the quantitative
approaches tend to over aggregate the dependent variable and treat civil wars as though they have
uniform causes. It ignores the fact that civil wars are just not the outcome of everyday encounters
between individuals; they are the result of interactions between the state and ethno political
movements that challenge state authority. Thus, many observers that Collier's statistical model has
not generated convincing regularities. They highlight that the Gini coefficient used in the model
captures vertical inequality as opposed to horizontal inequalities telling us very little about the
actual condition that may trigger rebellion. Indeed comparison of such data across a range of
countries is problematic. For example, Rwanda had low Gini coefficient prior to the war and
genocide. However, Collier et al. accepts that any quantitative analysis based on global data sets

36
Charles T Call, ‘Why Peace Fails’’, 61.
15
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

could have severe limitations and suggests that the findings arising out of such quantitative
research be seen as ‘'complementing qualitative in-country research rather than supplanting it’’.

What explains civil war?

Notwithstanding its intellectual rigor and methodology, the ‘greed or grievance’ model is neither
absolute nor most obvious. The model is also at odds with other economic approaches to civil war
that views resource wars not only as a function of predation but also as a distributional conflict,
poor governance or as a response to the structural adjustment. The unjust distribution of revenue,
economic vulnerability, weak governance structure, inefficiency, corruption, institutional
conservatism, disruptive social impacts, subsidization of politicized schemes, leadership etc. can
foment greed and generate grievance, creating conditions for civil war in countries endowed with
abundant natural resources.37 The persistent diversion of state assets by neo-patrimonial rulers in
'shadow states' and their subsequent fragmentation can breed 'war entrepreneurs' and creates
condition for civil war. Such internal decay coupled with global and regional connections is a
reality but does not feature in Collier's model. Thus, it is not surprising to see that contrary to the
Collier's contentions, Lebanon and Indonesia (Aceh province) experienced civil wars at a time
when both the countries were having high growth. In case of Aceh, the expansion of extractive
resource industry flocked migrant workers leadings to a second order effect in massive land
seizures which reinvigorated the Acehanese rebel movement coinciding with the high growth
period. 38 However, Kenya, despite a weak economy and low GDP per capita, has managed to
avoid civil war arguably through a strong authoritarian rule. Again Papua New Guinea with strong
income from primary commodity export (over 40 percent of GDP) for two decades could not avert
the rebellion. All these suggest that state capacity may be less of a function of its resource base
than its ability to manage the resources effectively. 39

37
As Le Billon contends that ‘‘even the well governed countries have found it difficult to break out of
resource dependence when faced with unfavourable geography and a legacy of resource based public
welfare’’. See Philip Le Billon, Wars of Plunder- Conflicts, Profits and Politics of Resources, (New York:
Columbia University Press, 2012), 9-33 and 224-225. Also see M. Humphreys, ‘‘Natural Resources and
Armed Conflict: Issues and Options’’, in Ballentine and Nitzschke, eds., ‘‘Profiting from Peace’’, 25-41;
and Jeffry Herbst, States and Power in Africa. (Princeton U.S.A.: Princeton University Press, 2000),
38
Sambanis, ‘‘Conclusion: Using Case Studies’’, 306
39
Ballentine, 2003, p.265.

16
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

Joel Migdal (1980) and K.J. Holsti (1996), are the two early proponents who attempts to links
intra-state war with state weakness in general and the legitimacy dimension – i.e. the degree of
state-society cohesion, in particular. According to Holsti, a deficiency in vertical and horizontal
40
legitimacy reflecting weak state-society cohesion is the main cause of civil war. Vertical
legitimacy is the belief that a select group of individuals and organizations (such as state actors)
have the right (i.e., it is appropriate) to act as representatives of society. Horizontal legitimacy is
the belief that all people (or peoples) within the boundaries of the nation-state are acceptable
partners in the collective endeavour fundamental to the political, economic, social, and cultural
health of a society. Thus lacking such legitimacy fragments state-society relationship. As a result,
even when a state is very powerful in its ability to command over the instruments of social
surveillance, coercion and terror, it remain fragile and can slip into the abyss of civil war. Evidence
supporting this hypothesis is overwhelming. Many post-colonial states invested heavily on security
sectors to achieve better social control with the assistance of foreign donors and governments. Yet,
the rapid growth of the security forces in these post-colonial countries failed to achieve the degree
of social control and were affected by coup incidences, political repressions often facilitated or
triggered by external actors.41 Indeed, as Powell and Thyne in their study on global incidence of
coup reveals that within two decades of their entry onto the world stage, the newly created Third
World states experienced a total of about 200 coup incidents which had a lasting impact on state-
42
building. Between 1950-2015, Africa (42.3 percent) and the Americas (30.9 percent)
experienced the most coups, while Asia and the Middle East accounted for 13.1 percent and 10.14

