1.1 Understanding-Conflict-and-Violence-Theoretical-and-Interdisciplinary-Approaches-Tim-Jacoby
1.1 Understanding-Conflict-and-Violence-Theoretical-and-Interdisciplinary-Approaches-Tim-Jacoby
1.1 Understanding-Conflict-and-Violence-Theoretical-and-Interdisciplinary-Approaches-Tim-Jacoby
Violence
The causes of violence and conflict are often left untheorised, or they are
discussed as an existent problem assumed to be an inevitable part of human
interaction. This book seeks to examine, interpret and analyse a wide range
of approaches to the causes of violence and conflict. Adopting an accessible
approach, it presents readers with a clear understanding of these by high-
lighting their evolutionary roots and illustrating them with in-depth case
studies and examples.
Tim Jacoby addresses the fragmented nature of the literature on conflict
theory by drawing upon a wide range of disciplinary traditions. This text
seeks to reflect the fact that international relations, history, economics, de-
velopment, politics and sociology all share a longstanding interest in the
study of conflict and violence, and that common concerns make interdisci-
plinary approaches stimulating and productive. It is, therefore, intended for
students and scholars across these disciplines.
Tim Jacoby is Senior Lecturer at the Institute for Development Policy and
Management at the University of Manchester, UK.
Understanding Conflict
and Violence
Theoretical and interdisciplinary
approaches
Tim Jacoby
First published 2008
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
270 Madison Ave, New York, NY 10016
Routledge is an imprint of the Taylor & Francis Group
1 Introduction 1
2 Dimensions 18
3 Structure 34
4 Functions 51
5 Innate 67
6 Learnt 85
7 Grievance 103
8 Mobilisation 124
9 Crises 144
10 Hegemony 160
11 Conclusion 179
References 192
Index 233
Figures
Much of the research for this book was undertaken in preparation for the
lecture programme of a postgraduate course entitled Conflict Analysis which
I teach at the Institute for Development Policy and Management at the Uni-
versity of Manchester in the UK. In devising this course, I owe much to
Keith Webb, who introduced me to many of the ideas included in this book
during my time at the University of Kent at Canterbury in 1997 and 1998. I
also received a lot of support from my colleague at Manchester, Sam Hickey,
who, especially during my first year, was particularly selfless with advice and
support. In running the Conflict Analysis course, I have learnt a great deal
from my students. On the basis of this, I have tried to revise and (hopefully)
improve my treatment of each of the following topics. If this has proven to
be the case, then I owe much to their enthusiasm and endeavour and for
that I am very grateful. I am also indebted to Roger MacGinty, Eric James
and Jonathan Goodhand for taking time to comment on a partial draft of
the manuscript and to Routledge’s anonymous reviewers for their correc-
tions, criticisms and a raft of helpful suggestions. In addition, I am indebted
to Harriet Brinton and Heidi Bagtazo, from Routledge, who have made the
writing of this book as easy and pleasurable as such an undertaking can be.
My biggest vote of thanks must, however, go to my family: to my mother,
my father (who has meticulously proofread my prose and ironed out some
of its errors) and my wife and our three children, all of whom have contrib-
uted immensely to this book and all of whom deserve a considerable share
of whatever merit it deserves.
1 Introduction
This book aims to bring some order to the vast and fragmentary literature
concerning the study of conflict and violence. It considers a diverse range of
perspectives, models and theories that share a common attempt to explain
why people engage in conflictive behaviour in general and violence in par-
ticular. In doing so, it presents nine broad headings under which an eclectic
body of material is covered. By focusing on theories rather than theorists and
on one overarching issue rather than a disparate array of topics, it hopes to
encourage readers to compare various factors and mechanisms, to appraise
common analytical themes and to develop a deeper understanding of the cur-
rent landscape of conflict studies. To this end, a wide assortment of academic
enquiries is presented, examined and illustrated with case studies, data and
examples drawn from various times and disciplines. Research ranging from
1950s functionalism to the latest explanations of civil war, on the one hand,
and from ethology to constructivism, on the other, informs the discussion
of each interlinked approach. Inevitably, the book is selective in the sources
it chooses and, deliberately, it claims neither to offer exhaustive analyses of
each topic it handles nor to cover every scholarly tradition. Rather, it intends
to bring a degree of lucidity and connectivity to a disjointed and frequently
abstruse corpus of literature drawn from the explosion of academic interest
in conflict studies since the Second World War. With this in mind, the chap-
ter to follow endeavours to locate the themes that will be addressed in the
rest of the book within a broader socio-historical context by taking in some
of the more significant changes that have marked both the practice and the
study of violence.
Around the same time as Tzu was writing, but on the other side of the
world, the notion of rational human volition as a motive conceptually sepa-
rable from the influence of fortune or divine preordination underpinned the
work of many Athenian thinkers. Thucydides’ (circa 460–400 bce) account
of the 27-year Peloponnesian War between Sparta and Athens, for instance,
10 Introduction
is widely considered to be the first work in which the past is analysed in-
dependently from the supernatural intervention associated with the earlier
work of Herodotus (circa 484–425 bce) and others. Thucydides, like Plato
(427–347 bce), drew a clear distinction between the use of violence in the
pursuit of foreign policy objectives and the growing public disorder that was
increasingly characterising this period (Price 2001). For both Plato and his
student Aristotle (384–322 bce), such civil unrest had its roots in growing
levels of factionalism within Athenian society. The latter suggested that it
may be a result of three fundamental causes, all of which remain pertinent
today: the unequal nature of Athenian society, frustration with the weak-
ness and incompetence of Athens’ leaders and the desire for the wealth and
privilege that holding political office may entail (Kalimtzis 2000).
For many Roman writers seeking to explain the causes of similar social
problems, such dissent could best be dissipated through reform at the higher
echelons of political society. In rejecting what he describes as ‘the slave’s
ideal of a good master’, Cicero (106–43 bce), for instance, advocated a
concordia ordinum, a harmonious alliance between the senators and the
equites in which conflict would be resolved by selfless and honourable rule,
consensus and the extension of a fraternal omnium bonorum to all citizens
(Everitt 2001). Appian (circa 95–165) shared a similar concern to incor-
porate the views of the masses. His 24-volume historiography represented
the perspectives of the defeated as well as the victorious by organising its
narrative thematically around the histories of Rome’s conquered peoples.
In doing so, he succeeded, like his Greek predecessors, in demonstrating a
major discontinuity between the social impact of Rome’s internecine strife
and life during the expansionist phases of imperial development. This not
only had a profound impact on subsequent Roman historiographies (such as
Cassius Dio’s (circa 155–229) 80-book text), but has, in addition, influenced
contemporary historians and sociologists attempting to compile synchronic
accounts of the past based upon more ‘bottom-up’ discourses on war (Os-
good 2006).
Hellenistic and Roman social commentaries were also influential in Persia
under the Sassanid kings (226–650). There, a prevailing emphasis on the
moral over the natural (derived from the teachings of Zoroaster) gave rise to
a pronounced and eclectic interest in learning. Cities such as Gundeshapur
proved an attractive centre for some of the world’s foremost scholars, in-
cluding eminent Jewish thinkers, Neo-Platonists displaced from Athens by
Justinian in 529 and Nestorian Christians who brought Syriac translations
of classical Greek works (Durant 1980). The resultant dissemination of
philosophical and scientific research was extended by Muslim institutions
such as the Bayt ul-Hikma (or House of Wisdom) founded by the Abbasi
Caliph Abu Jafar al-Mamun ibn Harun (786–833) in 832. Through the work
of academics, such as Al-Shaybani (?–circa 804), Al-Tabari (circa 838–923)
and Al-Mawardi (circa 972–1058), codified limits (hoddod) to definitions of
self-defence and upon battlefield behaviour were elaborated. The sciences of
Introduction 11
tafsir (understanding the Qu’ran) and fiqh (jurisprudence) emerged along-
side research into the sunnah (the life and example of the prophet) and the
lives of the sahabah (companions of the prophet – especially the Khulafa ur
Rashidun or the ‘rightly guided’ leadership of the Caliphs Abubakr, Umar,
Uthman and Ali) to produce sophisticated moral treatises on the use of vio-
lence. A Muslim was, for instance, prohibited from attacking non-combat-
ants, destroying physical infrastructure and mistreating prisoners. Similarly,
war, it was said, could be legally declared only if it was waged in defence
of the faith, against a tyrant or to rescue others from repression (Hussain
2003: 51–8). These are comparable to the classical Jewish categorisation of
just warfare into obligatory (those undertaken at God’s behest), defensive
(including the use of pre-emption) and optional (those pursued for a good
reason) (Solomon 2005). Like all normative conventions, religious injunc-
tions have, of course, been widely flouted. Nonetheless, their enduring ef-
ficacy has been demonstrated repeatedly in a wide range of contexts – from
the refusal of Muslim Hutus to participate in the 1994 Rwanda genocide to
the careful targeting and penal policies of some mujahadeen guerrilla units in
Chechnya during the Russian war of the 1990s (The Seattle Times 9 August
2002; Yule 2000).
Similar ideas emerged within the Christian tradition. As the religion grew
and gained more power in Rome, it was realised that it might become neces-
sary to use violence to protect the innocent. Clement of Alexandria (circa
150–215), for instance, argued the need to defend the empire and to guard
the position of the emperor – just cause and rightful authority in other words
(Johnson 1987). These precepts were expanded and brought into contact
with the work of Cicero by Saint Ambrose (circa 339–97) who helped to
establish the idea that a just war could be undertaken to defend life as well as
religious orthodoxy (Johnson 1987). The five-million-word corpus of Saint
Augustine (354–430) – a convert from the Manichaean sect of Zoroastrian-
ism – extended the definition of just cause to include repelling aggression,
retaking something taken dishonestly and punishing wrongdoing, implying
that war should be proportional, a last resort and aimed at achieving peace.
During the medieval period, writers such as Gratian (circa 1140), Moses
Maimonides (circa 1135–1204), Huguccio (?–circa 1210) and Saint Thomas
Aquinas (1225–74) built on Augustine’s voluminous work to establish an
extensive revival of Roman thought (Swift 1973: 382).
The controversies of the later crusades and, particularly, the introduction
of firearms towards the end of this period augured a new era of battlefield
horrors which, accentuated by a rise in religious zealotry and polarisation
across Europe (illustrated in the brutal subjugation of the New World and,
later, the Thirty Years War), prompted a renewed interest in notions of
the just war from scholars more associated with secular humanism. Here,
Aquinas’ suggestion (drawn from Saint Ambrose) that, even between war-
ring parties, certain rights and covenants should be observed proved to be
profoundly influential. Francisco de Vitoria (1492–1546) and Hugo Grotius
12 Introduction
(1583–1645), for example, argued that sovereigns (ruling, as Francisco Sua-
rez (1548–1617) propounds, not by divine right, but by the consent of the
masses) should act in concert to prevent war’s excesses. This should, it was
suggested, be accompanied by limiting offensive war to just interventions de-
fined by natural law and reason rather than simply by religious difference – a
key premise of the modern Church’s discourse on warfare today. The Holy
See has, for instance, repeatedly condemned the use of religion as a justifica-
tion for violence and apologised for the persecution of heretics, apostates
and non-Catholics in the past. Through an unconditional condemnation
of nuclear weaponry (issued by Pope John XXIII in 1963) and a recurring
call for diplomacy in place of violence (most notably in Pope John Paul II’s
implacable opposition to Great Power policy in the Gulf), the Church has
significantly narrowed its definition of what constitutes a ‘just’ war (Duncan
2003).
Taken together with the earlier work upon which they build, these
propositions represent the philosophical underpinnings of much of what
the West currently defines as the legitimate use of force in the international
arena. They contain, in other words, the essence of jus ad bellum and jus
in bello (Walzer 2006). The former is normally construed as the idea that
sovereign states (not private armies) directed by a legitimate authority can,
once all other means of redress are exhausted, use proportionate violence
in response to aggression. This may be in self-defence or in the assistance of
another state (or, in some extreme cases (such as genocide), a group within
another state) aggressed against and must have the aim of returning to a
peaceful status quo. The latter is generally understood to consist of a set
of core principles guiding the conduct of warfare. Typically, these include a
responsibility to distinguish between soldiers and civilians, to deploy a level
of force commensurate with the tactical goal sought (thereby prohibiting
certain weapons) and to adhere to generally agreed rules regarding the treat-
ment of prisoners of war. As with other moral traditions, of course, Western
states have regularly flouted these conventions (and continue to do so), but
discussions of warfare remain closely connected to notions of justice (Frost
2004).
Indeed, as economic interdependence has grown during the industrial
era, the kind of pooled structure of sovereignty, which could replace the
signing of ad hoc peace treaties with a pacific federation of states, implied
by renaissance writers and articulated more fully by Immanuel Kant (1724–
1804) has gained popularity and come to underpin modern ideas of liberal
internationalism (Reichberg 2002). For writers such as Norman Angell, who
presciently pointed out the imminent folly of the First World War, the fact
that the wealth of states is no longer held by absolute monarchs relying on
the extraction of tribute renders redundant the idea that war can reduce
an opponent to an inferior position. Instead, he concludes, force should be
used only to increase international cooperation, in the same way as police
are needed because thieves refuse to observe the normative structures of
Introduction 13
domestic communities (1910). The human and material costs of the Great
War’s industrialised combat convinced many that Angell (who received a
Nobel Peace Prize in 1933) was correct, and his concept of collective security
was, partly at least, embodied in the League of Nations and, later, the United
Nations. His work also helped to establish the first chair in International Re-
lations at Aberystwyth to investigate the ‘causes of war, and the conditions of
peace’ in 1919, the British Institute of International Affairs in 1920 and New
York’s Council on Foreign Relations in 1921 (Groom 1992).
Not only did the First World War increase the attention given to the
causes of large-scale conflicts, it also provoked numerous efforts to moder-
ate the ways in which modern warfare was conducted through humanitarian
intervention. Although such activity during periods of combat has a long
history, the conflicts of the mid- and late Victorian age (of which the First
World War was the culmination) showed an increasingly internationalised
European middle class that the technical means of killing had far outstripped
any sense of restraint in their application (Lawrence 1997: 31). Through
the work of the Red Cross (founded in 1863) and others, conventions (es-
pecially that signed in Geneva in 1864) institutionalised battlefield access
for non-combatants to organise medical treatment, to monitor the conduct
of the conflict (including the use of certain types of weaponry) and to ar-
range prisoner exchanges. During the First World War, for instance, the Red
Cross was permitted access, on the basis of its demonstrable commitment to
impartiality, neutrality and independence, to over 500 penal camps where
it recorded the identity of more than two million captives. In the Second
World War, the Red Cross (along with Red Crescent societies) carried out
nearly 13,000 such visits and, despite failing to reach an agreement with
the German authorities over access to their camps, maintained a database
of around 45 million prisoners of war (Wylie 2002). Since then, the size
and scope of non-governmental and multilateral organisations involved in
humanitarian work have, along with charitable donations and high-impact
televisual reportage, enlarged considerably. The United Nations, for exam-
ple, has taken on an increasingly interventionist role in world politics – from
dispatching an observer mission to Palestine in 1948 to deploying an armed
contingent of 28,000 ‘peace enforcers’ to Somalia in 1993 (Jeong 2000).
Today, its Office for the Coordination of Humanitarian Affairs has some 860
staff and a budget of around $110 million.
As the joint award of the 1901 Nobel Peace Prize to Henri Dunant (the
founder of the Red Cross) and Frédéric Passy (the anti-war campaigner and
co-founder of the Inter-Parliamentary Union) illustrated, such efforts fre-
quently coincided with broader endeavours to prevent violence altogether.
These, too, have a long history. The Pax Dei, or ‘peace of God’, for example,
prohibited (from the mid-eleventh century onwards) violence in or upon
consecrated people, places and occasions and was gradually extended to in-
clude women, children and other days. During the Middle Ages, the Dutch
humanist and theologian Erasmus (circa 1466–1536) set out a reasoned case
14 Introduction
against warfare of all kinds, while Christian groups such as the Waldensians,
the Mennonites and elements of the Religious Society of Friends (the Quak-
ers) have continued to renounce all forms of violence – as have organisations
from all other faith groups (Barash and Webel 2002). While some writers,
such as Søren Kierkegaard (1813–55) and Leo Tolstoy (1828–1910), rejected
organised resistance in favour of a pacifism based on individual reflection
and God consciousness, other individuals have mobilised religious senti-
ment to achieve a considerable political impact. Prominent here have been
Ghandi’s swadeshi movement, Martin Luther King’s Southern Christian
Leadership Conference and the Dalai Lama’s Tibetan government-in-exile.
Alongside such organisations, others have achieved high levels of popular
support through a non-ecclesiastical peace agenda. These include the In-
dustrial Workers of the World (created in Chicago in 1905), the Women’s
International League for Peace and Freedom (founded in Washington, DC,
in 1915), the War Resisters League (established in London in 1923) and the
World Peace Council (convened simultaneously in Paris and Prague in 1949).
Both secular and faith-based civil peace organisations – as well as the senti-
ments they represent – remain, as the examples of Desmond Tutu and Aung
San Suu Kyi testify, profoundly influential today.
Since warfare entered the age of mass destruction weapons in 1945, the
anti-nuclear movement has arguably been the largest and most vociferous
of these. Early dissent took the form of a 1955 statement, signed by 11
intellectuals led by Bertrand Russell and Albert Einstein, which suggested
that ‘whatever agreements not to use H-bombs had been reached in time
of peace, they would no longer be considered binding in time of war, and
both sides would set to work to manufacture H-bombs as soon as war broke
out’. Two years later, this became part of the founding charter of the Pug-
wash Conferences on Science and World Affairs – an international forum of
scientists committed to disarmament. Having received a Nobel Peace Prize
in 1995 for ‘their efforts to diminish the part played by nuclear arms in
international politics’, it has now convened nearly 300 meetings involving
over 10,000 delegates. In the United Kingdom, the Campaign for Nuclear
Disarmament (established in 1958) has maintained a membership of over
50,000 for most of the last 25 years and currently claims to be Europe’s big-
gest single-issue peace campaign (Hudson 2005). In global terms, the Nobel
Peace Prize-winning (1985) International Physicians for the Prevention of
Nuclear War (founded by American and Soviet cardiologists in 1980) rep-
resents one of the most internationalised movements with a membership of
more than 200,000 people from over 60 countries (Wittner 2003).
However, while these organisations and the broader peace movement
have been highly influential within the academic study of conflict and vio-
lence, policy making has remained dominated by the belief that reducing a
state’s level of war preparedness is not the most effective way to minimise
the possibility of large-scale conflict. The idea of deterrence, and the school
of realism that underpins it, is grounded in a clear distinction between a
Introduction 15
forceful geopolitical state and a passive domestic polity that can be traced
back to the work of Dante (circa 1265–1321), Machiavelli (1469–1527)
and Clausewitz (1780–1831). For strategic studies researchers such as Hans
Morgenthau (1948) and George Kennan (1982), the gravest danger of war
emerges not from an aggressive assertion of sovereignty, but from misper-
ception and imbalances in the distribution of power between states. As we
shall see in Chapter 9, a nuclear arsenal, along with alliances and conven-
tional weapons, is regarded as having the potential to reduce uncertainty
about a state’s position and to ensure conforming behaviour in others. A
hierarchy of states may emerge in which the hegemon maintains a stable
and ‘realistic’ distribution of destructive power and alliances – ensuring, as
Chapter 10 outlines, that large-scale conflicts are restricted to a (possibly
cyclical) pattern of contest over the leadership position. The Hobbesian
conservativism of seeing states and individuals as immutably conflictive (and
therefore favouring robust precautionary and remedial measures) derives its
motivational premise from a view of human beings as having an innate drive
to aggress and dominate (see Chapter 5), which can be understood in terms
of Darwinian biology, evolutionary psychology or settled socio-historical
cultural legacies.
Such a position differs greatly from the behaviourist view of conflict, dis-
cussed in Chapter 6, as socially constructed, gendered and changing (Císař
2003). As the foundation of modern peace movements, this tends to see
violent behaviour as primarily driven by environmental factors, and thus
possible to extirpate. To do so, it was suggested that realism’s emphasis on
the Cold War’s East–West balance of power as the basis of, and a threat
to, global peace had to be replaced by a concern to reveal the frequently
subliminal conflict between the global North and the South. As Chapter 3
explains, Marxist-influenced writing emerged in Europe and moved defini-
tions of conflict beyond a focus on overt, behavioural hostility towards an
understanding of violence as part of the normal, yet unjust, operation of
the international system (Callinicos 2003). This questioned the notion that
‘peace’ in a world of such injustice was an objective worth pursuing. It also
contained a broader definitional challenge: that relying on the narrow per-
ceptions of actors to define their own situations inevitably obscures broader
concerns and restricts the analyst to dealing with extant power relations
(Lopez 1989: 69). The tension between such subjectivism and the poten-
tial dangers of its objectivist alternatives is pertinent to the social sciences
in general and, along with associated debates over the different levels of
investigation available to conflict analysts, is thus this book’s initial point of
enquiry as the primary concern of Chapter 2.
The idea that conflicts can be understood and compared at various levels
of analysis – from the macro-economic forces of the global system to the lim-
ited awareness of the individual – questions another of realism’s fundamental
tenets: that the domestic polity is conceptually separable from international
geopolitics. As increasing numbers of states in the world system produced a
16 Introduction
considerable rise in the frequency of civil wars following the Second World
War (Singer and Small 1982), interest grew in the relationship between con-
current processes of modernisation and decolonisation, on the one hand,
and enlarging expectations, inequality and grievance formation, on the
other (Dos Santos 1970; Blum 2003). Building on the frustration–aggres-
sion model developed by social psychologists in the 1940s and prior studies
on human needs, the notion of relative deprivation, the focus of Chapter
7, emerged as a key means of explaining the civil unrest of the period. This
prompted a re-evaluation of earlier work dealing with the functionality of
conflictive behaviour, particularly political elites’ use of conflict – both do-
mestically as a means of identifying an ‘other’ through which an enhanced
sense of ‘self ’ may be constructed and internationally as a similarly specious
method of distraction. The causes of conflict, as Chapter 4 demonstrates,
may thus be found in rulers’ desires to maintain their domestic authority
(Miller 1999). Indeed, the mobilisation of individuals into social movements,
national groups or bands of insurgents (to be considered in Chapter 8) may
be regarded as a better way to explain the causes of collective violence than
motivational studies of unrealised expectations or perceived frustrations that
do not account for power differentials and individuals’ rational calculation
of costs, benefits and probabilities of success (Buechler 1999).