40
See K. J. Holsti, The State, War, and the State of War, (Cambridge, UK: Cambridge University Press,
1996), 84, 90, Chapter 5. Also see K. J. Holsti, "International Relations Theory and Domestic War in the
Third World: The Limits of Relevance," in International Relations and the Third World, ed., Stephanie G.
Newman (New York: St. Martin's Press, 1998), 113-114, 119-122. Joel Migdal, Strong Societies and Weak
States: State-Society Relations and State Capabilities in the Third World, (USA: Princeton: Princeton
University Press, 1988).
41
See Joel Migdal, State in Society: Studying How States and Societies Transform and Constitute One
Another, (Cambridge: Cambridge University Press, 2001), And Jeffry Herbst, (2004), ‘‘Let them Fail :
State Failure in Theory and Practice’’, in When States Fail: Causes and Consequences, ed., Robert I.
Rotberg, (Princeton: Princeton University Press, 2004), 301 – 316. Also see Mohammad Z.I. Khan,
‘Paradox of State Fragility: The Role of International Structures in the context of Bangladesh, BIPSS
Monograph, available at, http://www.bipss.org.bd/images/pdf/monograph/monograph_200614.pdf/ ,
accessed on 12 December 2016.
42
See J.M Powell and C. L. Thyne, ‘‘Global instances of coups from 1950 to 2010: A new dataset’’, Journal
of Peace Research 48(2) (2011), 249–259. And Mohammad Z.I. Khan, ‘‘The Fallacies of State Fragility
Indices, Is there a Fragility Trap?’’, Upcoming BIISS Paper, 14-15.

17
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

percent coup attempts respectively. In contrast, Europe had the fewest (3.6 percent) coup attempts.
Such prevalence of coup incidences in weak states hindered developing strong state-society
relationship, weakening government’s legitimacy and ultimately contributing towards the
incidence of civil war.

Michael E. Brown and Kristian S Metternich and others highlights the role of regional and
international factors for the initial outbreak of civil wars and the escalation dynamics of conflict
once underway.43 In a similar vein, Christopher Cramer (2006) compares and contrasts the liberal
interpretation with the romantic utopian view of violence to forge an alternative view of
understanding civil war as a product of the transition to capitalism and of global economy. 44
Highlighting the need to contextualise civil wars, Nicholas Sambanis argues the need for case
study based analysis dispensing the large N analysis that tends to focus on generalization missing
the nuance aspects.45 ior However, Collier’s model fails to capture such international dimension
of civil war.

Some suggest that the onset of civil war and its link to the natural resource predation can be
explained by examining the ‘lootability’ of the resources. For example, Le Billon distinguishes
natural resources based on their proximity from the centre of power in both spatial and political
terms (i.e. proximate and distant) and also in terms of their physical characteristics and socio-
economic linkage (i.e. point and diffused). Projecting his classification with the help of a matrix,
Le Billon illustrate the type of conflicts associated with each set of resource categories. He
concludes that 'diffused' and 'distant' natural resources like gems and diamond are associated with
warlordism while 'point' and 'proximate' resources like oil, gas typically generates coup-d'état.
Conversely 'proximate' but 'diffused' resources like coffee generate peasant/mass rebellion and the
'distant' but 'point' resources generate secessionists’ movement. Such typology and classifications

43
See Michael E. Brown, (1996). International Dimensional of Internal Conflict. (Cambridge, MA: MIT
Press, 1996), 31-36; and Nils W. Metternich, Kristian S. Gleditsch, Han Dorussen and Andrea Ruggeri
‘‘International Dimensions of Internal Conflict’’, in Graham K. Brown and Arnim Langer, eds. Elgar
Handbook of Civil War and Fragile States. (U.S.A.: Edward Elgar Publishing, 2012), 214-222.
44
See Christopher Cramer, Civil War is Not a Stupid Thing: Accounting for Violence in Developing
Countries. (London: Hurst & Company, 2006), 7-8.
45
See Sambanis, “Conclusion: Using Case Studies’’, 303-329.