Interest in a section of this latter body of work – the so-called ‘greed’
thesis – has, to the detriment of other theories of group mobilisation, been
sharpened by the perceived ‘end of old-fashioned war between states’ in fa-
vour of ‘new wars’ (Kaldor 2001: 1). These are characterised as deliberately
focused upon civilians as a source of revenue (whose share of total fatalities
has increased from around half during the 1960s to over 90 per cent by the
early 1990s) and organised by avaricious conflict ‘entrepreneurs’ (Collier et
al. 2003). Recent studies have therefore tended to place responsibility for
such bloodshed largely upon the shoulders of local agents. This has, in many
cases, produced depictions of contemporary conflicts ‘as explicable simply
by reference to the narcissism of violence’, leading to absurd portrayals of
‘perpetually recidivist societies’ where ‘torture is exciting, rape is fun, and
looting is profitable’ and which can only be salvaged through the ‘imposi-
tion of imperial order’ (Gray 1999: 277; Shawcross 2000: 169–78; Cooper
2005: 464–5). The result is a partial abandonment of the intellectual pursuit
of ‘why’ conflicts start in favour of ‘how to’ organise a response.
This book is, in many ways, an attempt to return to ‘why’ questions as the
bedrock of conflict analysis. At the heart of preserving a wide-reaching and
independent discipline is the maintenance of a rich theoretical tradition and,
in particular, a broad understanding of the range of approaches to analysing
conflict instigation. What follows, then, is not intended to be a mere list of
possibly relevant topics. Instead, it is nine inter-related – and sometimes op-
positional – ways of theorising the causes of conflict and violence (italicised
below) which, together, constitute a comprehensive (though not exhaustive)
account of why different conflictive phenomena arise. Based on around 900
Introduction 17
academic sources drawn from more than half a century of research within
numerous academic traditions, this study proceeds by initially considering
the dimensions, and difficulties, of defining conflict at various levels of analy-
sis, before examining the notion of structure as a fundamental challenge
to the way in which conflict and violence is commonly understood. The
idea that the normal operation of the international system is intrinsically
violent leads to a more thorough treatment of who benefits from conflict
– its functions in other words. A key element in both the functional creation
of enhanced in-group coherence and the legitimisation of aggressive foreign
policies is the idea that there is an innate basis to ethnic or national solidarity
born of the immutably conflictive nature of the human condition.
Alternatively, understanding conflict and violence as essentially learnt
could reveal the ways in which individuals and groups are organised into
conflictive entities by elites. Neither of these approaches, however, explains
why the masses should wish to follow elites. Here, participation in organised
forms of conflictive behaviour may be determined by the presence of griev-
ances born of the unequal division of material wealth and social status or
by the mobilisation of individuals’ rational assessment of the prospective
benefits that conflict and violence might offer. Indeed, the distribution of
resources through the international system can also be understood to be the
fundamental cause of conflict and violence. Just as individuals perceiving
themselves to be threatened seek to improve security and lower their isola-
tion levels, so states respond to geopolitical pressures – particularly crises
– by acquiring weapons and forming alliances. These patterns of interaction
have produced hierarchical power blocs within and across which warfare
has been used to attain or defend hegemony, producing ever more severe
instances of violent conflict and raising the possibility that the next great
contest over global leadership may be the last.
2 Dimensions
Explaining why people engage in violent conflict has, as the previous chapter
outlined, absorbed academic thought for a great many years and has given
rise to a wide range of deliberations. Not least among these are the problems
of comparing conflicts at an abstract level, identifying conflicts’ component
parts, defining what a ‘conflict’ means and locating conflicts at various levels
of social interaction. This chapter proposes to introduce the reader to some
of the more significant elements of these discussions. It first presents some
ideas surrounding the problem of approaching conflict generally by looking
at the evolution of the literature concerning conflict dimensions. Here, there
continues to be considerable debate. Analysts disagree over how conflicts can
be identified, their origins, their consequences and even whether or not they
necessarily involve violence or aggression. Second, it will discuss issues sur-
rounding the definition of conflict. This is subject to a broader divide within
the social sciences; namely, between approaches that emphasise the impor-
tance of actors’ perceptions and those which argue that conflicts and other
observable social phenomena can be present and influential without the self-
awareness of those involved. Third, it will consider the ‘levels of analysis’
issue in conflict studies, first put forward by Kenneth Waltz in 1959 in his
book Man, the State and War. It will deal with the different analytical impli-
cations of examining conflicts from various viewpoints, from intrapersonal
dissonance to interpersonal disputes, civil unrest, interstate war and global
conflagrations. As such, looking at these three topics – generalising about
conflict, definitional debates and the unit of analysis – will help to introduce
concerns that run right through the rest of the book.
Situation
Behaviour Attitudes
Figure 2.1 The structure of conflict. Source: Redrawn from Mitchell (1981a: 16).
20 Dimensions
situations contain both realistic and unrealistic elements. They also tend to
involve the expectation of zero-sum and variable-sum outcomes. The former
relates to circumstances in which the total benefit to all actors always adds
up to zero. In other words, no gain can accrue to one party without an equal
loss to another. It is typified by conflicts perceived to be based upon limited
resources, such as land in Palestine or diamonds in Sierra Leone. The lat-
ter involves conflicts over goods which are not, inherently, in short supply
– such as status, legitimacy, religious expression and dignity – and so do
not, theoretically at least, inevitably mean that one actor must lose in order
for another to gain. In this sense, perceptions of scarcity within conflictive
situations may be over material goods like oil, lootable objects and so on
or positional goods such as access to political representation and economic
management.
There are also qualitative differences in the type of goal pursued which
determine the nature of a conflictive situation. Different interests may lead to
conflict, although parties basically concur about the value of some position,
role or resource and the various factors that may have led them to disagree.
Conflict here arises over issues of distribution within a broadly agreed frame-
work. Examples might be a husband and wife arguing without considering
divorce or academics discussing their profession while agreeing that higher
education provisions are worthwhile. Consequently, a compromise is usu-
ally possible. On the other hand, conflicts of value exist when parties differ
fundamentally about the nature of desirable end-states or social and political
structures. The conflictive relationship between the goals of the World Trade
Organization and the Anti-Globalization Movement is an example of this,
although, as Anthony de Reuck points out, ‘the distinction is rarely clear-cut’
(1984: 97). Within the latter, there is liable to be a conflict over what has
given rise to the set of circumstances from which a dispute or problem has
arisen – the question of attribution. Here, parties may hold each other or a
third party responsible. Protagonists are also likely to disagree over the best
means of dealing with the problem – often stemming from a different view
of its causes. The ways in which interests, value, attribution and means inter-
relate within a conflictive situation are illustrated in Table 2.1.
Conflictive attitudes are closely related to conflictive situations. Mitchell
defines them as ‘common patterns of expectation, emotional orientation
and perception that accompany involvement in a conflict situation’ (1981a:
28). Here, anger, resentment and suspicion are common, as are cognitive
processes such as stereotyping and selective approaches to new information.
These frequently become self-perpetuating in the sense that previous experi-
ences of a conflict will reinforce or exaggerate conflictive attitudes in the
future. As a source of conflict, attitudes, in the form of internal drives and
thoughts, may push individuals towards conflictive behaviour. In this sense,
there is a fundamental division (which will be looked at in more detail in
Chapters 5 and 6) between analyses that focus upon actors’ response to their
Dimensions 21
Table 2.1 Aspects of conflictive situations
Agreement on basic values
Conflict aspects and ends Agreement on causes
Value No No
Interest Yes Yes
Attribution Yes No
Means Yes Either
Source: Mitchell (1981a: 42).
Defining conflicts
General models also tend to contain three elements apparent before the
manifestation of conflictive behaviour. In the case of Mitchell’s work, these
Dimensions 23
are, as Figure 2.2 illustrates, incipient, latent and suppressed conflict (1981a:
51). The first can broadly be regarded as a situation in which a conflict is not
recognised by one or both parties. The second emerges when goal incompat-
ibility is perceived, but not sufficiently motivating to give rise to observable
conflictive behaviour. In the third scenario, one or more parties are aware
of a conflictive situation, but the costs of pursuing their goals are too high
to produce conflictive behaviour. In Mitchell’s model, these three elements
are not, however, theorised. Broadly speaking, he and others who adopt a
subjectivist approach to identifying conflicts argue that actors must not only
believe that they are in a conflictive situation, they must also manifest this
belief in a way discernible to others. This position is subjectivist in the sense
that it relies on the actors, or subjects, involved in the conflict to define their
situation as conflictive.
An alternative approach to relying on the actors’ own perceptions is to
define what constitutes a conflictive situation itself – the object as opposed
to the subject. It is often argued that this offers a more inclusive way of
identifying conflicts as it takes into account the frequently imperceptible
formations which surround actors. In this sense, conflicts are ‘not seen as a
matter of subjective definition but as determined by the social structure. In
other words, conflict is incompatible interests built into the structure of the
system where conflict is located’ (Schmid 1968: 226). Such a ‘maximalist’
agenda includes ‘the effects of social and economic exploitation’ within a
definition of a conflictive situation – regardless of whether or not this is per-
ceived by those involved (Rogers and Ramsbotham 1999: 744). A slave may
not, for instance, be aware that she is in a conflictive relationship with her
mistress. This may be particularly well hidden if the mistress defines the role
of the slave and teaches her that being a slave is part of the normal or natural
way of the world. The slave is unlikely to question its legitimacy so long as
the mistress reinforces the structure of their relationship with kindness and
No contact or
complementary
goals
Suppressed
conflict
Figure 2.2 Conflict processes. Source: Adapted from Mitchell (1981a: 51).
24 Dimensions
palliatives. The result may be that the slave feels devotion to her mistress and
may, if offered, reject the chance of freedom. Thus, the relationship between
the two is cooperative. Both actors conform to agreed rules, and there is
little or no conflict over values, interests, goods and so on. For a subjectivist,
this situation can only be identified as incipient, latent or suppressed and is
thus not, for the most part, the subject matter of the conflict analyst. It is, in
other words, ‘peace’.
Such a view has considerable implications. If, for instance, the slave per-
ceives that she is in a conflictive situation and acts by effecting some form
of conflictive behaviour, then, according to the subjectivist position, she has
changed her goals and manifested a conflict. By implication, then, it is her
behaviour that is the primary concern of the conflict analyst rather than
the role of the mistress or the nature of their relationship. As Anthony de
Reuck puts it, ‘the first party whose conflict behaviour becomes conscious
and deliberate is often labelled the aggressor, whether or not he is the ag-
grieved party’ (1984: 100–1). This is obviously problematic, as it suggests
an assumption that conflictive behaviour represents an unwelcome event per
se and that it is the role of the analyst to try to find ways of reducing the
intensity of this behaviour, rather than to attempt a reconstruction of the
conflictive relationship itself.
A further problem with the subjectivist position is that the differences
in power that exist between conflict actors are often ignored in favour of
the presumption of a shared commitment to ‘peace’. As, in order to analyse
different aspects of conflictive situations, attitudes and behaviour accurately
and ultimately suggest ways of reducing the intensity of that conflict, the
researcher must have comparable levels of access to the actors involved, sub-
jectivist models mostly advocate various notions of neutrality; an idea that
is supported by the adherence to the idea that the subjects themselves must
provide the contextual details of the conflict (Burton and Dukes 1990). By
and large, then, it is assumed that it is both possible and desirable for the
researcher to be positioned equidistantly from each of the conflict actors.
Of course, some conflicts, such as the Cold War confrontation between the
Warsaw Pact and the West, do involve approximate power parity, and thus
may, for researchers from non-aligned states, offer a viable, ‘neutral’ ground
roughly halfway between each. Mostly, though, conflicts are between parties
with asymmetrical capabilities.
So, in terms of power, to be equidistant from each actor, the researcher
must, as Herman Schmid has observed, actually be closer to the less power-
ful party in order to be in a mutually neutral space (1968). However, is such
a location likely when the powerful are both better equipped to commit
resources to the analysis process and more likely to welcome a return to
the pre-existent situation? To explain this, Keith Webb offers the example
of an industrial dispute in which a conflict analyst is only acceptable to the
management if she accepts the basic structure, and inherent inequalities, of
capitalist production and exchange. Bringing an end to the dispute can thus
Dimensions 25
only hope ‘to restore the status quo, leaving the relationship still asymmetri-
cal’ (Webb 1986: 432). So, given such power differentials, Schmid is perhaps
correct to assert that the subjectivists’ ‘claim of political neutrality is hardly
satisfied by leaving the evaluations to the decision-makers’ (1968: 220).
The concept of a collective dedication to peace is similarly tricky. Even
where parties possess broad power parity, as in the bipolar blocs of the Cold
War, it is uncommon that a shared commitment to a positive future is suf-
ficiently potent to provide a normative basis for a reduction in conflict inten-
sity. Subjectivist analyses of conflict therefore often resort to the ‘minimalist’
objective of war avoidance – in this case, the universally catastrophic effects
of a Superpower nuclear exchange (Rogers and Ramsbotham 1999: 744).
The fear of a mutually harmful outcome is thus presented as the primary
grounds for the work of the conflict analyst, for the tractability of the con-
flict itself and for the amenability of the conflict actors. For many, though,
the mere absence of overt, behavioural violence does not constitute a satis-
factory definition of peace. Adam Curle, for instance, suggests that accept-
ing such a negative notion of peace frequently becomes a means for more
powerful parties to obscure objective goal incompatibilities. He points to
the South African government’s policy of granting a number of ‘Bantustan’
settlements some degree of autonomy during apartheid. Although these had,
by 1994, been abolished and reabsorbed into the South African state, ‘the
illusion of a certain degree of independence for the non-whites blunt[ed]
perceptions of the deep conflict of interest between black and white South
Africans’ (Curle 1971: 9).
As a negative peace may not have any actual manifest conflict, a positive
peace is not merely an absence of conflictive behaviour, but involves coop-
erative relations leading to an end to the exploitation of the powerless by the
powerful. Curle puts it like this:
Such a future, does not, for Curle, necessarily imply an idealistic egalitarian-
ism or a vision of future socialism. Positive peace may contain goal incompat-
ibilities born of varying power capabilities and access to material resources.
The key difference lies in the management of the difference in power. For
Curle’s positive peace, the powerless are helped to develop their capabili-
ties by the powerful. An example might be a parent–child relationship or a
pluralist society in which latent, incipient and suppressed conflicts do not
emerge.
Herman Schmid offers a different perspective. He argues that, first,
26 Dimensions
inequitable structures and institutions, which are broadly accepted by Curle
and others, are a fundamental part of any given conflictive situation. Second,
he suggests that the way in which these are understood is primarily drawn
from ‘the wealth of social science theory developed for the control and in-
tegration of the national system’ and is thus a component of conflictive at-
titudes (1968: 219). Consequently, conflict studies should not, he continues,
simply attempt to analyse conflicts, nor should they seek to make policy rec-
ommendations to governments who frequently represent the powerful side
of an asymmetrical conflict. The fact that the bulk of this type of research is
located in, and frequently directly financed by, the United States and north-
ern Europe – those with the greatest vested interest in maintaining the status
quo – adds considerable credence to Schmid’s position. The overall result,
he suggests, is that conflict analysts have a strong propensity to adopt ‘a
system perspective and a value orientation which is identical with those of
the existing international institutions and lies very close to those of the rich
and powerful nations’ (Schmid 1968: 221). Given that, throughout history,
positive social change has generally occurred through an increase in conflict
levels rather than a decline, should, Schmid asks, conflict analysts continue
to advocate reductions in conflict intensities?
In conclusion, it is clear that the objectivist critique of subjectivism is both
comprehensive and an ongoing feature of the discipline. It has not, however,
passed without rebuttal. As Mitchell notes, objectivism’s tendency to conflate
inequality with conflict contains the real danger that the peculiar importance
of conflict analysis is lost within broader discourses on the egregious effects
of injustice and imparity – a particular problem for notions of structural
violence discussed in Chapter 3 (1981b). Moreover, does differential status
and wealth inevitably imply an incompatibility of goals? The elites of the
developing world are, for instance, mostly poorer and less powerful than
the elites of the West, yet does this mean that there is a conflict between
them while both are prospering from the international terms of trade and
relations of production? Simply because actors operate within an unequal
environment, it does not self-evidently follow that they will wish to redress
the inequities of the negative peace in which they operate. To assume that
they are blind to the ‘objective’ truth of their position is, it has been claimed,
to give conflict research ‘the qualities of an intellectual black hole’ (Lawler
1995: 237). Moreover, such a positivist faith in a ‘reality’ obvious only to
the initiated contains the danger that conflict analysts will arrogantly export
their vision of equality, peace, etc. without regard for contextual specificity,
thereby escalating, or even fomenting, conflictive behaviour – Schmid’s ul-
timate conclusion. As adherents of the objectivist position are neither free
from value-driven judgement nor able to predict the outcome of any change
in social structures and institutions, they tend to become entrapped by their
own critique of subjectivism. In other words, they too attempt to base their
analysis on questionably universal notions of a positive future, in this case
the value of wholesale systemic change.
Dimensions 27
Levels of analysis
The following section presents an overview of some of the ways in which
these controversies and ideas have been structured by attempts to identify
the causes of conflict at various levels of analysis, first put forward by Waltz
in 1959. He suggested that the causes of conflict could be found within
three different, but interlinked, social strata – the individual, the state and
the international system. Since then, commentators have expanded these to
include various substate and supranational groupings which, depending on
the research question posed, can be employed as taxonomical devices to
categorise and compare instances of conflict. Sometimes, these may be dis-
crete and distinct. Assessing macro-changes in the frequency of warfare over
the last few centuries would, for example, clearly necessitate studies linked
to the level of the system, whereas the reasons for a government choosing
certain foreign policy options would tend to lead the analyst towards a con-
centration upon state-level concerns. It is, however, generally not necessary
to include all levels in an analysis and, most often, conflicts are examined as
an interaction of a number of different levels. Liberal theories of economic
competition, for instance, combine models of global exchange with both
state-based processes of expansion and recession and rational actor accounts
of individual behaviour (Hayek 1991).
Indeed, as Table 2.2 illustrates, conflicts frequently take different forms at
different levels as they proceed. In this case, it would appear that the actors’
values are, in Mitchell’s terms, going through a process of intensification
followed by de-intensification, thereby causing the conflictive situation to
escalate, de-escalate and, ultimately, to end in some reparatory interaction.
Categorising conflictive behaviour like this offers a fruitful way of consider-
ing inter-relations between various parties. Each row represents a particular
way in which actors respond to goal incompatibility. Costs and benefits are
conferred or withheld, through either action or threat. Each column shows
how this behaviour may be manifested at different social levels. The first
of these outlines a common pattern of interpersonal conflictive behaviour,
while the second and third columns use the example of industrial and inter-
national relations respectively.
At both the systemic and the state levels of analysis, a longstanding ten-
sion has existed between realist, liberal and socialist traditions. As an ap-
proach to foreign policy, the first tends to take states to be the key actors and
assumes that they attempt to behave rationally and to advance their security
and wealth in an environment which, in the absence of a legitimate source
of coercive power (a police force in other words), is perceived to be anarchic
and intrinsically threatening (Levy 1996). As wealth is born of trade, which
in turn relies on pacific markets, it is thus argued that the primary role of
governments operating at the international level should be to ensure their
country’s own defence through armaments (Sheehan 1996). To do this, as-
sessments of other states’ power should proceed on the basis of ‘worst-case
28 Dimensions
Table 2.2 Types of conflict process
Conflict level
Conflict behaviour Individual Collective International
Reduce benefits to Reduce contact Work to rule Delay aid deliveries
adversary with person
End interaction Send to Coventry Strike Stop aid, trade and
with adversary diplomacy
Begin imposing Insult person Prosecute union Confiscate local
costs on adversary leaders assets
Intensify cost Physically assault Hire violent strike- Deploy armed
imposition person breakers forces
Reduce cost End aggression End strikers’ sit-in Reduce conflict
imposition outside self- intensity
defence
End cost imposition End all aggression Stop prosecution Armistice
Begin benefit- Find excuses Limited return to Resume diplomacy
conferring action for adversary’s work aid and trade
conduct
Intensify benefit- Praise adversary Improve pay and Increase trade and
conferring action conditions aid to new levels
Source: Adapted from Mitchell (1981a: 127).
Conclusion
In keeping with the rest of the book, this chapter has covered an eclectic
selection of literature dealing with a wide range of debates and problems
broadly connected to the difficulty of undertaking conflict research. In par-
ticular, it has looked at three aspects of conflict dimensions. The first relates
to the existence of fundamental insights that seem to be true of all conflicts
– an attempt, in other words, to identify essential patterns of similarity be-
tween ostensibly very different social phenomena. Iraq’s invasion of Kuwait,
the Khmer Rouge’s activism in Cambodia, the French Revolution and do-
mestic violence may appear to be utterly unlike, yet at their heart lies a basic
struggle for status recognition, for resources or, even more broadly, for the
realisation of an incompatible set of goals. To make sense of this, Mitchell’s
triangular model of situation, behaviour and attitudes is, this chapter has
suggested, an efficacious means of understanding both the various elements
that make up an analysis of a conflict and the ways in which these change
over time.
The second section above highlighted some of the implications of analys-
ing the components and dynamics of conflictive behaviour, attitudes and
situations. Here, a fundamental divide was noted and explained. On the
one side, the need to empower protagonists to respond to their own social
environment underpins a tendency to define a conflictive situation as exist-
Dimensions 33
ent only at the point of actor perception; when, in other words, conflictive
behaviour manifests itself. By and large, this approach focuses on the demon-
strable actions of subjects, thereby maintaining an interest in the ‘minimal’
objective of reducing conflict intensities and scope as the foremost priority
of the parties involved. Objectivism, by contrast, tends to concentrate on the
conflict as a definitional unit commonly independent of parties’ perceptions
and often identifiable only by a third party. Conflictive situations can, there-
fore, be latent, hidden from view or endemic to the ‘normal’ functioning of
society. Such ideas are frequently found in notions of ‘false consciousness’ as
well as other ‘universalisms’, such as inalienable human rights, and are thus
associated with a more ‘maximalist’ agenda for social change, revolution
and peace.