18
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

surely adds to our clarity of understanding on civil war but does not go further to facilitate our
understanding about the contextual and ideational factors.

However, the most formidable challenge to Collier’s model perhaps comes from the theories that
connects inequality and exclusionary behaviours as prime source of grievance causing civil war.
Notable early efforts along this line includes Ted Robert Gurr's work that explains civil war based
on relative deprivation theory. 46 Analysing the behaviour of 275 politically active ethnic groups
during the 1990s, Gurr lists four main determinant of civil war: ethno cultural identity, level of
actual or perceived grievance, mobilizing capacity of the ethno political groups and the available
opportunities for political actions by each group. His views are consistent with constructivists
approach as it focuses on the political mobilization of elites by political entrepreneurs and the
social construction of identity that is often used as a mobilization device. Charles Tilly in his book
Durable Inequality offers another prominent model rooted on inequality. 47 Tilly highlights people
and societies form ‘categorical pairs’ by drawing boundaries between ‘us’ and ‘them’. The
institutionalisation and sedimentation of these paired categories generates ‘‘durable inequality.’’
The diffusion and institutionalisation of such inequality takes place over time as modes of
exploitation and opportunity hoarding are extended, copied, and adapted, leading to violent
conflicts. Focusing on the political instability and regime transitions, Håvard Hegre and others
models civil war as the result of opportunity for violence as opposed to opportunity for predation
to settle prior grievances.48 They argue that the probability of civil war is a function of a host of
factors that reduce the net expected costs of political violence and generate an expectation that
violence will resolve prior grievances. A similar conclusion is drawn by Call (why peace fails) as
he reports that the most important factor that plays a common causal role in civil war recurrences
is political exclusion and not the economic and social factors.49

46
See Ted Robert Gurr, Why Men Rebel. (Princeton, NJ: Princeton University Press 1970), 24-32. And
Ted Robert Gurr, People Versus States: Minorities at Risk in the New Century, (Washington D.C.: Institute
of Peace Press, 2000).
47
See, Charles Tilly, Durable Inequality (Berkeley, CA : University of California Press,1998),
48
Håvard Hegre, Tanja Ellingsen, Scott Gates, Nils Petter Gleditsch, ‘‘Toward a Democratic Civil Peace?
Democracy, Political Change, and Civil War, 1816-1992’’, American Political Science Review, Volume
95(1) (Mar 2001), 33 – 48.
49
Call, ‘‘Why Peace Fails’’, 4.

19
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

In sum, all these models informs us the common causal role of inequality in civil wars. The
centrality of inequality in human convulsions was succinctly put by Alexis De Tocqueville, nearly
181 years ago: ‘‘Remove the secondary causes that have produced the great convulsions of the
world and you will almost always find the principle of inequality at the bottom.’’ 50 However, as
noted before, greed or grievance model finds (vertical) inequality ‘insignificant’ to explain civil
war. Thus these theories are at odds and challenges the primacy of economic dimension in civil
wars as espoused by Collier’s model. Frances Stewart’s horizontal inequalities (HIs) model,
introduces in 2008 reconciles these differences as it goes beyond the traditional methods of
capturing inequality. Stewart’s model is built on Gurr and Tilly’s work focusing on group
inequality and extends to link civil war with state weakness and authoritarian repressions offering
rich and diverse analytical tools to examine civil wars that we examine next. 51.

Horizontal inequalities (HIs) and civil war

HIs model argue that significant presence of HIs contributes to civil war. The root of the model
can be traced to Gurr’s relative deprivation and Tilly’s categorical inequalities theory. Combining
these two, HIs intimately relates to the issues of power and wealth distribution in the societies. At
the heart of the HI model is the principles of inequality manifested in the political, economic, social
and cultural status of different groups (figure 4).