Third, this chapter also considered the levels of analysis debate. This was
initiated by Kenneth Waltz, who suggested that the causes of collective vio-
lence could be found in three interlinked strata: (1) individual belief systems,
instincts, personalities and psychological processes; (2) societal factors such
as government, economic exchange and national ideology; and (3) an inter-
national system structured by power differentials, alliances and trade. Other
writers have augmented these by disaggregating intrastate groups and by
emphasising the independent role of suprastate formations within the glo-
bal system. In some ways, concentrating on one particular level of analysis
reinforces disciplinary boundaries; states and systems constitute the basis of
international relations, while the individual level informs much of social psy-
chology’s contribution to conflict studies. Introducing them together does,
however, help to highlight their comparative strengths, and placing them
within the context of this chapter’s overall focus on generality and definition
also serves to mark out the parameters of conflict studies and to frame the
context of the rest of the book. In the next chapter, for instance, we will
look at the notion of the structure at various levels of analysis. Particularly
focused upon will be objectivist visions of how the causes of violence might
be understood and measured across the international system, inside states
and within families once the need to define them through subjective percep-
tion and direct, interpersonal agency is abandoned.
3 Structure
By the late 1960s, however, the conflict between the West and the Warsaw
Pact, along with associated weapons transfers and highly destructive proxy
wars, combined to define developing countries’ domestic policy and nullify
much of the possible benefits of economic assistance. A growing awareness
that, despite considerable economic growth over the postwar period as a
whole, the divide between rich and developing countries was widening led
former Canadian Prime Minister, Lester Pearson, to conclude, in 1969, that
attempts to bring stability and prosperity to the developing world were ‘heavy
with disillusion and distrust’ (quoted in UNICEF 1986: 308; Singh 2002:
297). This combined with vociferous social forces in the West, such as the
radicalisation of black consciousness, the anti-Vietnam movement, student
protests and industrial unrest, to produce a sharp rise in civil dissent in what
Jürgen Habermas called ‘the first bourgeois revolt against the principles of a
bourgeois society’ (1971: 28).
The sense that Western capitalism was facing a crisis of legitimation also
influenced the social sciences (Bendix 1967). Within conflict studies, the
discipline’s apparently close relationship with Western interests came in for
particular attention. For much of the postwar period, analyses of conflict
had focused disproportionately on the study of Superpower politics and
the nuclear threat (Dunn 1978). Predominant was the view that interna-
tional tension could be best reduced through the establishment of a powerful
multilateral authority. For the most part, issues of domestic civil disorder,
36 Structure
revolutionary irredentism and protest movements were either ignored or re-
duced to a problem of counterinsurgency. The use of force by Western states
therefore remained undertheorised, leading to a commonplace assumption
of legitimacy, public authorisation and, as the 1960s progressed, a sense that
a normative emphasis upon the universal value of pacifism was becoming
increasingly irrelevant (Garver 1968).
Dependency theories, which first appeared within networks of South
American writers, led to a greater focus on the extent to which policies
pursued by rich countries actively prevent poorer countries from developing
(Valenzuela and Valenzuela 1978). In general, these owed much to Lenin’s
work on the economic causes of imperialism and Marx’s theory of colonial-
ism. They refuted modernisation theories’ emphasis on the internal charac-
teristics of developing countries in favour of a structural, internationalised
analysis in which global inequality, as the stimulus of much of the world’s
conflict, is perpetuated by the normal operation of the economic system and
its trading structures – thereby ensuring that developing countries remain
dependent on Western power (Smith 1979). The protection of European
agriculture, for instance, excludes competition from the South and ensures
that most added value is located in the West and its manufacturing sector
(cocoa is grown in Ghana, sold to the West at raw material prices, processed
and distributed to Western consumers at high prices, while locally produced
chocolate is denied access to Western markets by exorbitant tariffs) (Griffin
and Gurley 1985).
This is, it was suggested, backed up the West’s policy of aggressively
exporting its culture to stimulate sales of its consumer goods (which are
often manufactured in the developing world to undercut local competition
and institutionalise further exogenous control of developing markets). Fur-
thermore, dependency is enhanced by the international legal regime and
its capacity to facilitate penetration of developing economies through land
purchases, military bases, World Trade Organization rulings, United Nations
missions and so on. These are, it was argued, frequently supported by the
arming of internal dissenters, exerting diplomatic or economic pressure,
invading, bombing, forming hostile alliances with unfriendly neighbours or
demonising local cultures and leaders through Western media outlets. Final-
ly, dependencia theorists argued that the development industry itself serves
to maintain Western supremacy. As aid now props up many of the poorest
countries, this too can be used to exert enormous influence. Development
agencies can thus become states within states or, in the longer term, under-
mine developing economies by distorting the labour market and suppressing
the emergence of domestic skills (Dos Santos 1970).
Within the West of the late 1960s, the influence of the dependency school
was discernible in a growing tendency to question the ‘senseless reproduction
of [the] now superfluous virtues’ of liberal capitalism and an acknowledge-
ment of the ‘untruth of [its] prevailing legitimations’ (Habermas 1971: 25).
Mainstream sociological trends were, as Peter Lawler notes, ‘increasingly
Structure 37
cast as an uncritical, technological and scientistic [sic] rationalization of the
status quo’ (1995: 68). This was most strongly felt in Europe where the
postwar emphasis on positivism had not taken hold with the same potency
as in the United States. There, the purportedly ‘value-neutral standpoint of
an impartial observer [began to] give way to the subjectively open, value-
committed attitude of an interlocutor in a shared practice’ (Ingram 1987:
4). This permeated the lower echelons of the main conflict research centres
– the Peace Research Institute in Oslo (the forerunner of which was founded
in 1959), the Polemological Institute and the International Peace Research
Association (both established in Groningen in 1961 and 1963 respectively),
the Peace Research Society (International) (convened in Malmö in 1963) and
the Stockholm International Peace Research Institute (formed in 1966).
Within these organisations, there emerged a pronounced sense of dis-
satisfaction with what Lars Dencik (one of the leading writers of the new
movement) called (in a paper written in 1969) ‘conventional or conservative
peace research’ (1982: 177). Strongly influenced by dependency theory, this
group of mainly young European academics and activists accused fellow aca-
demics of being ‘an unwitting tool of American policy’ and ‘in the service of
capitalist and neocolonial interests and purposes’ (Dencik 1982: 195). Fol-
lowing a Peace Research Society (International) conference in 1967 on the
war in Vietnam, at which papers were presented predominantly focused on
game theories and security studies (thereby bypassing a more fundamental
evaluation of American involvement in the conflict), they concluded that,
in general, conflict analysis ‘has a adopted a system perspective and a value
orientation which is identical with those of the existing international institu-
tions and lies very close to those of the rich and powerful nations’ (Schmid
1968: 221). At a subsequent meeting of the organisation in Copenhagen in
1969, discontent took the form of a strongly worded petition signed by a
large number of delegates. It proclaimed that ‘conferences like the one on
Vietnam will only serve to discredit peace research’, which should, instead,
seek to offer ‘active solidarity with the peoples struggling against imperial-
ism and super-power supremacy’ (cited in Dencik 1982: 194–5).
The impact of such vociferousness was considerable. In a remarkable
tirade first published in 1970, Kenneth Boulding, President of the Interna-
tional Peace Research Association, characterised the ‘radical school’ thus:
They tend to wear beards and have a fancy for what might be called
academic guerrilla theatre. They are not inhibited by the customs of
personal courtesy which tended to characterize the older generation and
they have moral feelings which are so strong that morals are regarded as
a substitute for manners . . . . They regard the older generation of peace
researchers as obsessed by the cold war and by the necessity for resolv-
ing conflicts, as they regard this conflict essentially as no longer crucial
and the thing that interests them is how to increase conflict which is
‘objective’ but of which people are not aware.
(1982: 82–3)
38 Structure
Unsurprisingly, perhaps, such a divergence proved irreconcilable and a split
appeared between a ‘narrow’ group of predominantly American-based re-
searchers and a ‘radical’, mostly European, network of theorists influenced
by ideas of dependency. The former continued to maintain a commitment
to behavioural scientism (the Peace Research Society (International) was re-
named the Peace Science Society (International) in 1973 and moved to Cor-
nell University) and to reject ‘political action or polemical discussion’ (Boul-
ding 1977: 76). For the latter, on the other hand ‘pacifism . . . [was] replaced
by Marxism, conflict resolution by class-struggle, peace by revolution and if
necessary bloody revolution’ (Goldmann quoted in Dencik 1982: 177).
Despite the animosity with which the two factions communicated their
point of view and the polarisation that resulted, the influence of the radical
newcomers could not be fully resisted. For instance, The Journal of Conflict
Resolution, which had emerged from Boulding’s Centre for Advanced Stud-
ies in the Behavioural Sciences, announced in 1973 that, along with deter-
rence and disarmament studies, ‘the journal must also attend to international
conflict over justice, equality, and human dignity’ (quoted in Rogers and
Ramsbotham 1999: 745). It also prompted a reconsideration of ‘violence’ as
a concealed and structural, rather than overt and relational, phenomenon.
The notion of ‘peace’ was also extended to be understood as a discrete con-
cept, a pursuant set of policies and an ideal. Adam Curle’s explication of
negative and positive peace, published in 1971 and discussed in the previous
chapter, was seen as an example of such a more nuanced understanding.
The overall result of the radical challenge was that conflict analysis began
to include
structural in the sense that no specific actors are indicated, and in the
sense that for the concrete actors that happen to be performing roles in
that structure no specific motivation is necessary. [Once the structure]
. . . has started operating it is not necessary for those who are acting
within it to will all the consequences.
(Galtung 1980: 183)
This ‘seems to suggest an initial creation of structure after which the rela-
tionship between structure and action is one-way only’ (Brown 1981: 223).
Such a rejection of agential utility serves to distance Galtung’s model of struc-
tural violence from the Marxist emphasis on revolutionary action stressed by
more radical conflict analysts. In particular, it avoids the difficult proposition
that behavioural violence might be a way of bringing about a reduction in
structural violence. Indeed, given that, in one calculation, structural violence
kills over 1,000 times more people than behavioural violence each year, there
is, perhaps, some credibility to the argument that ‘revolutionary violence
against structural violence aimed at reducing the sum-total of violence in
the world should also be fairly evaluated in relation to reactionary violence
which is aimed at maintaining or increasing structural violence’ (Eckhardt
and Young 1974: 93). For Galtung, though, ‘the elimination of the members
of an oppressive elite would not necessarily end structural violence if new
forms of it were latent within the ideological perspective of the new holders
of power’ (Lawler 1995: 82). Beyond a normative commitment to pacifism
and its moral posture of refusal, then, Galtung offers no positive philosophy
of agential action. This leaves the peace analyst to define a priori the differ-
ence between the potential and the actual and the means to reconcile the
two – a sense of vagueness potentially efficacious to both liberal notions of
redistribution and Marxist accounts of structural transformation.
Indeed, the elasticity of Galtung’s terminology and taxonomy has proved
an advantage as well as a limitation, doubtless contributing to the wide
number of applications that his work has attracted. Unsurprisingly, given
Galtung’s concern with global exploitation, a number of subsequent stud-
ies have focused on the level of the international system and its structural
violence between the developed and the developing world (Farmer 2003;
44 Structure
McGregor 2003). Accompanying these has been a more recent body of
work looking at structural violence at the level of states (Jacobs and O’Brien
1998; Mazurana and McKay 2001; Preti 2002), as well as numerous stud-
ies concerned with substate issues such as individual rights and the protec-
tion of children (Farmer 1996; Kent 1999; Kostelny and Garbarino 2001).
Within each analytical framework, gender, ethnic and class axes are common
structural determinants of violence. The following section will outline some
examples of the ways in which various conceptions of structural violence
have informed case studies at different levels of analysis.
Soon after the split in the peace research community, Galtung, in keep-
ing with his commitment to empiricism, sought to enumerate structural
violence in an arithmetical essay (co-authored with Tord Høivik) published
in 1971. Later, Høivik (1977) and other writers, such as William Eckhardt
and Christopher Young (1974), attempted to operationalise more complex
quantifications of those whose lives have been shortened by inadequate ac-
cess to a range of otherwise available resources. Working on the assumption
that there is no innate deficiency within the physiology of the poor and thus
no reason why they cannot achieve the same lifespan as the world’s wealthi-
est societies, analysts compared the violence of structures with the violence
of behaviour – an idea illustrated in Table 3.1. In what is arguably the best
articulated study of its kind, Gernot Köhler and Norman Alcock used two
methods to ‘present some estimates of the fatal consequences of structural
violence in global society’ (1976: 343). First, in the Swedish model, they
took the country with the highest life expectancy as a benchmark for the age
Structure 45
Table 3.1 Indicators of violence
Violent input Violent output
Direct violence Deployment of armed People killed by war
men, shelling, bombing,
etc.
Structural violence Malnutrition, lack of People killed by a lack of
shelter, health care, etc. necessities
Source: Adapted from Köhler and Alcock (1976: 343).
(or potential) that people could achieve. This was fixed at Sweden’s 1965
average of 74.7 years (Es). They then took the country with the lowest life
expectancy (again using data from 1965). This was Guinea’s average of 27
years. To calculate Guinea’s level of structural violence, they divided the
country’s overall population (3.5 million) by 27 to give the number of Gui-
neans who actually died in 1965 (129,630) and by 74.7 to give the number
of Guineans who would have died in 1965 had they enjoyed the same life
expectancy as Swedes (46,854) – the difference is the number of people who
died as a result of structural violence in 1965 (129,630 – 46,854 = 82,766).
Second, in the egalitarian model, Köhler and Alcock estimated the effects
of a complete and equal redistribution of the world’s resources. To do this,
they plotted every country’s gross national product (GNP) on a scatter-gram
(Figure 3.1), before adding each of these together to give 1965’s gross glo-
bal product (the entire world’s economic product) which, once divided by
the population of the planet, could be used to calculate a per capita figure
of $651 (g*), equating to a life expectancy of 68.3 years (c*). Using this
figure, the number of deaths due to structural violence in Guinea came out
at 78,385. As it produces slightly lower results, the egalitarian model was
then run throughout the world giving a total of those killed due to structural
violence for 1965 of 14 million – the figure for the Swedish model was 18
million – amounting to a shortfall of over 300 million life–years. Of this,
480,000 lives (or 3.42 per cent) were lost in the global North compared
with over 96 per cent in developing countries. Both figures far outweighed
the numbers of people killed as a result of warfare for 1965 – 2,207 in the
North and 113,000 in the South.
An important aspect of the Köhler and Alcock egalitarian model is the
position of its predicted life expectancy – 68.3 – in relation to existing mor-
tality ranges. While it is more than 40 years higher than in Guinea, it is less
than six years lower than in Sweden. This leads the authors to conclude that
‘under conditions of complete global equality, the rich countries would lose
only minor amounts of life expectancy, whereas the poor would gain tremen-
dously’ (1976: 355). So, for global equality to come about, life expectancy in
the North would have to drop only by just over five years. Furthermore, Fig-
ure 3.1 reveals that, for the North, economic growth is associated with only
a limited increase in life expectancy, whereas for countries in the $40–600
46 Structure
90
‘South’ ‘North’
population= population=
80 2+ billion 1+ billion
Japan Sweden
Es France
Greece
c* 70
USSR USA
Brazil Jamaica
Life expectancy (E) (years)
60
China Syria
50
India
40
Nigeria
30
Guinea
20
10
0 1000 2000 3000 4000
g*
Gross National Product per capita (G/P) (in 1965 U.S. dollars)
Figure 3.1 An empirical table of structural violence. Source: Redrawn from Köhler
and Alcock (1976: 353).
per capita range, an increase of 7.68 per cent in GNP correlates to a rise in
life expectancy of one year. ‘In other words, wealth cannot only buy a higher
standard of living, it also buys life itself ’ (Köhler and Alcock 1976: 355).
Although similar types of approach based on the value of economic
growth were often hailed as potential ways of measuring social develop-
ment empirically (Galtung et al. 1975), more differentiated methods began
to emerge during the 1980s and 1990s (summarised in Alkire 2002). As it
became increasing clear that rises in GNP were not being converted into
quantifiable forms of social development, per capita measures of economic
wealth came to be regarded as especially crude. By 1992, for instance, life
expectancy in Sri Lanka had reached 72 years on the basis of a per capita
income of only $540 per annum, while Gabon’s yearly per capita income of
$4,450 was producing an average life expectancy of only 54 years (Sen 1998:
10). It therefore became common for studies to emphasis intrastate factors in
examining development in general and conflict variables in particular. This
was, not least, due to the growing cost of warfare to civilian populations and
its accompanying media attention. Since the Second World War, for instance,
over 170 million people have been killed by their own governments and, of
these, non-combatants have, as Chapter 1 outlined, absorbed ever increasing
proportions (Sivard 1993: 20; Rummel 1994).
Consequently, academic interest began to focus on axes of structural
violence within states. It was argued that certain groups could be structur-
ally repressed, yet individuals from those groups could still be immensely
powerful. An example of this might be the gendered nature of structural
Structure 47
violence in Pakistan where poor health care and other social factors have
combined to reduce the country’s overall female population by almost 1 per
cent (1993 data) at a time when Benazir Bhutto was beginning her second
term as prime minister. Similarly, the Afro-American community, who have
a mortality rate almost double that of the white majority, is often regarded
as being subject to extensive structural violence despite recently supplying
two of the state’s most powerful leaders, Condoleezza Rice and Colin Powell
(Sen 1998: 10–17). These prominent exceptions combine with the banality
of social exclusion to make such structures hard to recognise (although they
are rarely hidden – rather, they are ‘right before our eyes and therefore
simply taken for granted’ (Scheper-Hughes 1996: 889)). Such habituation
is grounded upon the capacities of ideologies, languages, arts and sciences
– the aspects of what Galtung identified as ‘cultural violence’ – to make
exploitation look and feel right (1990: 291). For instance, the structural vio-
lence of being forced to swear allegiance to the monarch in order to take up
a seat in the British parliament or to apply for UK citizenship is sufficiently
obscured by ritual and tradition to appear superficially innocuous, yet it has,
for many years, served both to exclude Irish republican parliamentarians and
to present a considerable obstacle to religiously observant Muslims.
The ordinariness of structural violence is well illustrated by Peter Uvin’s
study of Rwanda (1998: 103–60). Reviewing a series of World Bank reports
during the run-up to the 1994 genocide, he finds that, by relying on of-
ficial data and the stated aims of legislation rather than academic research
and actual policy implementations, they were able to reach the mistaken
conclusion that ‘land is less unequally divided than elsewhere, . . . house-
hold expenditure is relatively evenly distributed in Rwanda, and . . . govern-
ment expenditure and tax policies are income neutral’ (cited in Uvin 1998:
110–11). In reality, ‘the privileged of the regime’ had been able to garner aid
rents of enormous proportions leading to ‘the constitution of large pastoral
domains’ under their clients’ control and landlessness as high as 75 per cent
in some regions (International Fund for Agricultural Development 1992:
8; Pierre Erny cited in Uvin 1998: 114). The patrimonial nature of urban
politics produced comparable disparities in income levels with the wealthiest
decile in Rwanda increasing their share of the country’s income from 22 per
cent in 1982 to 52 per cent in 1994 (Jeff Maton cited in Uvin 1998: 115).
Despite the fact that the great majority of those benefiting in this way
were Hutus and that much of the resources that the Hutu political elite (a
third of whom were drawn from the President’s own province of Gisenyi)
were distributing came from international aid donors, spending patterns
continued to favour ‘the smallest groups in society, that is, the richest 1 per
cent or so, composed of technical assistants and their “homologues”, plus
merchants and high-level government officials’ (Uvin 1998: 143). In many
ways, this is closely in keeping with Galtung’s notion of ‘exploitation’ as the
causal mechanism behind his propositions of structural violence. According
to Uvin’s analysis, much of the international development effort in Rwanda
48 Structure
worked with domestic political elites to institutionalise Western ‘penetra-
tion’ (defined by Galtung as ‘implanting the top dog inside the underdog’),
‘segmentation’ (‘giving the underdog only a very partial view of what goes
on’), ‘marginalisation’ (‘keeping the underdogs on the outside’) and ‘frag-
mentation’ (‘keeping the underdogs away from each other’ (1990: 294)).
So, while donors were insulated by a moralising afflatus of philanthropy in
which power relations were comprehensively ignored and their response to
‘ordinary’ communal violence in Rwanda was simply an apolitical ‘mirror
image of need’ (Duffield 1999: 32), the actual effect of development policy
was a growth in domestic exploitation and structural violence.
Here, then, structural violence may be a facilitative or a linear causa-
tive element in behavioural violence. Studies that focus on the latter tend
to consider behavioural violence to be a response to the inequalities and
exploitation that underpin structural violence. As we shall see in Chapter 7,
these frequently build on theories of human needs (particularly in terms of
the frustration of material wants, civil rights and the deprivation of higher
social requirements) to suggest that structural violence contains the intrinsic
capacity to provoke behavioural violence from the deprived group (Khan
1978: 836). In terms of the former, the potential of structural violence to
lead on to direct bloodshed is, in contrast, contained within its capacity to
create an atmosphere of ordinariness through which elites can both mobilise
their clients and prosper politically – a capability related to both diversionary
theories of rule (dealt with in Chapter 4) and the construction of social iden-
tity (the topic of Chapter 6). Uvin argues that the subjugation of non-Hutus,
for example, lowered social barriers to greater tyranny, thereby preparing
the ground for behavioural violence on a grand scale. He concludes that
Conclusion
The notion of structural violence emerged during the late 1960s largely as a
response to a perceived failure of academia to take up a position critical of
contemporary Great Power policy – a situation perhaps similar, as Chapter
1 suggested, to that of today. It sought to expand definitions of conflict be-
yond apolitical and scientistic studies which tended to avoid moral concerns
and thus, in the view of some, served to endorse the subjectivist preferences
of the powerful. Violence may thus be psychological as well as physical, it
may be contained within rewards and not simply punishments, and it may
be present even though someone is not hurt and there is no subject-to-object
relationship. It may also emerge from non-violent intentions, be latent as
well as manifest and include many of the results of the international system’s
normal operation. Exerted at the level of the structure and not simply the
individual behaviour of aggression and warfare, violence may be regarded as
present whenever damage is done to a person’s potential.
Structural violence thus represents one of the key conceptual advances in the
study of conflict. It has influenced, and continues to influence, a considerable
50 Structure
amount of important analysis with a wide range of concerns and approaches.
Its opposition to the staid empiricism and uncritical pacifism so often found
within the canonical establishment of peace studies remains a pertinent and
powerful challenge today. Yet, in many ways, it has succeeded in straddling
both methodological and ideological divides. Its elasticity clearly appeals to
proponents of reasonably conservative policies of redistribution as well as to
those who espouse more radical transformative agenda. This is evident in its
antecedents, in its theoretical focus, in the critical responses it has attracted
and absorbed and in the various applications it has provoked or informed.
These include studies at the global, national and individual levels of analysis.
The first reveals mortality differentials resulting from low life expectancies in
developing countries and quantifies what might be required to redress such a
massive loss of human potential, the second casts light on the way in which
the ground might be prepared for behavioural violence (including incidents
of an unprecedented scale), and the third helps to increase our understand-
ing of the impact of socio-economic stratification on the lives of children.
Structural violence is thus an arithmetic tool, a conceptual framework and a
concrete social phenomenon. It offers a way of quantifying exploitation and
inequality at various analytical levels. Conceptually, it can be understood as
a result of latent conflict and, in its capacity both to facilitate and to provoke
conflict, as a cause of behavioural violence. As an issue for modern society,
it has helped to maintain attention on discriminatory axes of class, gender,
race, age and culture.
4 Functions
Evaluating conflict
During the 1950s, writers began to question the generally held view that
conflict (in both its violent and non-violent forms) was a universally de-
structive feature of human life and one to be avoided or minimised. Writers
such as Rolf Dahrendorf offered a challenge to the predominant view of
52 Functions
society as structured by a tendency towards equilibrium which, as Table 4.1
illustrates, tended to see conflict as an aberration or temporary pathology
(1959: 159–61). Indeed, ‘the equilibrium model, by stressing integration
and consensus, leads directly to the position that conflict, which is a threat to
stability, must be curtailed in order for the integration of the social structure
to be maintained’ (Stohl 1976: 7–8). As it is the state that is responsible for
the maintenance of order in society, such a position is inherently conserva-
tive. Dahrendorf ’s alternative model, in contrast, tends to view equilibrium
or passivity as abnormal and born of repression. Here, conflict is regarded as
a vital part of societal change.
So, with this in mind, the primary question becomes how to measure the
effect of conflict on society rather than simply how to reduce its impact.
Writers started to classify conflicts as ‘functional’ if their benefits outweighed
their costs and ‘dysfunctional’ if their costs outweighed their benefits. Typi-
cally, four groups have been identified (Mitchell 1980: 71–2). The first are
those who gain from the existing distribution of values and resources. These
may, for instance, be large landowners in South America who have an in-
terest in preventing certain types of conflict. Their interest is in the status
quo. In order to protect this, they may engage in a variety of conflictive be-
haviour including coercion, threats and compromise. The second are those
who lose from the existing distribution of values and resources. To continue
the previous example, agricultural workers in South America who wish to
amend their position may also engage in conflictive behaviour in order to
effect this change. A third group consists of those who do, or will, derive
benefits from a conflictive situation. This may include a range of actors, from
glaziers in Northern Ireland prospering from bomb damage to organised
crime syndicates in south-east Turkey who use Kurdish activism to traffic
narcotics. A fourth group are those who are directly, or indirectly, harmed
by a conflictive situation. This might be the deceased and bereaved or (often
utterly unrelated) actors such as employees in the Turkish tourist industry
whose livelihoods were damaged by the conflicts in Iraq.
Using these categories to evaluate conflictive behaviour is, however, prob-
lematic. The fact that they are not mutually exclusive – actors are likely to
Subscribing to this view, Mitchell puts forward four ways in which inter-
est conflicts may benefit society as a whole (1980: 64–5). The first is that
they establish numerous patterns of crosscutting antagonisms and coalitions.
These tend to prevent the emergence of fewer, more acute divisions ground-
ed upon overlapping value-based differences. The second is that the pres-
ence of interest conflicts helps to offer members of society a ‘safety valve’
through which grievances can be legitimately channelled without disrupting
the normal operating patterns of the system itself. Third, conflicts can es-
tablish or promote mechanisms for managing grievances and disquiet. This
may be either through the creation of new institutions and measures, such
as the appointment of a commission of inquiry, or by emphasising existent
norms such as campaigns to improve road safety, food standards or public
accountability. Fourth, conflicts may create relationships, or improve aware-
ness, between previously disassociated groups. Although this may involve
Functions 55
some initial costs, benefits in the form of enhanced interaction may outlast
the conflict by many years. For instance, periodical race riots in British cities
have helped to highlight the institutionalised prejudices that exist within
sections of the United Kingdom’s police force and have arguably led to an
improved relationship between minority groups and the constabulary’s sen-
ior management.
Conflicts of value may also be regarded as containing functional facets
if a quantifiably ‘better’ future can be expected to ensue. Such ideas can be
broadly separated into three categories. First, from a liberal international-
ist perspective, many people would argue that the devastating conflict with
fascism in the 1930s and 1940s was necessary and, therefore, broadly func-
tional as it led to an international system in which fascism is no longer a
major part. In addition, it could be said to conform with Mitchell’s first and
third points above, in the sense that the new world order that emerged from
the Second World War replaced the polarisation of the interwar period with
numerous crosscutting relations and helped to establish multilateral regimes
such as the United Nations, the EU and normative conventions concerning
human rights, refugee protection and so on.
Second, conflicts of values can be seen to underpin dialectical visions of
society and the ways in which different systems and structures have emerged
over time. Both Hegel and Marx regarded conflict as a functional way for
societies to attain greater sophistication and development through trial and
error. Each saw conflict between different theses, or classes, as important in
perceiving the ‘real’ nature of modernity and in moving towards a superior
synthesis in which existent inequities and stratifications are absent. Third,
writers from a socio-biological perspective have suggested that conflict is
an effective means of reducing demographic pressures. This is derived from
the economist Thomas Malthus (1766–1834), who argued that exponential
increases in human populations born of the improvements in public health
provided for in the industrial society could be offset (though not entirely
thwarted) by the ‘positive checks’ of pestilence, famine and warfare. Here,
collective violent conflict has replaced the predation threat of large carni-
vores. Although all human beings belong to a single species, non-biological
differentiations such as religions, languages, cultures, national identities and
so on are, as Chapter 5 will show, held to constitute ‘quasi-species’. In killing
each others’ members in war, these are deemed to self-limit human impact
on the ecosystem (Hutchinson 1965).
Indeed, it is perhaps at this level of the substate group that most attention
has been focused upon the functions of conflict. Five basic ideas predominate
– all of which may be considered alongside Mitchell’s triangular model of
mutually interactive conflictive attitudes, behaviour and situations looked at
in Chapter 2. First, participating in conflictive behaviour may, as Raymond
Mack has noted, help to coalesce groups by making individuals aware of
their shared goals and mutual interests (1965: 335). Afro-Americans’ resist-
ance to racism has, for instance, helped to extend notions of black identity
56 Functions
across the United States. Second, engaging in conflictive behaviour can in-
crease the critical focus on views that are held within the group, thereby
bringing problems to the surface. In this way, group structures can become
more egalitarian through a shared commitment to a cause. Examples of this
include the enhanced capacity of female insurgents in Guatemala and Turkey
to challenge traditional patriarchal hierarchies and the enfranchisement of
British women following the boost that the First World War gave to the
suffrage movement. Third, conflicts may increase motivation levels within
groups, which may then stimulate industry, solve technological problems
and lead to innovative thinking. The Second World War was, for example, a
period of both scientific advance and very high per capita outputs. Fourth,
the experience of conflict may oblige parties to establish contact with oppos-
ing groups, to modify previously unobtainable goals or to assess and gain a
greater understanding of their, and others’, goals – thereby ending a stalled
relationship (Mitchell 1980: 62–3).
First, the threat and danger come from outside and the causes can clearly
be seen and specified. Second, the immediate needs are clearly recog-
nizable, and direct action can be undertaken with discernible results.
Third, all are affected indiscriminately and thus the danger and suffer-
ing become public phenomena equally shared. The resultant solidarity
eliminates even social distinctions.
(1976: 151)
For Stein, the elimination of social distinctions does not imply the presence
of the kind of societal equilibrium discussed at the start of this chapter. If we
take Ernst Haas’ definition of cohesion as ‘the likelihood of internal peace-
ful change in a setting of groups with mutually antagonistic claims’, then a
group subject to the integrative effects of conflict engagement would remain
fundamentally conflictive (1961: 367). In assessing this situation, it is useful
to distinguish between attitudes and behaviour by returning to Mitchell’s
triangular schema presented in Chapter 2. His model suggests that a con-
flictive situation can lead to a change in attitude whereby individuals, first,
seek security and prestige in identifying with groups and, second, tend to
experience personal challenges as a threat to the values of the group as a
whole. This cohering effect is thus likely to impact upon conflictive behav-
iour within the group. Of the three types of conflictive behaviour that he
identifies – coercion, persuasion and compromise – group members are thus
more likely to select a combination of the last two during periods of exter-
nal conflict. Bearing in mind Schmid’s point, elaborated in Chapter 3, that
persuasion and compromise tend to be strategies preferred by those seeking
a return to a favourable status quo, such an effect clearly has a number of
potential benefits for leaders wishing to institutionalise a pacific domestic
environment. Indeed, in order to ensure in-group integration, leaders may
58 Functions
‘actually search for enemies with the deliberate purpose, or the unwitting
result, of maintaining unity and internal cohesion’ (Coser 1956: 104). As
James Madison told Thomas Jefferson in 1798, it is, perhaps, ‘a universal
truth that the loss of liberty at home is to be charged to provisions against
danger, real or pretended, from abroad’ (quoted in Smith 1995: 1048).
For the philosopher Leo Strauss (1899–1973), whose students included
the architects of the recent invasion of Iraq, Paul Wolfowitz and Abram Shul-
sky (The New Yorker 5 May 2003), liberalism required politicians to iterate
‘noble lies’ in order to give people moral certainty in a system which, in el-
evating individual liberty to the zenith of human achievement, contained an
intrinsic tendency towards relativism. In a reworking of Nietzsche’s ‘deadly
truths’, he argued that leaders – those able to comprehend the deeper mean-
ing in philosophers’ deliberately esoteric erudition – had a responsibility
to inculcate group identities in their followers in order to prevent a liberal
society from descending into Huxleyan vacuousness and chaos. In fact, he
went further and questioned the distinction between values and facts that
underpins post-Enlightenment rationalism, thereby implying that all claims
of objective truth are self-delusional and dangerous (1965). Such a vision
of political leadership has proved to be highly influential not only for con-
structivists and post-modernists who, as Chapter 6 notes, have elaborated
accounts of how the manufacture of conflictive environments takes place,
but also for a generation of American neo-conservatives, such as William
Kristol (the former chief of staff to Secretary of Education William Bennett
and Vice President Dan Quayle), and the present Deputy National Security
Advisor, Elliot Abrams, both of whom studied under Strauss’s protégé, Har-
vey Mansfield, at Harvard.
The influence of Strauss can be seen in Mary Kaldor’s study of what she
calls the ‘imaginary’ Cold War (1990). Here, the leaders of the two blocs
of ‘Stalinism’ and ‘Atlanticism’ used their huge conventional and nuclear
arsenals to control domestic political life and marshal economic resources
in ways that would not have been tolerated without a war mentality. It was
born, she argues, of the need for a new unifying principle of action following
the Second World War. The Soviet threat helped Western leaders to discredit
isolationism as well as to repress domestic leftism. The maintenance, design
and deployment of both sides’ weaponry also stimulated economic growth
and secured party loyalties. Kaldor’s claims have, more recently, received
support from the partly autobiographical accounts of President Carter’s
National Security Adviser, Zbigniew Brzezinski (1997), and former United
States Ambassador to the Soviet Union, George Kennan (1996).
An associated body of work puts forward the idea that external conflict
increases the internal coherence of alliance structures – one of the key focus-
es of Chapter 9. George Liska, for instance, notes that ‘reciprocal pressures
between roughly equal alliance systems tend to consolidate both’ (1962: 26).
Terry Hopmann builds on this work by hypothesising that ‘the greater the
East–West tension, the greater the degree of cohesion within the Communist
Functions 59
system’. He concludes that, during the premiership of Joseph Stalin, the
eight core countries of the Soviet-led bloc (China, Albania, East Germany,
Poland, Hungary, Rumania, Bulgaria and Czechoslovakia) were relatively
well integrated behind a structure of Politburo control based upon percep-
tions of an acute American threat. Following Stalin’s death in 1953 and the
Geneva Conference in 1955, however, a greater degree of diversity within
the alliance’s view of the United States emerged. He suggests that a possible
way to view Moscow’s decision to intervene in Hungary and Poland in 1956
is as a means of solidifying intra-alliance unity through the heightening of
East–West tensions. Indeed, strains placed on the Sino-Soviet relationship
following the latter’s signature of the 1963 Test-Ban Treaty did appear to
be considerably alleviated by the United States’ decision to begin bombing
North Vietnam in February 1965 (1967: 216).
100
65
55
Satisfaction (%)
45
35
25
0
1979 1980 1981 1982 1983 1984 1985
Year
Figure 4.1 The British public’s satisfaction with the Conservative government 1979–
85. Source: Redrawn from Norpoth (1987b: 953).
Functions 61
ing from 47 to 44 per cent and from 39 to 38 per cent, respectively, and
President Mitterrand enjoying a modest rise of one point from 58 to 59 per
cent during this period. In the United Kingdom, on the other hand, 61 per
cent of those questioned named the Falklands crisis as ‘the most important
problem facing Britain today’ with only 25 per cent citing unemployment
(The Economist 8 May 1982). By 26 June, over 250 British troops were
dead and more than one billion pounds had been spent, yet the same survey
found that 76 per cent of the British public remained in agreement with
government policy over the issue. While this does not, of course, prove that
involvement in the war caused the rise in positive public opinion, nor that
the British government undertook this conflict in order to gain popularity or
to divert attention from the miserable economic climate, numerous writers
have concluded that the conflict helped to secure a Conservative victory in
the 1983 elections and that, in the eyes of many, Mrs Thatcher ‘emerged as
a remarkable war leader’ (Hastings and Jenkins 1984: 355–6). As such, the
result of the Falklands/Malvinas crisis offers some support for the general
comment that ‘a fully specified theory of the causes of international conflict
will require an understanding of the role domestic politics plays in foreign
policy decisions’ (Morgan and Bickers 1992: 26).
Constructing such a general theory has, however, proved extremely dif-
ficult. Successive quantitative studies have failed to establish a systematic link
between internal political problems and the external use of force. An analysis
of nine indicators of domestic strife and of 13 indicators of external conflict
undertaken by Rudolph Rummel during the early 1960s, for instance, con-
cluded that ‘foreign conflict behaviour is generally and completely unrelated
to domestic conflict behaviour’ (1963: 24). Similar results were arrived at
using various methodologies over the next ten years or so (Tanter 1966;
Wilkenfeld 1972). More recently, though, these studies have been brought
into question. Clifton Morgan and Christopher Anderson, for instance, note
that, in many cases, their findings were based on an inability to establish a
clear linear relationship. As the previous section outlines, however, group
cohesion cannot be simplistically related to participation in external conflict,
and thus the search for such a strong association is, almost inevitably, futile.
If, for instance, the integrative effects of conflict are believed to vary across
different groups and group constituents, one might expect leaders not to en-
gage in external uses of force if internal cohesion is very weak. Furthermore,
if it is assumed that already well-integrated groups would not be significantly
affected by conflict participation then, again, leaders would not be expected
to deploy such a strategy. The external use of force may therefore be more
likely to occur when internal dissent is moderate, producing a curvilinear,
inverted ‘U’ relationship rather than a linear correlation (1999: 801).
Consequently, many writers have argued that leaders’ propensity to choose
to distract dissatisfied groups by using force externally on another party, as
opposed to using the more obvious methods of compromise or repression,
is heavily reliant on the type of regime in which they operate (Bueno de
62 Functions
Mesquita and Siverson 1995). The greater autonomy of authoritarian lead-
ers, plus the absence of significant reform pressures may, for instance, make
diversionary conflict easier to instigate. Equally, the institutional restraints on
democratically elected leaders may mean that the state elite is more likely to
acquiesce to the demands of dissenters. Alternatively, it is frequently argued
that both types of leadership structure rely on support coalitions, a stable
economy and a pacific workforce for survival and thus have a comparable
incentive for diversionary conflict (Downs and Rocke 1995). In contrast,
other writers, such as Christopher Gelpi, suggest that the relative ease with
which authoritarian leaders can repress internal dissent, coupled with the
less extensive need for a ‘rally-around-the-flag’ effect, means that democrati-
cally elected leaders will be more likely to use diversionary tactics (1997:
260–1). This is especially so after internal dissent has become sufficiently
acute for democratic leaders to regard the granting of concessions as a pos-
sible incentive for others.
It has also been noted that the decision to embark on a diversionary con-
flict rests, first, on the degree to which leaders can be certain that such a
course of action will lead to the clear identification of an internal enemy
among the public. As some dissenting factions within states may regard an
external foe as an ally in their struggle, this policy will only be effective ‘if
domestic actors view the foreigners as worse than their domestic antagonists’
(Morgan and Anderson 1999: 803). Second, leaders must believe that the
identification of an internal enemy will solidify their own position in relation
to their constituents. As the allegiance of the entire polity is neither pos-
sible nor necessary, however, this rests on the perceived relative importance
of the leadership’s allies. A leader with a large parliamentary majority may
not, for instance, be overly anxious at the dismay of opposition parties, but
would probably view Cabinet fractures over their choice of external enemy
with more concern. Thus, a diversionary strategy becomes especially likely
if the leadership support base is perceived to be under threat. Third, leaders
will not embark upon any hostile act simply because it offers the potential
to divert. The expected utility of an external conflict is a vital considera-
tion. It must, for instance, be eminently winnable without the danger of a
potentially costly or prolonged engagement. Conversely, though, it must be
sufficiently high profile to have an impact on the target dissenters within the
leadership’s support base.
The diverse nature of leaders’ internal constituencies brings us to another
problem with much of the statistical literature – its failure to acknowledge
that approval for war leaders tends to be short lived (Cotton 1986). The
state’s extraction of soldiers and taxes is frequently resisted by elites and
unpopular with the masses. Furthermore, a focus on external conflict may
reduce a government’s capacity to repress its opponents, leading to greater
internal conflict and instability – particularly when coupled with the de
mobilisation of large numbers of combatants commonly troubled by feelings
of social dislocation. In many cases, these factors have led to the downfall
Functions 63
of governments. It is not without reason that warfare is often called the
midwife of revolution. The disastrous involvement of American forces in
Vietnam, for instance, is reckoned to have cost President Johnson over 20
percentage points (Mueller 1973: 196). Indeed, in the case of the Falklands
War, it is quite clear that
They then proceed on the basis of seven presidential decision premises de-
rived from the domestic political environment. These are outlined in column
one of Table 4.3. Column two sets out Ostrom and Job’s prediction of each
premise’s impact on the executive’s decision-making processes. Correlating
these to the 226 occasions on which the United States has deployed the
‘political use of major force’ overseas, Ostrom and Job find support for the
first premise, concluding that public perceptions of international tension are
‘negatively related to the probability of a political use of force’ (1986: 557).
64 Functions
Table 4.3 American presidential decision-making over the political use of force
Expected impact on presidential decision-making
Decision premise processes
Public attitudes to Public concern over international tension will lower the
international tension president’s propensity to use force
Public attitudes to the Public perceptions that the United States is becoming
strategic balance disadvantaged by changes in the strategic balance will act
as an incentive for the president to use force
Public aversion to The president is unlikely to use force following American
warfare involvement in warfare
The condition of the Deteriorations in the domestic economy will encourage
domestic economy the president to use force as a diversionary measure
The perceived level of The president will be more likely to use force if he
public support perceives himself to have a buffer of support
Past political success A decline in presidential success will make him more risk
acceptant and thus more likely to use force
The current position in The propensity to use force during important periods in
the electoral calendar the electoral calendar
In terms of the second premise, no relationship was found between the stra-
tegic balance and political uses of force. Both the third and fourth premises
did, however, find support, leading Ostrom and Job to note that war weari-
ness and periods of economic prosperity acted as inhibitors on presidents’
use of major political force between 1946 and 1978 – matters that we shall
return to in Chapter 9.
Most significant in their statistical analysis were the fifth and sixth premis-
es. While declines in presidential backing almost always occur during an
incumbent’s time in office, the use of force was found to transpire most com-
monly when popularity levels were waning, yet a buffer of support remained.
Taken together, however, ‘the more negative the president’s overall record
in office (as represented by declining popular support), the more likely he is
to act in the absence of a popularity buffer’ (Ostrom and Job 1986: 557–8).
The final premise was supported, in that the overseas use of major political
force was more likely to occur in the final quarter before an election than
in any other quarter of the electoral year. These findings are important as
they offer a direct challenge to the realist understanding of state behaviour
as primarily grounded on the pressure of international relations as determi-
nants of domestic politics. As Patrick James and John Oneal point out, if ‘the
desire to divert attention from a troubled economy is a greater influence on
the decision to use force than the level of international tension, assumptions
regarding the national interest are clearly undermined’ (1991: 308).
Question marks do, however, remain over the transferability of these
results. Bruce Russett has, for instance, noted that, as the success of diver-
sionary measures relies on a rapid victory, more powerful states are more
likely to engage in such actions than those with weaker military capabilities
Functions 65
(1989). Extrapolating from Ostrom and Job’s state-level study to the level
of the international system without controlling for this would introduce a
serious bias into the analysis (Levy 1993: 277). Moreover, their finding that
a buffer of support during periods of declining popularity encourages the
presidential use of force has been countered by realists. Hans Morgenthau
and Kenneth Thompson have argued that this is not a domestic variable at
all (1985: 158). They suggest that national support is primarily an indica-
tion of leaders’ capacity to act internationally. Here, then, internal support
is construed as a means of strengthening national security and the ‘search
for policies which will maintain the leader in power against domestic op-
position’ (Bueno de Mesquita and Siverson 1995: 853). Such a concession
does, however, confirm the broadly non-realist notion that ‘decisions regard-
ing international conflict are, as with any policy decisions, affected by the
pushing and pulling of competing domestic interests’ (Morgan and Bickers
1992: 26).