50
Alexis De Tocqueville, Democracy in America, trans., Herny Reeve, (New York, Bantam Classics),
chapter XXI.
51
Frances Stewart, ‘‘Horizontal Inequalities as a Cause of Conflict A review of CRISE Findings’’, available
at, https://assets.publishing.service.gov.uk/media/57a08b0e40f0b64974000936/CRISE-Overview-1.pdf/ ,
accessed on 12 December 2016. Frances Stewart, ‘‘Horizontal inequalities as a cause of conflict’’, Bradford
Development Lecture 2009, available at http://www.bradford.ac.uk/social-
sciences/media/socialsciences/BDLStewart.pdf/, Accessed on 12 December 2016; Frances Stewart,
‘‘Horizontal Inequalities and Conflict: An Introduction and some Hypotheses’’. In: Frances Stewart, ed.,
Horizontal Inequalities and Conflict: Understanding Group Violence in Multi-ethnic Societies, (New York:
Palgrave Macmillan, 2008), 3-23. Frances Stewart and Graham Brown, ‘‘Fragile States’’, available at
http://www3.qeh.ox.ac.uk/pdf/crisewps/workingpaper51.pdf/ , accessed on 12 December 2016

20
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

Figure 4: Dimensions of horizontal inequalities (His)

Economic
entitlement of Political
Assets, participation,
incomes and opportunities
employment and power

Cultural status
in terms of
Social access
languages,
and situation
customs,
norms

However, defining group boundaries is always challenging if not impossible since identities are
malleable, fluid, multiple and may even be endogenous. However, many have argued that group
identity intensifies in times of social, economic and political crisis. In the context of civil war in
Bosnia Herzegovina, the influential Canadian political scientist Janice Gross Stein reports:
‘‘Muslims in Bosnia Herzegovina identified themselves as Serbs or Croats until it was no more
possible.’’ 52 Similarly, in the context of Chittagong Hill Tracts insurgency in Bangladesh, Eva
Gerharz stresses the importance of political power and socio-economic entitlement concluding that
‘‘Being a Chakma, Marma, Mro or Lushai is (thus) more important in people’s everyday lives than
collectivising notions.’’ 53 Contrasting these examples is the Orang Asli, an active indigenous
group in Malaysia, which was developed as a group out of at least eighteen different identities to
help people mobilise, first against the British and subsequently for their rights in independent

52
See, Janice Gross Stein, Image Identity and Conflict Resolution, in Managing Global Chaos, Sources of
and Responses to International Conflict, eds., Chester A. Crocker, Fen Osler Hamson and Pamela Aall,
(US Institute of Peace, Washington DC, 1998), 93-107.
53
Eva Gerharz, ‘‘What is in a Name? Indigenous Identity and the Politics of Denial in Bangladesh’’,
available at, http://edoc.hu-berlin.de/suedasien/band-4/115/PDF/115.pdf/ , accessed on 15 December 2016.

21
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

Malaysia.54 Similarly, the struggle for liberation of Bangladesh was also a mix of different groups
and ethnicity who felt indifferent with the Pakistani oppression. This raises the question as to what
should be a valid measure to draw the group boundaries. Addressing this problem, HIs model uses
the ‘‘felt differences’’ criteria arrived at variety of survey methods. It is argued that ‘feeling
different’ is ‘‘important enough and clear enough in many societies to make it possible to measure
group performance.’’55

Despite its inheritance from Gurr and Tilly’s work, HIs theory differs from them on two important
areas. First, HIs theory presupposes the notion that relative rich or poor could both initiate conflict.
As a result it provides a framework to explain why rich Tutsi's attacked the poor Hutu's in Burundi
in 1993 or why Côte d’Ivoire remained peaceful under the politically inclusive regime of Félix
Houphouët-Boigny and yet degenerated into chaos after his death in 1993 under successive
regimes who adopted a policy that barred and curtailed the opportunity of the northerners to
participate in the political process.56 Second, HIs is also viewed as close cousin of social exclusion
(SE) hence one could relate it with Call’s (‘‘why peace fails’’) arguments on civil war recurrence
that focuses on exclusionary behaviours. While defining the size of a ‘‘critical mass’’ required for
the HIs to trigger any civil conflict, Stewart argues that it would require at least 30 percent or more
people subjected to such inequalities. Of course, it would be a matter of judgment, he adds.57

HIs theory stand out from the ‘greed or grievance’ model on several counts. First, it focuses on
culturally defined groups instead of individuals and measures group inequalities in economic,
political, cultural and social dimensions to explain civil war onset. Second, in HI model, the proxy
variables are relatively nuanced, contextual and politically relevant. For example, the economic
dimensions focuses more on the power relationship and entitlement instead of crude economic
indicators such as income level, Gini coefficient. It includes issues like: access and ownership of