Conclusion
This chapter looked first at issues related to the evaluation of conflict. In the
first section, it traced the various means through which these have been ap-
proached over the years and how different methods have informed the way
that the impact of conflictive behaviour on group formation, structure and
dynamics has been assessed. It contrasted these approaches with the equi-
librium model and with simplistic categorisations of conflictive behaviour
as either functional or dysfunctional, suggesting that, instead, engaging in
conflict may involve both types of outcome depending on who is analysed,
when and by whom. Apart from the obvious costs of conflictive behaviour,
various material, political and psychological benefits have been identified.
These may be at the level of the system (in terms of driving forward change
and reducing population pressures), the state (through the consolidation of
alliance structures), the group (as a means of cohering hitherto disparately
organised elements) and, in providing both an emotional catharsis and a way
of responding rationally to duress, the individual.
In the second section, one of these functions – the idea that conflict par-
ticipation can cohere or integrate social groups and systems – was taken for-
ward and explored in more detail. Evidence from evolutionary science was
discussed in the light of an assortment of well-known anthropological studies
with the aim of developing a nuanced comprehension of this proposition. It
was argued that existent structures, previous experiences and socio-historical
trajectories are important in understanding the changes in behaviour that
commonly result from, or coincide with, the impact of conflict engagement
upon groups, alliances and systems. Here, the presence of structural violence
can, as Uvin’s study of Rwanda examined in Chapter 3 illustrates, provide a
facilitative environment for social differentiation to occur. Galtung’s model
of cultural violence is particularly apposite here, as a means of revealing the
66 Functions
banality of scapegoating ‘others’ so that groups may feel more ‘self ’. As we
saw in Chapter 3, the intensification of group delimitations can rely on the
exclusionary organisation of religious, gender, class, racial or political identi-
ties and can be underpinned by the, frequently subliminal, use of discrimina-
tory communication media. Here, the capitalist mode of production and its
emphasis on value maximisation, competitiveness and the ‘naturalness’ of
unequal production factors may help to normalise the ‘noble lies’ identified
by Strauss (1965). A similar function might be ascribed to the production of
knowledge through science and its power to confer legitimacy upon those
elites for whom conflict and violence is efficacious.
The third section focused on the idea that conflictive behaviour stems
from political elites’ need to divert attention from the domestic arena by
pursuing aggressive foreign policy objectives. At its broadest, this body of
work merely seeks to establish a connection between domestic and interna-
tional state actions as a challenge to realist notions of international primacy.
In some formulations, however, direct causal linkages are deemed to exist
between government unpopularity or internal dissent and the deliberate pur-
suit of crises or even warfare abroad. A widely cited example of this is the
British administration’s handling of the Falklands/Malvinas crisis in 1982
and its failure to prevent the dispute from escalating to war. Much clearly
depends on the type of regime in which leaders operate. In many cases,
brinkmanship might serve a similar purpose to military engagement and may
thus be preferred. Yet, the notion that elites are prepared to provoke con-
flicts in an attempt to obfuscate and cause their citizens to rally around the
flag remains an intuitively appealing and important area of research.
5 Innate
A key element in both the creation of enhanced in-group coherence and the
legitimisation of aggressive foreign policies considered in the previous chap-
ter is the idea that there is a scientific basis to ethnic or national solidarity
born of the immutably conflictive nature of the human condition. This chap-
ter argues that, within the current landscape of conflict analysis, there are
two basic facets that build on these premises – constituting the two sections
to follow. The first is a focus on the biology of the human body, its evolution-
ary origins as well as its neurological, psychological and hormonal functions.
With its origins in nineteenth-century scientism and the development of the
experimental method, this corpus of work is often associated with studies of
animal behaviour, physiology and psychiatry and tends to focus on human
phenomena as a branch of sensate life. The second set of ideas promulgat-
ing the immutable nature of the human propensity for violent conflict is
grounded upon the premise that deeply rooted aspects of human experience
– namely culture, history and religion – are relatively constant determinants
of human behaviour. It has been particularly deployed to cast light on very
long-term conflicts of apparent intractability in which language, ethnicity
and human rights play a part. In many ways, these two approaches represent
not so much theory as such, but rather a disparate, at times quite specula-
tive (albeit extremely popular and widely read), spectrum of opinion. The
unifying, or at least common, features which these views share are, first, an
assumption that violent conflict is an inevitable part of human interaction,
second, that it can only be imperfectly prevented rather than avoided alto-
gether and, third, that the root causes of conflictive behaviour are in natus,
or present at birth, and are thus to be found in unchanging, or very slowly
changing, features of human development or physiology. Conflict is, in other
words, inherent; ‘the potentiality for it always exists and actuality can only
be obstructed’ (Eckstein 1980: 139).
68 Innate
Socio-biology
The attribution of social ills to the fixed nature of humanity is far from new.
Basing their work on the writing of Saint Augustine and others, Christian
theologians have long promulgated the idea that humanity has inherited the
original sins of Adam and Eve, the only two people born free of sin. People
are thus not sinners because they sin, but rather people sin because they are
sinners – or as John Calvin put it in the sixteenth century, humanity’s ‘whole
nature is, as it were, a seed of sin’ (1949: 8). In many ways, this view of hu-
man behaviour as immutably prone to, among other dysfunctions, violent
conflict captures the essence of this chapter’s topic. While the various ideas
and approaches discussed below mostly have their origins in the less exegetic
concerns of the nineteenth century, they share Christianity’s tenet that, at
their most fundamental, the causes of violence are to be found within human
beings and not within the social dynamics of human interaction. Moreover,
in policy terms, efforts to end the human propensity for violence are, for
both old and new accounts, largely futile. So, although evolutionary change,
ethological instincts, psychological drives and fixed socio-cultural impera-
tives have replaced divine ordinance and creationism as the primary determi-
nants of individual action, the best that can be hoped for is that dysfunction-
ally conflictive behaviour may be temporarily arrested by ritual, cathartic
release, religious or secular education, imminent or deferred punishment
and therapeutic procedure.
A salient element of this literature is socio-biology, defined as ‘the system-
atic study of the biological basis of all social behaviors’ (Crawford 1987: 3),
and its pioneer, Charles Darwin (1809–82). Its master-concept is the idea
that organic adaptations, including patterns of behaviour, have their origins
in an evolutionary course of change. This has two fundamental tenets: that
variation exists at the outset of a process of achieving a better fit between
organism and environment and that this character is transmitted between
parent and offspring. As Patrick Bateson explains:
culture is created by the communal mind, and each mind in turn is the
product of the genetically structured human brain. Genes and culture
are therefore inseverably [sic] linked . . . . The mind grows from birth
to death by absorbing parts of the existing culture available to it, with
selection guided through epigenetic rules inherited by the individual
brain . . . . The quicker the pace of cultural evolution, the looser the
connection between genes and culture, although the connection is never
completely broken.
(1998: 127–8)
Innate 75
Culture, historiography and ancient hatreds
Socio-biology’s turn towards culture is born, in part, of the perennial prob-
lem of variation. It has long been noted that conflict ‘does not invariably or
universally lead to the same behaviour’ and that some societies display both
higher frequencies and greater intensities of violence (Boring et al. 1939:
163). Indeed, as a subset of human action, conflict is relatively infrequent.
While each of us may have disputes with other individuals quite regularly,
aggressive behaviour is, for most people in most cultures, quite unusual and
violence is rarer still. This is even true of the vast majority of XYY men (well
over 90 per cent) who never acquire a conviction for violent crime and who
respond well to interactive therapy (Manning 1989: 54–5). Indeed, as Keith
Webb observes, modern people probably spend more time watching televi-
sion than behaving aggressively, yet nobody is arguing that this is born of an
innate propensity (1992). Moreover, at the state level, the majority of coun-
tries have experienced a declining rate of warfare over the last few centuries,
despite considerable rises in human population (Levy and Morgan 1984). In
fact, across the global system as a whole, the incidence of large-scale wars
may be cyclical rather than deterministically constant – a proposition ex-
plored in Chapter 10 (Väyrynen 1987). In all, this suggests that ‘whatever
the bases of human aggression, it is within the capacity of humans to do
away with it’, or at least prevent its expression (Goldstein 1989: 15).
For psychologists and social scientists supportive of the innateness para-
digm, the societal provision of individual cathartic releases may be one such
way to prevent the expression of inner aggressive drivers. Here, psychoanaly-
sis and the work of Sigmund Freud are important. He argued that individuals
are born with an inherently self-destructive tendency, Thanatos, which works
‘to reduce life to its primal state of inert matter’ – the evolutionary origins of
humanity, in other words (quoted in Strachey 1964: 75). Buried deep within
the unconscious id (over which biological and cultural evolution has laid a
thin veneer), the influence of Thanatos is, it is argued, reduced by displacing
its destructive qualities on to other individuals, thereby causing conflict (Kull
1990). Society must, Freud argued, institutionalise cultural practices that can
offer a functional outlet for pent-up emotion. This remains an important
idea today and underpins much of the justification for competitive, including
violent, sports as providers of a ‘satisfactory outlet for the instinctive aggres-
sive drive’ (Menninger cited in Berkowitz 1990: 30).
Other writers argue that the cathartic control of inner aggression can be
bolstered by developing complex social structures of dominance, defined as
the ‘probability that the dominant animal will have preferential access to
some good to which both animals aspire’ (Somit and Peterson 1997: 52).
While also providing an enhanced means of reproductive success (for those
who are on top), these structures serve ‘to contain male aggressiveness and
subordinate it to the needs of the group’ (Corning 1971: 342). They do so
by increasing the predictability of social relations and by obviating the need
76 Innate
for continual disputes over resource access (Jolly 1985). As ‘stark coercion,
unsupported by other devices, is usually unsuccessful over time’, it is con-
cluded that dominance has evolved in order to ensure that male aggression is
contained by norms of obedience and authority (Wilkinson 1969: 8). While
modified by regime variation in a constant struggle between the strong and
the weak, such an iron law of oligarchy is considered to be universal, ineluc-
table and, insofar as it ensures passivity, socially functional (Frank 1985).
The failure of some societies to institutionalise cultural patterns that con-
trol, or provide functional outlets for, male aggression is often seen as a
fundamental explanation of substate level violence. Group formations are,
for observers adopting this approach to conflict analysis, ‘controlled prima-
rily by the values, norms, or duties imposed by the sociocultural structure
. . . distantly related to the basic propensities of individuals on which natu-
ral selection operated’ (Hinde 1993: 49). As Umberto Melotti notes of the
work of the evolutionary biologists, Richard Alexander and Gerald Borgia
(1978),
the great divisions among humankind and the dominating source of con-
flict will be cultural. . . . [T]he principal conflicts of global politics will
occur between nations and groups of different civilizations. . . . These
include Western, Confucian, Japanese, Islamic, Hindu, Slavic-Orthodox,
Latin American and possibly African civilizations.
(Huntington 1993: 22, 25)
Conclusion
This chapter has considered the idea that conflict is in natus, or present at
birth. A bedrock of the West’s analysis of the ‘ethnic’ conflicts of the South
and frequently implied in many other studies of contemporary conflict, this
basic premise has its roots in Christian theology and the notion of trans-
ferable sin. Two bodies of work have been focused upon here. Although
derived from a wide range of different disciplines and traditions, both draw
on the acute impact that violent conflict has on the human psyche and the
intense pessimism it can engender. Their basic postulate, that the innateness
of the human propensity for conflict makes violence an inevitable aspect of
social interaction, is sustained by socio-biology’s scientific aura, which has
remained an important part of the social sciences for much of the last 100
years. After all, refutation is, as Gerhard Lenski observes, difficult given the
widely held belief that many social studies from outside the tradition of ex-
perimental science lack ‘substantive conceptual links to established theories
in other scientific disciplines [and are not] . . . falsifiable in the same unam-
biguous manner as theories in the natural sciences’ (1988: 163).
84 Innate
The first section above looked at the biology of the human body, its evolu-
tionary origins as well as its neurological, psychological and hormonal func-
tions. The emergence of the discipline of socio-biology, as the inheritor of
Christianity’s understanding of the human tendency to violence, was traced
alongside an account of its early and contemporary applications to the social
sciences. Darwin’s work, along with eugenics, genetics and ethology, was
considered with a particular focus on the causes of conflict within human
society. Especially important here are notions of dominance, altruism and
kin recognition. These have been subject to concerted challenges seeking
to rebut some of their more illiberal findings and to question the authority
of natural scientists to carry social research, given the fact that they have
spent several hundred years unsuccessfully endeavouring to elucidate the
general laws they so frequently claim guide human society (Jacoby 2004a).
Although frequently dismissed as an unscholarly attack orchestrated by ‘the
poets and artists among us’, they have, nevertheless, obliged socio-biolo-
gists to incorporate more aspects of cultural adaptation (Blalock 1984: 25).
Ideas such as the leash principle and the inclusive fitness of dominance and
cathartic release have consequently been used to elaborate models of conflict
and violence that emphasise adaptive influences.
The second body of work examined in this chapter relates to the way in
which culture and historiography are understood as causes of violent con-
flict. Although this literature frequently owes its foundational assumptions
to the ‘hard’ sciences of Darwin and his successors, it is, especially in its
populist forms (from journalists and politicians), more commonly applied to
the seemingly intractable dynamics of the ‘new wars’ introduced in Chapter
1. As the sustained popularity of Huntington’s clash of civilisations thesis
demonstrates (particularly since the attacks on Washington and New York
of 2001), however, the idea of immutable cultural forces driving individuals
to commit organised acts of violence does not only serve to explain substate
conflicts. History also weighs heavily upon the level of the system which,
in this characterisation, is marked by a millennial struggle between an in-
herently reasonable West and the immutable barbarism of the (frequently
Muslim) ‘other’. Here, the function of culture is quite different from the
models looked at in the next chapter. Rather than formed by the settling of
generations of broadly unchanging social practices, these regard culture as
requiring constant substantiation and legitimisation in order to exert an in-
fluence on human behaviour. As individual action is not linked to hereditary
endowments, conflict and violence are, as we shall see, regarded primarily as
heuristic responses to contingent changes in the social environment.
6 Learnt
As we saw in the previous chapter, it has long been agreed that most forms
of aggressive behaviour are influenced by learning. As early as the 1950s, the
zoologist John Paul Scott was able to conclude that, among dogs, ‘the mo-
tivation for fighting is strongly increased by its success, and that the longer
success continues, the stronger the motivation’ (1958: 194). Today,
It is these latter attempts that will make up the focus of this chapter. In many
ways, they are summed up in the Seville Statement on Violence of 1986.
Signed by 20 scientists from 12 countries and from a wide range of disci-
plines, it put forward five basic postulates:
1 that we have not ‘inherited a tendency to make war from our animal
ancestors’;
2 that ‘war or any other violent behavior is not genetically programmed
into our human nature’;
3 that it is ‘incorrect to say that in the course of human evolution there
has been a selection for aggressive behavior more than for other kinds
of behavior’;
4 that it is ‘incorrect to say that humans have a “violent brain” ’; and
5 that war is not ‘caused by “instinct” or any single motivation’.
(UNESCO 1986).
Social learning
Theories of social learning generally work from the premise that behaviour
in general and aggression in particular are acquired from the environment
through conditioning rather than driven by innate features of human physi-
ology or psychology. Such an idea is far from new. Edward Thorndike pre-
empted the work of more famous behaviourists (such as Burrhus Skinner’s
analysis of external stimuli (1938)) with his studies of trial and error asso-
ciation published in 1898. According to this, and a growing body of more
recent literature, behaviour can be conditioned in two ways. First, condi-
tioning may occur directly through rewards and punishments experienced
as part of an internalised process of heuristic experimentation. For instance,
studies have found that children who respond aggressively to school bul-
lying are more likely to increase their aggressiveness if they perceive the
response to be successful – if, in other words, the bullying ceases (Fox and
Boulton 2003). This may be taken as an indication of the reinforcing effects
of either the reduction of annoyance or the attainment of a reward. Either
way, aggression is reinforced not by satisfiers inherent to the aggressive act
itself, but by its capacity to meet an individual’s immediate needs which,
in most cases, are not biologically essential (Tapper and Boulton 2005). As
such, aggression is, according to many writers, deployed selectively in order
Learnt 87
to obtain a reward such as wealth, prestige or status, the value of which is,
itself, reinforced by social learning mechanisms. This can be offset by nega-
tive punishment in the form of the withdrawal of a reinforcing influence or
positive punishments such as the imposition of noxious sanctions. In Thorn-
dike’s early experiments, positive punishments were regarded as a symmetri-
cal influence upon response – carrots persuade, sticks dissuade (1932). More
recently, however, it has been widely accepted that the effectiveness of both
forms of punishment depends largely on the identity of the punisher and on
the punisher’s assessment of desert. If the former lacks authority and legiti-
macy or the latter is found to be arbitrary and unwarranted, aggression is
unlikely to be inhibited and may, in fact, be exacerbated or even instigated
(Pisano and Taylor 1971).
The importance of legitimacy and authority in modifying behaviour has
been famously demonstrated by the experiments of Solomon Asch and his
student Stanley Milgram. The former presented a number of individuals (of
whom all but one – the subject – had been prebriefed) with a flipchart upon
which were drawn two lines of various lengths and two of identical lengths.
He then asked each individual to tell the group which the identical lines
were, leaving the uninformed subject until last. In keeping with Asch’s earlier
instructions, each individual initially identified the two identical lines cor-
rectly with the subject following suit. After some time, however, the group,
again as instructed, all began to identify two obviously different-sized lines
as identical, thereby pressuring the subject and causing them to doubt their
own judgement. During 12 trials, 76 per cent of subjects went along with the
groups’ incorrect selections. When the experiment was repeated, first, using
a secret ballot and, second, with another individual going against the group,
the subject always identified the correct answer. Thus Asch concluded that
individuals do not simply follow crowds blindly; rather, they assess the level
of disapproval they are likely to face from challenging established, legitimate
and authoritative norms (1951). If this can be mollified by political elites,
then, as Chapter 4 revealed, the construction of internal foes and the pursuit
of overseas distractions can be successfully deployed to buttress their leader-
ship positions.
Stanley Milgram was interested in the extent to which this kind of social
pressure could instigate violence (1963). He was inspired by the trial of
Adolf Eichmann, who, having been responsible for the transportation and
extermination of German Jews during much of the Second World War, had
been captured by Israeli agents in Argentina in 1960 and returned to Jerusa-
lem for prosecution. The world’s media were in attendance and struggled to
explain why he had done such terrible things: was he a deranged monster,
as much of the press suggested, or, as Eichmann himself claimed, was he an
ordinary person simply following orders for which he was not responsible
(Wistrich 1997)? In order to examine the credibility of Eichmann’s defence,
Milgram placed a newspaper advertisement offering US$4.50 for one hour’s
participation in what was called a psychology experiment investigating
88 Learnt
memory and learning. Respondents (identified as ‘S’ in Figure 6.1) were in-
troduced to a stern-looking experimenter in a white coat (‘E’) and a friendly
co-subject (‘A’), both of whom had been selected and briefed by Milgram.
Fabricated lots were drawn to ensure that the subject ‘S’ took on the role
of ‘teacher’ and the co-subject ‘A’ (who was taken to an adjoining room,
strapped to a chair and attached to a fake electrode) was to be the ‘learner’.
The former was taken to a desk containing a generator, instructed to read a
list of word pairs and then told to ask the ‘learner’ to read them back. If the
‘learner’ responded correctly, the ‘teacher’ should move on to the next word
but, in the case of an incorrect response, the ‘teacher’ was directed to deliver
an electric shock to the ‘learner’ using the equipment provided. This had 30
labelled switches in 15-volt increments up to 450 volts. Each switch also had
a rating, ranging from ‘slight shock’ to ‘danger: severe shock’ (Blass 2004).
The ‘teacher’ was told to increase the shock each time the ‘learner’ missed
a word in the list (the equipment was, of course, harmless and the ‘learner’
was an actor feigning electrocution). Once the ‘teacher’ had been assured
that the experimenter assumed full authority for the consequences of the ex-
periment, 65 per cent delivered the maximum 450 volts, and none stopped
before reaching 300 volts – demonstrating that, even in liberal democracies,
Figure 6.1 The set-up of Milgram’s experiment on obedience and authority. Source:
Redrawn from http://en.wikipedia.org/wiki/Milgram_experiment/.
Learnt 89
people are willing to obey authority figures and commit acts of violence that
they would normally find morally repugnant (Harrington 2004: 56; Post
2004).
A second type of conditioning may occur vicariously through the obser-
vation of models, which, if successful, have ‘the effect of a positive rein-
forcement on the viewer’ and, if unsuccessful, have a deterring outcome
(Björkqvist 1997a: 32). Current writing in this area of social psychology
owes much to the work of Albert Bandura. In his formulation, knowledge
about the likely result of aggressive behaviour is acquired by witnessing its
consequences upon others. This was demonstrated in a series of experiments
using Bobo dolls (Bandura et al. 1961). These were large toys that were
filmed being beaten and abused by adults. This was then shown to three
groups of children accompanied by three different endings. The first film
showed the adult being rewarded for his aggressive behaviour, the second
showed the adult being punished and the third showed no consequences.
The children who watched the video in which the person was rewarded for
his actions were found to be more likely to duplicate the behaviour than
those children who witnessed the adult either being punished or receiving
neither punishment nor reward. From this, four conclusions were drawn
regarding the power of such modelling: that it teaches new behaviour, that it
influences the frequency of previously learned behaviour, that it may encour-
age previously forbidden behaviour and that it increases the frequency of
similar behaviour (Bandura 1973).
Extrapolating these findings to the level of society, Bandura hypothesises
that ‘there are three major sources of aggressive behaviour [or models], which
are drawn upon to varying degrees’ (1976: 124). First, the correlation of
anti-social behaviour in children with homes in which aggression, in words
as well as deeds, is predominant suggests that familial influences are highly
significant. Parents who attempt to influence their children using coercive
methods are presenting a positively reinforcing model of aggression which
conditions directly, if experienced by the child personally, or vicariously, if
the child observes siblings adjusting their behaviour in response to parental
aggression (McCord et al. 1959). Second, variations in the expression of ag-
gression indicate that, where it is highly valued (often through competition,
sport and historiography), violent subcultural influences may vicariously
reinforce individual behaviour. Here, cultural features are not expressions
of innate tendencies fixed by the weight of history, but influential only if per-
petually reinforced by successful modelling (Wolfgang and Ferracuti 1967).
Differences in violence levels between cultures are not then a consequence of
the failure to institutionalise catharsis or dominance, but a result of continu-
ally constructed, socially contingent and constantly changing conditioning.