54
Frances Stewart, Graham Brown and Luca Mancini, ‘‘Why Horizontal Inequalities Matter: Some
Implications for Measurement,’’ (CRISE Working Paper number 19, June 2005) available at,
http://www3.qeh.ox.ac.uk/pdf/crisewps/workingpaper19.pdf/ , accessed 14 December 2016.
55
Ibid, 8
56
For a detail account of ethnic representation in government of Côte d’Ivoire between 1980 – 2003, see
table 21 in Stewart and Brown, ‘‘Fragile States’’, 68.
57
Stewart and Brown, ‘‘Fragile States’’, 15

22
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

financial, human, natural and social assets as well as income level and employment opportunities.
Social HIs are counted against education, health care and housing while political HIs reflects the
distribution of political opportunities and power among groups and their representation and
participations in government organizations; Finally the cultural status of HIs is measured in terms
of recognition and standing of different language, religion, customs, norms and practices of the
respective groups. Leveraging the country specific Demographic and Health Survey (DHS) data,
researchers have used innovative questionnaires such as economic household assets (whether or
not each household has electricity, a radio, a television, a refrigerator, a bicycle, a motorcycle and/
or a car), educational, vocational and employment opportunities (percentage in public/private,
skilled/unskilled etc.) , infant mortality rates and the possession of children’s health cards etc., to
capture the power and entitlement aspects of HIs present in a society. Such methods allows us to
go beyond capturing the vertical inequalities and provide a nuance understanding of group-identity
and polarization dynamics of civil war and the complex interplay of both greed and grievance.

Based on a combination of hypotheses HIs theory asserts that convergence of political and social
exclusion together with economic deprivation can significantly contribute to the civil conflict
while cultural differences alone do not lead to violent conflict unless there are also major economic
and/or political causes. Analysing global data, Lars-Erik Cederman, Gudrun Østby, Stewart,
Brown and Mancini among others provide substantial evidences including econometric and case
study based analysis supporting HIs theory. 58 Thus, taken together, HIs theory appears to be
multidimensional and inclusive of both greed and grievance and better poised to explain the risk
and process of civil war.

End thoughts
Be that as it may, the interplay of greed and grievance in civil war is dynamic and shares a
symbiotic relationship. They are often 'shades of same problem' as we see more of ‘‘political greed
and economic grievance.’’59 The potential of organized crime, rent seeking attitude by the state

58
Lars-Erik Cederman, Niles B. Weidmann and Kristian Skrede Gleditsch, ‘Horizontal
Inequalities and Ethnonationalist Civil War: A Global Comparison’. American Political Science
Review 105, no. 3 (01 August 2011): 478–95. Gudrun Østby, ‘‘Horizontal Inequalities and Civil
Conflict’’, A paper prepared for presentation at the 13th Annual National Political Science
Conference, Hurdalsjøen, Norway, 5–7 January, 2005.
59
Sambanis, ‘‘Conclusion: Using Case Studies’’, 329.

23
Khan, M. Z. Islam, (2016), ‘‘Beyond Greed or Grievance Theory – What Explains Civil War?’’ Peace
and Security Review, Volume 7, Number 15, Fourth Quarter, 2016, pp.26-49.

itself together with corruption and manipulation of market may create 'grievances' that would in
turn facilitate the rebels to materialize their 'greed'. Indeed in countries based on ''the politics of
bellies'', grievance can be as much as political and greed for the control of primary commodity can
often be associated with such political grievances. Collier's absolute rejection to objective
grievance as a potential cause of civil war also stands in sharp contrast to several schools of thought
on civil war. Thus Collier's conclusions that civil war onset is based exclusively on economic
characteristics and not on exclusionary behaviours, horizontal inequalities that triggers objective
grievance seems to be an exaggeration. That however, is not same as saying the model has lost its
relevance in entirely. While it makes sense to link civil wars with economic decline, growth
collapse, high level of poverty, unemployment and feasibility of resources predation by the rebels
in a world where civil wars are increasingly understood and explained as a 'rationale economic
choice', the empirical certainty postulated in the model depicting that the material factors alone
contributes in rebel mobilization is far from being convincing. The reality of political economy,
regional and international commerce, together with HIs and the dynamics of group identity impels
us to reject the juxtapose dichotomy as stipulated by Collier and his team and rather embrace the
fact that both greed and grievance are at work simultaneously in civil war.

(Total words: 9,244)

24