A third factor, symbolic modelling, involves the legitimisation of aggression
through the indirect example of leaders, celebrities, fictitious heroes/hero-
ines, sports stars and so on – influences frequently intensified by favourable
media portrayal. In this way, both children and adults have virtually limitless
90 Learnt
ways of receiving televisual and pictorial models of aggression, inputs that
have been shown to have a demonstrable effect on both the quantity and the
quality of interpersonal aggressiveness (Felson 1996).
Recent studies have extended Bandura’s findings to establish more nu-
anced accounts of familial conflict. Kaj Björkqvist, for instance, finds that
children are more influenced by their mothers’ behaviour outside the home
and their fathers’ approach when they are angry at home. Moreover, while
mothers generally exert greater reinforcing authority, sons tend to imitate the
aggressive behaviour of their fathers more than daughters, who, outside the
home at least, demonstrate a stronger maternal influence (1997b: 79–80).
Analyses of a wide range of deviant behaviour, from terrorism to drug use,
have also made use of Bandura’s research (Ruby 2002; Lee et al. 2004).
The criminologist Ronald Akers, for instance, finds that the ‘probability that
persons will engage in criminal and deviant behavior is increased and the
probability of them conforming to the norm is decreased when they dif-
ferentially associate with others who commit criminal behavior or espouse
definitions favorable to it’ (1998: 50). This may be the direct influence of
rewards and sanctions or, as in the case of televisual portrayals of criminal
aggression, desensitisation through vicarious observation (Eron et al. 1994).
In direct contrast to the immutable characterisation of aggression of the last
chapter, such an understanding of behaviour offers numerous methods of
treatment (Cunningham et al. 1998). Rebecca Dobash and her team have,
for instance, used social learning approaches to treat violent offenders by
reinforcing alternative response models through education, self-assessment
and peer pressure (1996).
These studies, along with those of Asch and Milgram, have also been of
interest to military institutions. They, too, have sought to modify individu-
als’ aggressive response through reinforcement and modelling. Rather than
reducing aggression, though, they have used behaviourist experiments to
overcome the perennial problem of non-firing. Of the 26,000 muskets col-
lected from the field following the Battle of Gettysburg in 1863, for instance,
over 24,000 were still loaded, 12,000 were loaded more than once and more
than 6,000 were loaded between three and ten times (Lord 1976). This
was particularly puzzling given the labour and time involved in preparing
muzzle-loaded weapons for use. As Dave Grossman explains, ‘the obvious
conclusion is that most soldiers were not trying to kill the enemy. Most of
them appear to have not even wanted to fire in the enemy’s general direc-
tion’ (1996: 22). A similar phenomenon was observed by Brigadier-General
Marshall during the Second World War. He claimed that, of his study group
of American combat soldiers, at least 75 per cent did not fire their weapons
at all during battle (1947). It would appear that ‘when soldiers discover their
vulnerability then the fear of being killed, rather than killing, becomes the
more oppressive problem’ (Newsome 2003: 33).
Military institutions’ response to this issue has been complex and var-
ies considerably from country to country. Three general approaches can,
Learnt 91
however, be linked to theories of social learning. The first is group identity.
As Asch demonstrated, individuals will conform to the views of a proximate
group by assessing the level of disapproval they will incur by deviating. To
increase such pressure and thus soldier participation on the battlefield, the
United States army introduced small, cohesive fire-teams following the Sec-
ond World War (George 1971: 297; Kennett 1997: 137). This could not,
however, deal with the type of problem experienced at Gettysburg where
the closeness of firing formations meant that ‘if a man truly was not able
or willing to fire, the only way he could disguise his lack of participation
was to load his weapon . . . bring it to his shoulder, and then not actually
fire’ (Grossman 1996: 23). To tackle this problem, the United States army,
in keeping with the findings of Milgram, adopted a second approach – the
devolution of authority to these small companies of buddies. Non-commis-
sioned officers, such as corporals, were trained to develop high levels of
legitimacy with which to elicit commitment from their troops (Shalit 1988).
Assuming the role of Milgram’s ‘experimenter’, they both take responsibility
for the actions of their soldiers and issue fire commands which, given their
proximity and authority, are difficult to ignore – particularly given the exten-
sive vicarious reinforcement of associating status with rank hierarchies. Both
these factors – the intensification of group identity and the decentralisation
of authority – have considerable implications for the mobilisation of social
groups (looked at in Chapter 8) that are too large to be directly influenced
by charismatic or zealous leaders.
Third, the conditioning of enlisted men before arrival at the combat arena
has been revolutionised. The direct reinforcement of the need to engage the
enemy on the battlefield has been stepped up through the increased use of
medal citations (Kellett 1990: 220–1). Subcultural and symbolic modelling
have also been developed in what Grossman calls the ‘boot-camp deification
of killing’ (1996: 252). This has mixed the dehumanisation of the enemy
through the constant use of derogatory language (thereby reinforcing a divi-
sion between the collective self and the other) with highly repetitious and
blood-thirsty chanting. Soldiers are thus ‘being indoctrinated in the most
explicit fashion (as previous generations were not) with the notion that their
purpose is not just to be brave or fight well; it is to kill people’ (Dyer 1985:
123). This type of conditioning has, in some cases (such as the preparation
of Liberation Tigers of Tamil Eelam cadres in Sri Lanka), been accompanied
by the use of televisual media through which highly graphic portrayals of
battlefield violence are repeatedly shown to new or potential recruits with
the joint intention of instilling feelings of anger and of desensitising.
Most armed forces have also replaced static target training with ‘quick-
shoot’ ranges employing pop-up models of enemy soldiers. This has had two
outcomes. First, it offers immediate gratification in that an accurate round
produces an instant and very obvious indication of success (a reward posi-
tively reinforced by considerable regimental kudos and negatively reinforced
by mild punishments). Second, ‘in addition to traditional marksmanship,
92 Learnt
what is being taught in this environment is the ability to shoot reflexively and
instantly. . . . The man shape popping up . . . [makes] the conditioned stimuli
more realistic and the conditioned response more assured’ (Grossman 1996:
254). The effect on the United States military was that, by Vietnam, firing
rates had increased to between 90 and 95 per cent – thereby confirming
Bandura’s fundamental hypothesis that ‘people are not born with preformed
repertoires of aggressive behavior. They must learn them . . . either through
observation of aggressive models or on the basis of direct experience’ (Ban-
dura 1976: 122; Holmes 1985).
Constructed identities
Such studies have had a profound effect on the way in which collective
identity and its value in the eyes of political leaders are studied. Bandura’s
observation that societies which ‘provide extensive training in aggression
and make it an index of manliness or personal worth’ spend more ‘time
threatening, fighting, maiming and killing each other’ than ‘cultures where
interpersonal aggression is discouraged and devalued’ underpins the idea
that the efficaciousness of violence is constructed (or learnt) rather than in-
nate (1976: 128). Constructivism, as an approach to the analysis of collec-
tive identity (national, religious, ethnic, gender and so on), therefore rests on
the assumption that both the categorisation of, and the boundaries between,
social groups are fluid and subject to constant manipulation. As Jennifer
Sterling-Folker notes, ‘this means that there is no “middle ground” between
the biological and the social in most constructivist narratives. Human in-
teraction is instead treated as if it springs forth from some unknown source
that has no implications for species-wide behaviour’ (2002: 92). So, rather
than offering a static account of fixed antagonisms, constructivism seeks to
explicate ‘a specific process by which identities are produced and reproduced
in action and speech’ (Fearon and Laitin 2000: 850). In keeping with their
behaviourist roots, collective identities are therefore expected to be stronger
‘the greater the tangible rewards, perceived tangible rewards or expected
tangible rewards associated with the ingroup and the loyalty to it’ (Rosen-
blatt 1964: 132). ‘Rather than reflecting deep, historic passion and hatreds’,
civil wars between self-identified groups are, accordingly, more usefully con-
sidered to be a ‘result of a situation in which common, opportunistic, sadis-
tic, and often distinctly nonideological [sic] marauders [a]re recruited and
permitted free reign by political authorities’ (Mueller 2000: 43).
Such leaders might ‘construct’ and demarcate a collective identity for a
wide range of strategic reasons, some of which (including diversionary mo-
tives) have been discussed in Chapter 4. The concern for constructivists,
however, is more ‘how’ this is achieved than ‘why’. The latter is mostly
assumed to follow the basic tenets of rationality – self-preservation, material
profit, status acquisition and so on (Tambiah 1996). Assuming that rigorously
Learnt 93
exclusionary collective identities do not serve popular interests, the chal-
lenge is, therefore, to understand the ways in which leaders convince people
of abstractions such as nationalism, racial purity, gender roles and such like
and then persuade them to act upon these beliefs. One of the simpler ways
in which elites might achieve this is, as Milgram demonstrated, to create
a sense of moral resignation by using their authority over their immediate
followers to provoke cycles of violent action and reaction (Kalyvas 1999).
In cases in which their constituents stand to prosper from a collapse in civil
order or from a redistribution of wealth, such a tactic can reduce internal
dissent within, and external pressure upon, the in-group by appearing to
consolidate the idea that support for the existing leadership structure is the
only viable way to preserve its collective identity (Kapferer 1988: 102).
In Rwanda, for instance, Gérard Prunier argues that Hutu extremism was
based, in part at least, on the need to demonstrate both to the peasantry and
to international donors that their policies (and not those of moderate Hutus
who had negotiated a ceasefire with the Tutsi-influenced Rwandan Patriotic
Front two years earlier) were the only feasible means to protect the majority
of Rwandans (approximately 80 per cent of whom were Hutu) and thus to
bring stability to the country as a whole (1995: 141–3). To reinforce this
idea, the President’s wife, Agathe Habyarimana, and her three brothers es-
tablished small death-squads which, by selectively killing Tutsis, succeeded in
ending the ceasefire, fomenting an environment of acute fear and gathering
support for the Hutu Interahamwe militia (Gourevitch 1998). The result-
ant sense of collective peril among Hutus reduced perceptions of individual
responsibility for the escalating violence and enhanced the authority of the
extremists (despite the fact that over 90 per cent of Hutu men did not ac-
tively take part in the genocide) (Des Forges 1999).
It has frequently been observed that a key element in the construction of
such conflictive social identities in general, and in Rwanda in particular, is
the control of the mass media (Hobsbawm 1990: 141–2). Although graphic
televisual portrayals of violence have long been shown to be among the
most powerful vehicles for vicarious behavioural reinforcement, literary and
verbal communication of violently exclusionary ideas, commended by de-
centralised, non-state authorities, can also exert a subcultural and symbolic
modelling effect on opinion and thus potentially on behaviour (Lagerspetz
1989; Brass 1997). Studies of the United States, for instance, have, as per
Milgram, shown that the public is swayed most powerfully by the testimony
of authoritative and legitimate experts (Page and Shapiro 1992: 339–54).
This can both negatively reinforce out-groups (thereby demobilising opposi-
tion to the elite’s version of reality) and positively reinforce the character
and boundaries of in-group self-perception. Such a process may occur either
as part of a restrictive environment in which ‘official control of information
makes public opinion highly manipulable’ or ‘during incipient democratiza-
tion, when . . . the state and other elites are forced to engage in public debate
94 Learnt
in order to compete for mass allies in the struggle for power’ (van Evera
1994: 26–33; Gagnon 1995; Human Rights Watch 1995: xiv; Snyder and
Ballentine 1996: 6).
In Rwanda, both scenarios were apparent during the build-up to the
genocide. The abandonment of the state’s media monopoly in 1990 had
produced ‘an explosion in the number of newspapers and journals’ which,
in presenting polarised interpretations of the unfolding conflict, radicalised
and pressurised the incumbent government and helped to manipulate public
opinion (Africa Rights 1994: 150). The extremist Hutu journal Kangura,
for example, ‘warned Habyarimana not to flinch from the destruction of
the Tutsi’ (Snyder and Ballentine 1996: 33). With only around 60 per cent
literacy, though, a more important vehicle of persuasion was the radio,
over which Hutu extremists obtained a virtual monopoly. For example, the
Radio-Télévision Libre des Mille Collines, established by Agathe Habyar-
imana, became an instructive voice of elite Hutu radicals through which the
boundaries between Rwanda’s ethnic groups were redrawn. From Kigali and
then from within the French army-controlled zone, it helped to mobilise
the genocide through a combination of authority, threatened punishments
and promised rewards – leading Holly Burkhalter, the director of Human
Rights Watch in Washington, DC, to conclude that ‘the one action that, in
retrospect, might have done the most to save Rwandan lives’ would have
been the censure of its broadcasts (1995: 53).
At a more diffuse and general level, access to mass media outlets can
also help to establish what James Fearon and David Laitin call ‘social con-
struction by discourse’. This, they continue, relies on ‘symbolic or cultural
systems that have their own logic or agency . . . . [I]ndividuals are pawns or
products of discourses that exist independently of the actions of any particu-
lar individual’ (2000: 851). These ideological frameworks provide meaning
to actions and can thus assist in explaining how individuals come to imagine
themselves to be part of an exclusive collective group – or, put another way,
how ‘the multiple identities of individuals come to be expressed in terms
of one dominant identity’ (Jabri 1996: 120). Here, models of social learn-
ing blend with accounts of globalisation and post-modernity to produce
analyses of conflict less based on truth, or causality, and more concerned
with the rise of fragmented supra- and substate identities and the content
of the competing narratives that they develop (Bilig 1995: 131). These, it
is suggested, tend to be ignored by commentators whose focus on ‘mod-
ern’ conflict attributes – professional military structures, scientific strategy,
field tactics, technology and clear war aims – causes them to overlook the
post-modern character of contemporary warfare (its diverse armed groups,
ethnicised politics, ritualised violence and transnationality) and to regard it
as an abnormality or social pathology evident simply in manifest adversity
(Duffield 1998a; Kaldor 1999).
In reality, many argue, today’s conflicts are a direct result of the routine ac-
tions, cultural references and linguistic content of all collective identity (Gid-
Learnt 95
dens 1984: 60). Often banal, subliminal and self-affirming, these combine
to construct what Vivienne Jabri calls discourses of ‘exclusion’ and ‘origins’
(1996: 128–41). The former relates to the dichotomous representation of
out-groups, and the latter refers to the homogenised traditions of forebears.
Each is held to be imaginary (and thus impervious to rational analysis) and
both are thought of as constituent parts in the legitimisation of conflictive
relations. In attitudinal terms, the conflict engendered by these is therefore
what Mitchell (considered in Chapter 2) might describe as ‘unrealistic’ as,
like the cathartic release of inner aggression, it is not caused by the demon-
strable pursuit of incompatible goals. Rather, the dynamics of the conflictive
discourse are represented in symbols, ritual, art, emblems, remembrance and
sovereignty as well as notions of societal interest and social difference (Onuf
1989). These are then disseminated by ‘the casting of thought in language
[which] makes the private and the individual public and collective by accom-
modating individual experience and subjectivity’ (Norton 1988: 46).
Gender
The gendered nature of this language has proved of particular interest to
feminist writers on violence and conflict. As Table 6.1 illustrates, commen-
tators from what Ted Hopf calls a ‘critical constructivist’ background have
identified binary gender oppositions within theories of conflict themselves
(1998: 181–6). These, it is argued, undermine the claim to truth inherent
in more positivist studies of conflict and identity and provide a challenge to
the implied gender essentialism of socio-biological research examined in the
previous chapter – both of which may be said to disguise their masculinised
character behind an afflatus of normality and ‘scientific’ authority (Peterson
and Runyan 1993: 22–5; Jacoby 2004b). In emphasising the intersubjectiv-
ity of meaning, the reflexivity of the self and the concealment of patriarchal
discourses, ‘feminism and constructivism [thus] share an ontology of becom-
ing’ that resists the ‘natural’ in favour of the socially acquired (Hoffman
1987; Locher and Prügl 2001a: 111). In this sense, constructivism’s account
of social identity has helped to establish mechanisms of individual agency
within a ‘naturalised’ social structure, while feminism has informed analyses
of the construction process by more fully incorporating gendered forms of
obviously there is for you some glory, some necessity, some satisfaction
in fighting which we have never felt or enjoyed . . . As a woman I have
no country. As a woman I want no country. As a woman my country is
the whole world.
(1938: 9, 166)
There is evidence to support the idea that women are socialised towards
greater passivity then men. For instance, although local variations are con-
siderable, only around 3 per cent of the 23 million or so uniformed soldiers
worldwide are women – of these, combat troops are 99.9 per cent male. In a
large scale study of values and policy preferences, Benjamin Page and Robert
Shapiro were able to conclude that in ‘practically all realms of foreign and
domestic policy, women are less belligerent than men’ (1992: 295). Indeed,
a survey of 285 American polls showed a regular male preference for a ‘more
violent or forceful option’ (Smith 1984: 384–95). As Table 6.2 illustrates,
this has been, in many cases, at a surprisingly consistent level, suggesting that
states in which the participation of women is high are less likely to take part
in wars than those where women are less prominent. Having reviewed over
2,000 military incidents between 1960 and 1992, Mary Caprioli supports
such a view. She concludes that ‘higher levels of gender equality correlate
with lower levels of military action to settle international disputes’ (2000:
65).
Indeed, the prominent role that women’s organisations have played in
peace and civil rights movements and anti-nuclear campaigns, along with
eco-feminism’s attempts to institutionalise less structurally violent ways of
98 Learnt
Table 6.2 Gendered survey support for American policy
Percentage in agreement
Year Issue Men Women
1939 Take strong measures against Japan 56 42
1960 Adopt a tougher policy towards the Soviet 60 46
Union
1968 Step up effort in Vietnam 50 32
1975 Wars are necessary to settle differences 55 38
between countries
1991 Use force to remove Iraqi troops from Kuwait 63 41
Source: Adapted from Goldstein (2001: 49); Nincic and Nincic (2002: 559).
transforming and utilising nature, would seem to substantiate the idea that
‘in most societies boys are encouraged more than girls to behave aggres-
sively’ (Segall 1989: 179). The apparent ubiquity of such gender differences
has led some to conclude that, once social development has ‘modernised’
gender relations, a more pacific, and consequently less resilient, environ-
ment will emerge (Fukuyama 1998: 27, 36). As Martin van Creveld puts it,
‘if only because research shows that going into combat is the last thing that
military women want [one study found that 52 per cent of female soldiers
would probably or definitely leave the service if forced into combat positions
(O’Beirne 1998)], the more of them there are around the less capable those
military are of acting as effective combat units’ (2000b: 442).
For many feminists, however, seeing the socialisation of women in pacific
terms unwarrantably universalises the experience of the Western middle
classes and serves to essentialise gender roles in ways similar to the biological
determinism of innate theorists. They point to research demonstrating that
a significant proportion of domestic violence is initiated by women against
men, against their own children or, in the case of same-sex relationships,
against other women (James 1996). In J. Ann Tickner’s view, then, ‘the as-
sociation of women with peace renders both women and peace as idealistic,
utopian, and unrealistic: it is profoundly disempowering for both’ (2002:
338). So, rather than seeking to reinforce male–female dualisms which, they
suggest, replicate the rigid and restrictive nature of the patriarchy, many
feminists see individual rights, freedom of choice and institutional access as
key to the advancement of women’s status. Their exclusion from warfare is
thus emblematic of a broader tendency to discount both the potential and
the actual contribution of women – a feature (of developed and developing
polities) more likely to provoke cynicism and withdrawal than peace activ-
ism (Tobias 1990: 181–2).
Where women have been permitted to participate in violent conflict, their
performance has been, it is argued, limited primarily by the discriminatory
and inhibiting effects of masculine identity. This was demonstrated on the
Learnt 99
home front during the World Wars when women in the United States con-
tributed actively to military production despite being restricted to a ‘menial
type of corps of low-grade personnel’ (Treadwell 1954: 12). In combat units,
it has often been argued that women perform better in guerrilla armies,
where a comparatively decentralised format and an ostensive commitment
to ‘fundamental alterations in the socio-political order’ produce a working
environment less conditioned by male identity than states’ imperative ‘of op-
timising military effectiveness’ (Enloe 1988: 161, 164). Although their duties
have often replicated the divisions of domestic labour, some revolutionary
movements have also made extensive use of female troops on the front line
(O’Gorman 1999: 92–6). The armed wing of the Nicaraguan Sandinistas,
for instance, was made up of 30 per cent women (Jones 1997: 103).
Women also demonstrate considerable support for wars in which their
participation is limited. Survey data on the Arab–Israeli conflict, illustrated
in Table 6.3, shows, for instance, no significant gender differences across the
Middle East. In the United States, too, differences in support for the hypo-
thetical use of force have been found to be slight. They are, Pamela John-
ston Conover and Virginia Shapiro conclude, ‘certainly not large enough to
warrant making the kinds of sweeping statements differentiating men and
women that have long been part of stereotype’ (1993: 1095). Indeed, this is
indicated by the vociferous campaigns that women have led to gain access
to the armed forces. In a challenge to the constitutional legality of all-male
military registration, the American National Organization of Women, for
example, argued that ‘the military is so central to the entire social order that
it is only when women gain access to its core that they can hope to fulfil their
hopes and aspirations’ (cited in Elshtain 1985: 43; see also Elshtain 2000:
445).
The grounds for resisting such access primarily rest on the biological
theses put forward in the previous chapter. It is claimed that the innately
‘superior ability of men to add muscle to their bodies’ makes them better
soldiers regardless of the social conditioning an individual has received
Table 6.3 The relationship between gender and attitudes towards the Arab–Israeli
conflict (all figures in percentages; date of survey in parenthesis)
Conclusion
This chapter has reviewed a number of ways of approaching the causes of
conflict and violence from the perspective of human learning. Although
none is, in any sense, uniform and, together, they are far from representing
a coherent position, all share an appreciation of conflict causality as fun-
damentally derived from the social environment. The tendency to behave
aggressively is acquired from individuals’ socialisation experience and de-
ployed selectively as part of a range of possible responses to given stimuli.
Early expressions of this basic premise can be found in the work of behav-
iourists such as Thorndike, Skinner and Asch, whose conclusions regarding
the functions of rewards and punishments – both material and non-material
– paved the way for later experiments aimed at understanding the role of
groups, obedience and authority figures. These established that behaviour
could be significantly affected by both direct and observational influences.
So, given the demonstrably compliant character of social interaction, such
a combination of experiential and vicarious pressures and inducements can
be seen as the primary or sole determinants of a considerable proportion
of interpersonal violence. Conflict may thus result directly from authority
and instruction as part of a direct relationship, as famously demonstrated by
Stanley Milgram, or vicariously from the symbolic and subliminal power of
society and culture.
102 Learnt
At the level of the individual, this helps to extend and elaborate Mitchell’s
conflict triangular introduced in Chapter 2. The common patterns of ex-
pectation, emotional orientation and perception that, he argues, accompany
involvement in a conflict situation can, perhaps, be more clearly understood
as not simply subjectively experienced states of mind, but collectively devel-
oped imprints of the broader social environment that surrounds conflictive
situations. In this sense, conflictive attitudes such as anger, resentment and
suspicion are not simply associated with other cognitive processes such as
stereotyping and selective approaches to new information, but products of
the formation of social constructs, which is, itself, guided by the expedien-
cies of the conflict dynamics. So Mitchell’s observation that the develop-
ment of conflictive attitudes frequently proceeds in a self-perpetuating and
exclusionary manner may be true but, without an account of how previous
experiences are used to reinforce or exaggerate individual outlooks, his
model lacks explanatory power – a criticism also pertinent to the appar-
ently cohering and distracting properties of conflictive situations presented
in Chapter 4.
At the collective level, similar forces may shape the construction, content
and delimitations of identity groups. Unlike innate models, these require
perpetual reinforcement to remain influential and are thus liable to change
markedly and rapidly. Such ideas have had a major influence on the way in
which soldiers are recruited and trained – both in a formal sense and in the
mobilisation of combat irregulars during the build-up to civil wars. Indeed,
the notion that the roles and identities that underpin wars are constructed
and ascribed rather than inflexible and fixed has been used to argue that
the patriarchal structures that underpin women’s vision of, and participa-
tion in, conflicts are not biologically based, but grounded upon the differing
socialisation experiences of boys and girls. This can, when either polarised
by elite manipulation or organised by seemingly benign social institutions,
result in both behavioural and structural violence. As such, the way in which
we learn to accept conflict and violence is difficult to separate from a general
acceptance of inequality and injustice. Clearly, the process of redrawing the
boundaries between ethnic groups in Rwanda, for example, cannot be fully
distinguished from the rise of Hutu political and socio-economic power. The
fact that such ascendancy relied, in Galtung’s terms, upon Western penetra-
tion, segmentation, marginalisation and fragmentation necessitated an ac-
companying ideological effort to ensure that the powerless remain partially
informed, passive and disunited (1990: 294). The learning methods that
underpin the formation of identity are therefore closely connected to the
notion of structural violence as a facilitative element in behavioural violence.
Their capacity to create an atmosphere of ordinariness can lower social bar-
riers to individual acts of violence (as Grossman points out), to collective
tyranny on a grand scale or, in keeping with Leo Strauss’ account of lib-
eralism’s mass malleability, a strata of political elites devoid of value–fact
distinctions.
7 Grievance
In the previous chapter, various ways in which political elites organise con-
flict and violence were considered. In Chapter 4, some of the reasons why
leaders might wish to do this were discussed, along with a range of ben-
efits that might accrue from such endeavours. As these mostly emerge once
the decision to become involved has been made, more analysis is needed to
understand what motivates participants in the first place. Here, research goes
back many years. Writing over 2,000 years ago, Aristotle, for instance, noted
that much social conflict was a result of grievances caused by the combined
effects of the fundamentally unequal nature of Athenian society and of the
perceived weakness and incompetence of the city’s leaders in responding to
this inequality. Comprehending the impact of grievances on human behav-
iour in general, and on the causes of conflict in particular, has made up a key
element of the social sciences ever since. This chapter looks initially at the
range of work, mostly emerging after the Second World War, which focuses
both on understanding the ways in which grievances are formed and on
the analysis of causal linkages between grievance formation and conflictive
behaviour. Of key significance here is the notion of relative deprivation and
its proposed relationship with frustration, action and rebellion – at the level
of both the individual and the group. The way in which these factors change
over time is also a fundamental concern. As the second section highlights,
it has often been noted that societies going through processes of ‘moderni-
sation’ or ‘development’ may be subject to greater disparities between the
satisfaction of human needs and the general desire of the citizenry to acquire
the perceived benefits of modernity. This can be seen as an important cause
of grievance formation, collective frustration and conflict.
Such appeals may, they continue (building on Gurr 1974: 229–31), take five
forms: (1) the establishment of a categorical identity, (2) the mustering of an
individual’s sense of relative deprivation, (3) the association of the existing
order with the source of discontent, (4) the normative defence of collective
action and (5) the utility of collective action. Although, in many cases, these
may be conceptually detached from the grievance–motive–action sequence
that underpins the approaches discussed here (and will therefore be returned
to in the next chapter), they remain an important element in understanding
how social movements maintain ‘an ongoing sense of legitimacy and efficacy
among movement cadres and members’ (McAdam et al. 1988: 722). This is
necessary not only to pursue group objectives (or values), but also to ensure
that deprivation continues to be associated with the level of the group and
not at the more dissonant level of the individual, which, as we have seen,
tends to provoke feelings of futility and self-doubt, leading to the percep-
tion that a grievance may be a result of unrealistic expectations, personal
shortcomings or the failure of the group to achieve a change in circumstance
(Turner 1969).
Much, of course, depends on the dynamic circumstances in which griev-
ances are formed. As we saw in Chapter 2, conflictive situations may change
rapidly. This is true not only of actual conditions, but also of individuals
and groups whose expectations can also grow sharply, thereby intensifying
perceptions of both egoistic and fraternal deprivation. After all, ‘the more
intensely people are motivated toward a goal, or committed to an attained
level of values, the more sharply is interference resented and the greater
is the consequent instigation to aggression’ (Gurr 1968: 257–8). Assuming
108 Grievance
that there is a notional equilibrium point at which a society is able to meet
its needs and therefore conflate, or at least narrow, any discrepancy between
its citizenry’s expectations and its capabilities (or vice versa), it is possible
to imagine a norm from which various scenarios of change – or growing
disequilibrium – can be distinguished. Gurr, after the work of Morrison and
Steeves (1967), points to three (1974: 46–58). The first, ‘decremental’ dep-
rivation, is illustrated in Figure 7.1. Here, expectations have not significantly
changed during a period in which people perceive themselves to be decreas-
ingly able to meet their needs. Such a sense of declining capabilities might be
caused by sudden shocks such as an economic recession, a collapse in civil
order, a natural disaster or a foreign invasion. It might also be the result of
incremental changes in the norms or beliefs of society leading to a lessening
of reciprocity and social capital or a gradual erosion of civil liberties by a
political elite. In either instance, people subject to decremental deprivation
develop grievances and become frustrated (and therefore prone to conflic-
tive behaviour) by the loss of what they once had.
John Booth’s study of Central America offers a comprehensive account
of decremental deprivation during the 1970s and early 1980s (1991). As
Table 7.1 illustrates, the entire region (except for Costa Rica) suffered both
a significant decline in real wages and a marked increase in unemployment
levels. In many areas, the concentration of land ownership intensified as
high cotton prices and an enlargement of beef production forced small-scale
farmers off their land and into saturated urban labour markets (Brockett
1988: 72–4). In Guatemala, for instance, the availability of arable land fell
from 1.7 to 0.8 hectares per capita between 1950 and 1980 (Hough et al.
1982). Costa Rica, in contrast, developed a comparatively successful land re-
form programme and, like Honduras, maintained higher welfare and lower
military expenditure levels than its neighbours. These factors combined to
reduce the impact of the 1970s energy crisis and to lessen rural labour vola-
tility. The result was that, in Nicaragua, El Salvador and Guatemala, vibrant
High
Collective value position
Value expectations
Value capabilities
Low
Time
Figure 7.1 Decremental deprivation. Source: Redrawn from Gurr (1974: 47).
Table 7.1 Unemployment levels (percentages) and real working class wage indices in Central America 1970–84 (1972 = 100)
Costa Rica El Salvador Guatemala Honduras Nicaragua
Year Unemp. Wages Unemp. Wages Unemp. Wages Unemp. Wages Unemp. Wages
1970 3.5 93 16.0 98 4.8 102 8.8 – 3.7 106
1980 5.9 125 24.0 87 5.5 79 8.8 101 17.8 56
1984 6.6 107 30.0 61 9.1 79 10.7 93 16.3 40
Source: Booth (1991: 42–3).
110 Grievance
and eclectic protest movements emerged from social groups most affected
by the economic downturn. White-collar workers, proletarianised by a sharp
drop in living standards, appeared alongside peasant groups and Church-led
organisations already opposing incumbent regimes and their cabal of clients.
While an increase in social activism was also noted in Costa Rica and Hon-
duras during this period, political fragmentation was mostly avoided and
instances of irredentist violence and state brutality remained significantly
fewer than in the rest of region. Booth concludes thus:
High
Value expectations
Value capabilities
Low
Time
Figure 7.2 Progressive deprivation. Source: Redrawn from Gurr (1974: 53).
they seem inevitable become intolerable when once the idea of escape from
them is suggested’ (1856: 214). Davies suggests that, during times of per-
ceptible improvement, expectations increase alongside capabilities, thereby
maintaining a tolerable gap between individuals’ wants and their ability to
achieve satisfaction. When an acute downturn occurs, however, there is, he
continues, an inevitable lag between material change and the collective re-
adjustment of expectation, producing a widening, and therefore increasingly
intolerable, gap between wants and satisfaction. Indeed, conflict participa-
tion, for Davies, ‘requires the continued, even habitual but dynamic expecta-
tion of greater opportunity to satisfy basic needs perpetuation’ (1962: 8).
The continued growth of expectation through the period of decline is, in
other words, necessary for frustration and a fraternal sense of deprivation
to occur.
prosperity,
Farouk takes throne and
1950–1
British troops withdraw Postwar unrest
to Suez, 1936
Egyptian
independence, War prosperity
1922
High
Social expectations
Social achievement
Low
Time
Figure 7.4 An improving J-curve. Source: Adapted from Feierabend et al. (1973:
407).
114 Grievance
High
Value capabilities
Low
Time
Figure 7.5 Aspirational deprivation. Source: Redrawn from Gurr (1974: 51).
This may happen in a number of overlapping ways. First, it is often the case
that such a change is prompted by the demonstration effect of modernisa-
tion processes’ differentiated influence. Some groups, in other words, ben-
efit disproportionately from change, leading to increases in inequality, envy
and resentment, grievances and on to conflict and violence (Nafziger and
Auvinen 2002: 156). Second, developmental processes inevitably involve
greater media linkages between the North and South and within developing
countries themselves. This makes previously unconnected people aware of
each other’s lives – particularly the apparently inspirational lifestyles of local
and Western elites.
Third, population increases and land shortages, which frequently result
from the modernisation of the agrarian sector, expose rural migrants ar-
riving in cities within developing countries to the Western lifestyles of the
urban wealthy and the expatriate community (Gurr 1973: 366). Fourth, po-
litical leaders may blame foreign powers and comprador classes (such as the
Gujarati community in Uganda during the 1970s) for domestic problems,
thereby highlighting differences in standards of living previously unnoticed
or disregarded. Fifth, the low administrative, economic and social capaci-
ties of many developing countries may prevent the state from responding
to development-led changes, with the consequence that capabilities fall be-
hind expectations (Holsti 2000). Finally, leaders may utilise the benefits of
development to increase their own prosperity or the material position of
their clients. This, in turn, can give rise to a social stratum of inordinate
wealth which then becomes a source of grievance and unmatched expecta-
tions (Keen 2000a: 292–4). As these changes are likely to occur in societies
undergoing rapid social change, frustrations may have a curvilinear relation-
ship with development. Countries at either end of the supposed tradition-
to-modernity continuum are, in other words, likely to be more stable than
those in transition. As the Feierabends explain,
Table 7.3 Political instability and Gini indices of land distribution in seven Latin American countries
Country Cuba Bolivia Colombia Argentina Venezuela Honduras Guatemala
Gini index 79.2 93.6 84.9 86.3 90.9 75.7 86.0
Political instability 2,900 663 316 217 111 111 57
Source: Adapted from Russett (1964: 451).
118 Grievance
100%
Equality
Land
Austria
Bolivia
0% 100%
Farms
Figure 7.6 Land distribution in Austria and Bolivia. Source: Adapted from Russett
(1964: 446).
precise link between inequality and conflict is, however, very problematic.
As Jonathan Goodhand has recently noted, ‘while there is some agreement
in the literature that conflict causes poverty . . . the argument that there is
a causal relationship in the reverse direction is more contentious’ (2003:
633–5). Findings generally disagree with one another, and single studies
frequently produce contradictory results. Ranveig Gissinger and Nils Pet-
ter Gleditsch, for instance, begin their study of the relationship between
globalisation and conflict with the comment that ‘a high level of trade does
generate more domestic peace; at the same time, direct foreign investment
also creates conditions conducive to political instability’ (1999: 327). Much
may depend on the nature of the material basis for such change and the kind
of inequality that is produced. The extraction and transformation of natural
resource endowments, an important concern of the next chapter, may, for
instance, lead to new grievances over issues of wealth distribution or, as in
the case of Angola, to an intensification of long-held political grievances
(Cramer 2003: 406). In such circumstances, heavily skewed patterns of land
ownership may, in situations where ‘the peasant lives in poverty and suffer-
ing’, make conflict ‘likely, if not inevitable’ (Huntington 1968: 375). This,
like ethnic or other communal linkages, may be considered a horizontal in-
equality, while disparities of income (also found to be an important causal
element in conflictive behaviour (Alesina and Perotti 1996)) can be regarded
as a vertical index of inequality (Murshed and Gates 2003).
In contrast, some writers have suggested that acute inequality does not
lead to conflict at all. Indeed, Paul Collier and Anke Hoeffler, pointing to a
statistically significant inverse relationship between the two variables, argue
that ‘greater inequality significantly reduces the risk and duration of war’
(quoted in Cramer 2003: 399). This, they explain, is because acute inequal-
ity implies the existence of a politically entrenched elite that is likely to
back, both materially and ideologically, the repression of any challenge to
Grievance 119
the status quo – from which the elite continues to prosper. Moreover, in
keeping with the premises of what Mark Irving Lichbach calls ‘the rational
actor scientific research program’ (also discussed in more detail in the next
chapter), potential protagonists are dissuaded from initiating conflicts aimed
at redistribution by observable increases in governmental/elite strength born
of inequality (1989: 459). As Collier concludes, ‘conflict is not caused by
divisions, rather it actually needs to create them’ (2000a: 13). These, as
Alejandro Portes notes of Santiago de Chile, may result from ‘the post-
factum self-legitimation of successful revolutionary movements’ (1971: 26).
Consequently, claims regarding preparticipatory grievances cannot be taken
as evidence of individuals’ actual motivations. This tends to undermine the
idea (central to the relative deprivation model generally and studies by Maro
Ellina and Will Moore (1990) and Douglas Hibbs (1973) in particular) that
political repression, as a grievance-inducing input, should be positively cor-
related with conflictive behaviour.
The result is, in many cases, an arbitrary tendency to classify people as
deprived and to see a wide range of post-frustration responses as aggressive.
In other words, comparatively affluent researchers may make social judge-
ments based on their own feelings of affinity and sympathy when record-
ing the sentiments and observing the behaviour of the poor (Bandura and
Walters 1963). Indeed, Richard Walters and Murray Brown (1963) found
that observers could be trained to adjust their evaluations of post-frustration
responses and that many used different criteria, based on past experiences,
to judge what amounted to ‘aggression’. Such objectivism can, as we saw in
defining conflictive situations and in the measurement of structural violence
(looked at in Chapters 2 and 3 respectively), become based upon overly
broad social judgements. These may then be deployed to label any indi-
vidual or group that, perhaps deliberately and contentedly, does not possess
‘sufficient’ material goods as inclined to behave aggressively and therefore
a potential threat to the established order. As Leonard Berkowitz puts it,
‘people have learned to categorise an individual’s rigorous behaviors . . .
as aggressive even when the frustrated person does not intend to commit
injury’ (1965: 306).
In fact, it would seem that, in empirical terms, the ‘vast majority of the dis-
advantaged do not engage in disruptive public protest’ (Bandura 1973: 169).
Throughout the urban upheavals in the United States during the 1960s, for
instance, it is estimated that only 15–20 per cent of residents from socially
deprived areas actively participated in the disturbances (Bowen et al. 1968).
The possible reasons why the poor (and, therefore, presumably aggrieved)
do not engage in conflictive behaviour are, however, largely ignored. As
Rod Aya notes of the work of the Feierabends, relative deprivation theorists
generally do not ‘specify the conditions under which expectations may be
frustrated without producing rebellion. [Instead] . . . they simply presume
a direct connection between frustration and revolt, and thus beg the ques-
tion they profess to have answered’ (1979: 57). Given that a number of
120 Grievance
studies have indicated that passivity is highest among the most deprived,
this criticism is particularly damaging. Edward Mitchell’s study of South
Vietnam, for instance, found that governmental control was highest (and the
recruitment powers of the Viet Cong were weakest) not in areas of compara-
tive prosperity, but in provinces where ‘few peasants farm their own land,
the distribution of landholdings is unequal, no land redistribution has taken
place, large French landholdings existed in the past, population density is
high, and the terrain is such that accessibility is poor’ (1968: 438).
Another key problem in associating frustration with conflict is the great
difficulty in measuring an individual’s (let alone a group’s) needs. How can
an accurate study of the social requirements for love and respect be ad-
equately operationalised? Even physiological needs are difficult to ascertain,
and there is little consensus among theorists on the comparative standards
used to analyse dissatisfaction. The notion of a need hierarchy of sustenance,
sexual gratification, self-actualisation and so on (which is fundamental to
many relative deprivation theorists) is especially problematic. As Peter Lup-
sha points out, such an ordering is likely to be ‘subject to change without
notice as one’s needs are satiated and others rise to the fore’, thereby making
longitudinal studies difficult to assess, compare and corroborate accurately
(1971: 93). Measuring and relating satisfaction levels are similarly challeng-
ing. What one subject might find highly satisfactory, another might consider
intensely frustrating. Yet writers such as Davies illustrate expected need satis-
faction as a straight-line function into which various revolutions are made to
fit. To do this, he and others are obliged to work backwards from the event
using post hoc and inferential methods, thereby undermining the prognostic
potential of their models.
An example of this kind of problem is the failure of the Feierabends’
basic assertion – that ‘once traditional societies are exposed to the modern
way of life, without exception, they desire benefits associated with moder-
nity’ (1972: 144) – to predict instability levels in the South. It is generally
acknowledged, for instance, that the last 30 years have seen a significant
increase in both adult literacy rates and the combined primary and second-
ary school enrolment ratio across much of the developing world. This, ac-
cording to the Feierabends’ reasoning, has enlarged the numbers of people
in contact with non-traditional concepts, thereby exacerbating a ‘process
during which former patterns of behaviour, outdated technologies, estab-
lished roles, status and norms must all give way to new, unfamiliar patterns’
and widening ‘the inevitable lag between aspiration and achievement’ (Feier
abend and Feierabend 1972: 144). The expected rise in ‘disruption, chaos
and personal discontent’, however, does not seem to have materialised (Feier
abend and Feierabend 1972: 149). Transitional countries are not, despite the
apocalyptic visions of Robert Kaplan and others discussed in Chapter 5, now
demonstrably more unstable than they were in 1970. Although there has
been an increase in civilian casualties that has disproportionately affected
the developing world, this is not a result of recent trends, but because poor
Grievance 121
and transitional countries have been home to the bulk of civil wars for most
of the post-Second World War era. Indeed, rather than a function of mod-
ernisation or development, these conflicts may be mainly a consequence of
Great Power policy, the spread of small arms, the enduring despotism of tiny
privileged elites often based on traditional societal structures or, as Stathis
Kalyvas contends, microfoundational logic (2006).
At the heart of this problem is, for critics of the relative deprivation re-
search programme, a basic confusion between a society’s tendency to de-
velop along modern lines and individuals’ concerns about their own life that
‘often occur within a still viable traditional context in which provision of a
minimum of economic, social and psychological security dampens whatever
frustrations unfulfilled demands generate’ (Oberschall 1969: 9). Indeed, in
Nigeria and Uganda, it was found that a high degree of exposure to urban life
and the media, which supposedly ‘prepare men’s minds for new desires more
rapidly than those new desires be satisfied’ (de Sola Pool 1966: 106), made
‘no marked difference in satisfaction level’ (Oberschall 1969: 12). Previous
positive correlations between these two variables tended to overlook a basic
difference between desires (such as owning a tractor) that are not seen as
likely to materialise and expectations (such as earning enough to feed one’s
family) which can be regarded as an imperative right. As Table 7.4 illustrates,
more subjects did not want, for instance, a car or truck than wanted one and
expected to get one. Consequently, Anthony Oberschall concludes that the
lack of ownership of a car or truck could not be regarded as an example of
relative deprivation. This, he explains, is because people living in developing
or transitional countries are unlikely to orientate their expectations towards
members of a comprador or overseas elite with whom they have little or
no social interaction. Instead, it is more likely that the assets of their own
immediate contacts will make up the primary reference against which their
own position is judged (1969).
Indeed, Derek Birrell reached a similar conclusion in his study of dep-
rivation levels in Northern Ireland (1972). Despite long being the poorest
Conclusion
This chapter has looked at some of the key debates regarding the formation
and marshalling of grievances as a causal element in individual and collective
conflict. First, it considered the notion of relative deprivation as a develop-
ment of human needs theory and the frustration–aggression mechanism to
which it owes much of its conceptual heritage. At its simplest, this is the
idea that people participate in conflicts because they are unable to satisfy
needs such as sustenance, security, affection and self-actualisation. There is,
in other words, a direct connection – initially conceived of in terms of frus-
tration always leading to aggression – between the inequitable distribution
of resources (both material and non-material) and the causes of conflict and
violence. Although this chapter has attempted to retain some of this explica-
bility within a particularly abstruse and fragmentary section of the literature,
it is important to note that the proposed relationship between perceived
needs, a frustrating instigator, a sense of grievance and conflictive behaviour
is, in fact, very complex.
There is, for instance, little agreement on whether or not perceived griev-
ances are derived from frustrated needs that are innate or socially acquired,
universal or culturally specific, rigid or fluid. Nonetheless, most writers are
agreed that those whose needs are not met may develop grievances in two
ways. The first – egoistic deprivation – is internally experienced and thus
tends to lead to individual responses such as a lowering of expectation, with-
drawal or, conversely, a reinvigorated pursuit of personal goals. The second
– fraternal deprivation – is experienced collectively and is therefore likely
to produce participatory responses including social movements, political
parties, trade unions and the like. In a sociological process related to the
integrative effects of conflict participation looked at in Chapter 4, much
depends on how deprivation is experienced and the web of group affiliations
with which an individual identifies.
Also relevant are the ways in which need satisfaction and individual or
collective expectations are changing over time – the primary concern of this
chapter’s second section. The ability of people to meet their needs may de-
Grievance 123
cline in relation to static expectations amid economic recessions. Equally,
people’s expectations may grow in a way that outstrips need satisfaction
during times of growth. The latter pattern is particularly relevant to periods
of political liberalisation, economic modernisation and social development,
all of which can increase expectations while also creating or exacerbating
collective differences – and therefore fraternal deprivation – within socie-
ties. As we saw in the case of Rwanda (discussed in Chapters 3 and 6), such
changes may lead to ‘transition’ periods in which inequality levels, collective
frustration, structural violence and the redrawing of social divides all inten-
sify, leading to a rise in instability, conflict and violence.
Specifying the causal dynamics of such relationships is, however, ex-
tremely difficult. As most countries might be said to be amid a period of
modernisation or development, instances of social instability do not reveal
any particular determinative influence. Indeed, problems in measuring hu-
man needs, deprivation and frustration (at the level of the individual, the
group and the state), coupled with the more fundamental problems of ob-
jectivism, have produced a contradictory body of inconclusive findings that
are difficult to link together longitudinally. Such problems have led some
analysts to question the value of basing analyses of conflict and violence on
the premises of need and frustration at all. Walter Korpi, for instance, states
that an overemphasis on grievance as a motivational cause of conflict and
violence ‘overlooks the actual possibilities for achieving the desired change’
and ignores the fact that achieving such an objective is ‘determined primarily
by the difference in power resources between the parties concerned’ (1974:
1569). This focus on the dynamics of collective mobilisation, along with an
associated emphasis on the expected utility of conflict participation, repre-
sents something of an elemental challenge to grievance-based approaches
and is thus the topic of the next chapter.
8 Mobilisation
In return for these benefits, individuals therefore come to value the preser-
vation of their group. They may overlook the opportunity to free-ride and,
for reasons of time, money and the need to pressure others to conform, they
may decide to commit their resources to the interests and objectives of the
group (Hechter 1987).
For Lichbach, this can happen in four, overlapping ways (1994b: 11–19).
The first he calls ‘uncoordinated market exchange relations’. These may act
to increase the benefits accruing to individual participation. If, for instance,
a group contains a high proportion of zealots (individuals for whom the
benefits of their contributions exceed the costs), then it is likely to mobilise
more easily – particularly if the object of mobilisation is made more valuable
through scarcity or policy (Olson 1971: 49). Similarly, a group’s capacity to
reduce or repay an individual’s participatory costs, such as start-up expenses
or forgone income (to be discussed in greater detail in the next section), may
permit it to mobilise with less difficulty (Rogowski 1985). Obviously, this
will be more likely to occur if the group can increase its resources, suggesting
that mobilisation becomes more probable when organised by prosperous
sections of society or when supported by larger concerns such as national
organisations, wealthy diasporas, sympathetic patrons or overseas states
(Azam 2001).
Additionally, an individual’s decision to participate may be influenced by
the perceived efficiency of the group’s manifesto, resources or plans; if, in
other words, more benefits can be obtained for a fixed cost (in terms of
time, money and so on) through advances in technology and refinements in
tactics (Muller and Opp 1986). These factors relate closely to the tendency,
130 Mobilisation
examined in Chapter 4, of conflict to exert a cohering effect upon members
of participant groups. The use of incomplete or deliberately manipulated
information, for instance, can be crucial in portraying groups’ objectives
favourably and improving their perceived likelihood of success, thereby so-
lidifying internal support and bolstering external recruitment. Misapprehen-
sions regarding the potential costs of becoming involved might inspire zeal-
otry and increase risk-taking, while a misleading impression of the group’s
internal structure can convince individuals that their participation could
make a material difference to the outcome of mobilisation (Moe 1980). Re-
stricting information or other material resources can also prevent individu-
als from selecting alternative options and increase general perceptions of
competition and polarisation (Gates 2002).
A second, inter-related solution to the free-riding dilemma is to be found
in groups that enjoy high levels of solidarity, strong institutions and a wide-
spread sense of common purpose (Taylor 1988). Here, the dissemination of
knowledge is important in the construction of a collective ideology, which
can promote conformity and the experience of self-actualisation, enhance
an individual’s capacity to endure hardship and add ethical dimensions to
the goals of mobilisation (Defronzo 1991: 314). Indeed, groups that con-
tain shared expectations based upon qualitatively similar ideas tend to make
comparable assessments of utility, especially if their members are subject to
appeals and pleas for action of the type discussed in Chapter 7. Addition-
ally, a broadening awareness of this mutuality through the evident engage-
ment of others can produce sequential forms of mobilisation known as
bandwagon, threshold or critical-mass examples (Oberschall 1979). Third,
free-riding may be overcome through the establishment of authoritative
rules within the mobilising group. These are, as Chapter 5 highlighted,
particularly important if groups are drawn from overcrowded and socially
diverse environments where personal autonomy is expected to be high and
reciprocity low. At a political level, such an oversight can help to ensure the
contingent cooperation of individuals who fear that their involvement will
not be matched by others, while, at an economic level, the enforcement of
exchange conventions can assist in solidifying a group’s resource base and
further consolidate mobilisation (Ostrom 1990).
Fourth, ‘a little power, authority, and, yes, dictatorship, can go a long
way toward solving the Rebel’s Dilemma’ (Lichbach 1994b: 17). In addi-
tion to the institutionalisation of doctrinaire regulations, personality cults
around charismatic and authoritarian leaders can facilitate mobilisation by
consolidating interpersonal bonds of expected reciprocity and thus helping
to alter an individual’s assessment of the potential costs of trusting others
(Hardin 1995). This may be strengthened by making the group more exclu-
sive through the formation of a centralised, compact revolutionary bureau,
the greater involvement of zealots or the establishment of a decentralised
network of small, highly coherent cells (Weede and Muller 1998: 45–9).
Top-down coercion, in the form of punishing desertion and sloth and enforc-
Mobilisation 131
ing stringent barriers between the group and the broader context in which it
operates, can, along with the types of direct and vicarious conditioning con-
sidered in Chapter 6, increase individuals’ sense of identity and lead them
to believe that successful mobilisation is dependent on their participation
(Moore 1995: 440–2). In Mozambique, for instance, young men recruited
to the Resistência Nacional Moçambicana (the Mozambican National Resist-
ance – RENAMO) were reported to have been forced to attack their home
village, thereby permanently severing their societal links and rendering them
more reliant on the group (Cohn and Goodwin-Gill 1994: 23).
Although these four sets of measures may be successful in offsetting the
effects of free-riding, they do not fully overcome a fundamental difference in
preference between the mobilising elite and the mobilised masses. As Patrick
Regan and Daniel Norton observe, ‘leaders seek authority and control . . .
they view success in terms of a new distribution of political power, whether
that is through a power-sharing arrangement or outright and total control
over the bureaucracy’. Those enlisted, they continue, ‘are motivated by
personal gains in the form of a minimally accepted improvement in their
personal standard of living’, suggesting that the distribution of ‘protection
and resources in the provision of selective benefits’ is extremely important
for sustained mobilisation to occur (2005: 323). This is very significant be-
cause, as we shall see in the next section, ‘an observable increase in the
self-financing nature of combatant activities’ since the end of the Cold War
has led to further inroads in the predominantly grievance-based idea that
warfare represents a frustrating breakdown in the supply of economic or
political public goods and is thus an instigator of restorative action (Bal-
lentine and Sherman 2003: 1). Rather, the imperative of ensuring an unbro-
ken supply of remuneration is seen as the primary means of ensuring group
integration, thereby transforming a war-affected economy into ‘an alternate
system of profit, power and protection’ (Berdal and Keen 1997: 797). As
Michael Pugh and colleagues conclude, ‘the conditions of war present new
commercial opportunities for the exploitation of assets, investment, services,
marketing, and welfare. Indeed, . . . armed factions are remarkably adept
at economic diversification and at seeking optimum gains in the changing
contexts of their struggle’ (2004: 19).
Kenya
Education 18.4/19.6 22.2/19.8 18.0/13.8
Health 12.2/13.0 10.3/9.2 9.8/7.5
Zambia
Education 15.8/10.2 8.6/5.8 5.5/4.0
Health 7.2/4.6 4.3/2.9 1.2/0.9
DR Congo
Education 5.6/2.9 1.1/2.0 0.0/0.0
Health 3.9/1.8 0.7/1.2 0.0/0.0
Congo
Education 56.5/18.7 39.5/9.8 19.2/4.9
Health 21.2/7.0 14.2/3.5 12.0/3.1
Source: Reno (2000b: 441).
for the selective benefits that they bestow. It may also result from collective
attempts to control the supply of commodities across borders and frequently
on to Western markets (Cater 2003: 32–3). In each case, the dynamics of
mobilisation are ‘deeply interconnected with both regulated and unregu-
lated global trade and financial flows’ (Pugh et al. 2004: 19).
Here, the presence of natural resources – a third factor in the emergence
of war-affected economies – is held to increase the destructiveness of con-
flicts once they have begun (de Soysa 2000). The availability of primary com-
modities may, it is suggested, ‘encourage [the] raiding of civilians to meet the
needs of the fighters’, giving rise to ‘competition and conflict over access to
and distribution of the economic proceeds of resource exploitation’ (Keen
1997: 72; Sherman 2003: 225). As Table 8.3 illustrates, these commonly in-
clude oil, natural gas, timber, gemstones, narcotics and timber. For instance,
the União Nacional para a Independência Total de Angola (The National Un-
ion for the Total Independence of Angola – UNITA) has controlled over 70
per cent of the country’s diamond exports during much of the country’s civil
war (Berdal and Malone 2000: 5). Similarly, the National Patriotic Front
leader, Charles Taylor, is said to have made hundreds of millions of dollars
a year (between 1992 and 1996) from his organisation’s administration of
134 Mobilisation
Table 8.3 Civil conflicts linked to resource wealth, 1994–2001
Duration Resources
Afghanistan 1978–2001 Gems, opium
Angola 1975– Oil, diamonds
Burma 1949– Timber, gems, opium
Cambodia 1978–97 Timber, gems
Colombia 1984– Oil, opium, coca
Congo 1997– Oil
DR Congo 1996– Copper, cobalt, diamonds, gold, coltan,
coffee
Aceh 1975– Natural gas
West Papua 1969– Copper, gold
Liberia 1989–96 Iron, diamonds, gold, timber, coffee, palm
oil, cocoa, rubber, marijuana
Papua New Guinea 1988– Copper, gold
Peru 1980–95 Coca
Sierra Leone 1991–2000 Diamonds
Sudan 1983– Oil
Source: Ross (2003: 49).
We have always maintained that the conflict in Sierra Leone is not about
ideology, tribal or regional differences. It has nothing to do with the
so-called problem of marginalized youths or, as some political com-
mentators have characterized it, an uprising by rural poor against the
urban elite. The root cause of the conflict is diamonds, diamonds and
diamonds.
(New York Times 6 July 2000)
Conclusion
This chapter has traced the emergence of a fundamental challenge to the
grievance model of explaining the causes of conflict. Rather than viewing col-
lective violence and social movements as emotional, cathartic and affective,
they are seen as rational, considered and self-regarding. Proceeding from a
Hobbesian premise of the avaricious nature of humankind, conflicts may be
seen as loosely organised predation with the availability of lootable resourc-
es a key consideration in individuals’ cost–benefit analysis of the prospects of
participation. Owing much to visions of human behaviour drawn from the
study of economics, these accounts stress the importance of resources – both
as a requisite element in the instigation, and as an objective, of mobilisation.
The key determining factor in the causes of conflict behaviour, then, is not
142 Mobilisation
the provision of public goods, the meeting of human needs or the manage-
ment of expectation, but the quality of the regime in which mobilisation
occurs. A highly repressive political system will, it is thought, act as a deter-
rent by imposing high costs upon, and offering little likelihood of success to,
prospective protagonists. Similarly, a very open polity may offer dissenters
alternative and lower risk/cost strategies through which to pursue change. It
is, therefore, regimes that are sufficiently exclusionary to provoke reaction,
yet insufficiently oppressive to deter action, that are predicted to suffer the
highest levels of irredentism.
Much, of course, also depends on the way in which mobilising groups
are internally organised. Elites may seek political objectives and positions of
power, whereas foot soldiers are seen as more likely to be concerned with
short-term material gain and remuneration. Here, then, greed and grievance
may not be fully separable. As we saw in Chapter 7, the capacity of leaders
to establish a categorical identity and to associate the existing order with
discontent as a normative defence of collective action might be as much a
means of mustering an individual’s sense of relative deprivation as persuad-
ing her or him of the utility of collective action. It could be, then, more
prudent to follow Jonathan Goodhand’s advice to ‘account for greed, but
don’t ignore grievance’ and thereby to adopt a more nuanced account of the
causes of conflict and violence (2001: 39). As Will Moore and Keith Jaggers
conclude,
Rejecting the countervailing hypothesis that such tightness may decrease sys-
temic uncertainty and therefore make it less likely that war can emerge from
Crises 151
crisis escalation, Bueno de Mesquita concludes that 84 per cent of wars in
the twentieth century began following periods of rising intrabloc coherence.
In direct contradiction of such findings, however, are studies that emphasise
the very low reliability rates of alliances (Sabrosky 1980). In fact, the inci-
dence of honouring state commitments during times of war may be as low as
25 per cent (Leeds et al. 2000: 686). Whatever the truth, the relationships
that these present are, of course, correlative rather than causal and so other
factors cannot be excluded as possible determinative influences. It may be,
then, that, as other accounts have proposed, ‘alliances neither limit nor ex-
pand conflicts any more than prevent them’ (Liska 1962: 138).
Similar problems afflict the study of warfare’s relationship with arms
races. Michael Wallace finds that, of 28 crises involving arms races between
1816 and 1965, 23 escalated to war, whereas, of the 71 crises that occurred
unaccompanied by arms races over the same period, only three resulted in
war (1979). He goes on to argue that arms races are significant predictors of
wars regardless of which of the parties acquires the greater or more effec-
tive weapons (whether or not, in other words, a major Power or revisionist
states emerge as superiorly equipped), thereby rejecting the peace-through-
strength maxim of deterrence theory (1982). Almost immediately, however,
these figures were challenged. Paul Diehl, one of the researchers who com-
piled the dataset that Wallace used, ‘discovered that only one-fourth of the
disputes preceded by military build-ups resulted in war, while ten of the
thirteen wars [under analysis] occurred in the absence of joint arms increases
by the dispute participants’ (1983: 210).
Much relies on how the variables are assessed and coded. In order to
avoid the problem of having to aggregate military capabilities, it is, for in-
stance, common to find large and complex wars analysed as a series of dyadic
disputes. The First World War may thus be considered to be either seven or
eight integrated, or 26 separate, wars depending on how the writer deals
with alliance structures (Weede 1980). The time lag between variables is also
highly significant. By changing Diehl’s approach from a three-year to a five-
year gap between increasing arms expenditure and war (and by controlling
for the presence of nuclear weapons as a variable with exceptionally deter-
ring properties), Susan Sample found that 47 per cent of the disputes that
had been thought not to escalate to war did in fact do so (1997). As Henk
Houweling and Jan Siccama note, however, these types of study do not dis-
tinguish between arms races as reflections of broader tensions, as catalytic
agents of strain in the international system and as primary causes of warfare
(1981). Without an account of the mechanism of change, it is very difficult to
eliminate the possibility that states perceive war to be imminent and prepare
accordingly, thereby rendering arms races merely an effect of other causal
factors (Diehl and Kingston 1987). Similarly, mutual military build-ups are
commonly assumed to be arms races even if it cannot be demonstrated that
such expenditure was directed at an opposing party or parties. As Paul Diehl
152 Crises
and Mark Crescenzi point out, both phenomena could be ‘manifestations of
the enduring rivalries [between states] and thus not directly related to each
other’ (1998: 113).
To understand governmental strategy in this regard, it is necessary to
have a clear comprehension of the intrastate dynamics that underpin arms
races. Yet much of the literature overlooks such factors; as, for example, in
Richardson’s highly influential technical model (1960a; McGinnis 1991).
Policy is, after all, ‘subject to multiple decision-making processes, each one
revolving around a group of relevant individuals, and then must undergo
some means of aggregation to arrive at a collective outcome’ (Bolks and
Stoll 2000: 582–3). Accumulating human inputs in this way may be subject
to a variety of factors that normally govern the treatment of crisis responses.
These influences, which Robert Jervis defines as ‘pressures that the actor
would not admit as legitimate if he were conscious of them’, will be looked
at in more detail in the next section (1969: 240).
ExCom’s ultimate decision to opt for a naval blockade and not air strikes
was, in Allison’s view, also a function of intrabureaucratic wrangles. It was,
he suggests, based on U-2 intelligence flights carried out after a number of
delays caused by the Central Intelligence Agency and the United States Air
Force being unable to reach agreement over whose responsibility the flights
were. Had they been undertaken earlier, it is possible that ExCom might
have favoured a less coercive option. Had the flights taken place later, then
the missiles might have been in place, making a naval blockade valueless.
Moreover, it is possible, Allison contends, that some of ExCom’s early vacil-
lations were a result of the Air Force having no contingency plan for such a
scenario. As they did have an organisational process – a standard operating
procedure in other words – for carpet-bombing Cuba in readiness for a land
158 Crises
invasion, Air Command repeatedly referred to the need for some 500 sorties
when only nine known missile sites existed, leading to considerable confu-
sion among the members of ExCom. Indeed, a key reason for dismissing the
air option was that the missiles were reported, again on the basis of standard
classificatory procedures, to be mobile – a fact that turned out, shortly after
the decision to impose a naval blockade had been reached, to be untrue. In
accordance with Allison’s predictions, the case for an air strike was not reo-
pened. Furthermore, once the blockade was in place, ExCom had great diffi-
culty in imposing its decisions upon the organisational echelons below it. For
instance, Kennedy, hoping to give the Soviet leader Nikita Khrushchev more
time to reach a decision, ordered the Navy to draw in its ships away from
the approaching Soviet vessels to 500 miles off Cuban waters. However, as
Krasner notes, ‘the Navy being both anxious to guard its prerogatives and
confronted with the difficulty of moving large numbers of ships’ failed to
implement the directive promptly (1972: 177–8; Smith 1980).
Conclusion
This chapter has outlined some of the key debates regarding conflict causal-
ity within international relations. It has focused on the dynamics of crises,
the way they emerge and their effects on state behaviour. Here, the literature
is highly fragmented with most writers preferring to undertake empirical
tests of particular variables without seeking to amass a coherent body of
theory. For this reason, presenting a varied account of conflict causality is a
challenge. However, two aspects have been considered in more detail as pos-
sible causal elements in the onset of interstate war – alliance structures and
military build-ups. Both have been analysed in terms of a range of domestic
and international pressures under which political elites take on treaty obliga-
tions and initiate armament expenditure programmes.
This chapter then looked at the dynamics of crisis escalation as an ex-
planation of how alliances and military spending may lead to polarisation
and arms races. Here, competing visions of national interest are important.
These include the notion of deterrence as a means of preventing war in an
ostensibly hostile geopolitical environment. Leaders may thus respond to
a perceived rise in national insecurity or international tension by pursuing
various forms of agreement with other states or by acquiring new weaponry
and/or updating their existing arsenal. Although these arrangements may be
ostensibly aimed at deterring others – thereby making escalation less likely
– they may threaten other states, which then respond with similar measures,
leading to a spiral of negative reciprocity, bloc polarisation and arms races.
This ‘security dilemma’ is rooted in the rational decision to forge treaty
commitments and to acquire weaponry in order to feel more secure, yet re-
sults in the irrational outcome of accentuated threat perceptions and greater
relational tensions.
The exact role of deterrence and the security dilemma in intensifying
Crises 159
or defusing crises is open to considerable debate, as is the precise relation-
ship between weapons, alliances and the causes of war. Such differing views
have rendered a clear understanding of conflict causality difficult to obtain,
particularly as the historical record appears to support various positions,
depending on how the empirical data are treated. Much clearly relies on
the bellicosity of state leaders – a factor which, in turn, depends on the way
in which decisions are made by governing elites. Traditionally, analyses of
this process tended to see the higher echelons of national administrations
as unitary actors seeking to maximise a clearly understood set of values.
However, as Graham Allison’s well-known study of the Cuban missile crisis
demonstrates, such a rationalist approach may not represent the actual dy-
namics of decision-making very accurately. As many have noted, the essential
weakness here is that, without an explicit account of decisional mechanisms
within states, associations between disputes and war will inevitably remain
correlative rather then causal. Indeed, organisational inertia and the dispa-
rate character of bureaucratic politics may combine to restrict elites’ capacity
to pursue their goals, thereby producing disjointed increments rather than
categorical and imminently applicable judgements.
For this reason, the second section of this chapter has focused on analyses
of crises that are not limited to rationalist understandings of the state as
value-maximising unitary entities. Decision-making processes that place po-
litical elites within broader organisational and bureaucratic structures were
considered. The Cuban missile crisis was discussed as a pre-eminent example
of how such an approach could cast new light on the connection between
individual rationality, group dynamics, implemented outcomes and many of
the issues looked at earlier in the chapter. It involved large states with consid-
erable military and economic reserves, a history of conflictive relations (most
recently in the abortive Bay of Pigs invasion and more generally in south-east
Asia) and extensive borders potentially contiguous with members of the op-
posing bloc. Both Cuba and the United States perceived the crisis to involve
a wide range of vital interests, concerning the homeland security of both
themselves and their allies. Although there was not a significant horizontal
escalation of the crisis, vertical escalation involved a high degree of magni-
tude and, with the deployment and testing of American ballistic missiles in
Italy and Turkey continuing throughout the autumn of 1962, considerable
intensity. Seeing these events in acute and precipitous terms is, however, not
the only way to understand their broader implications. The contest between
the Soviet Union and the United States against a backdrop of periodic con-
frontation and numerous proxy conflicts may be seen, instead, as part of a
long-term crisis intrinsic to global leadership. Put another way, crises could,
as Chapter 10 will explain, be regarded as key markers in an age-old struggle
between an incumbent hegemon and its nearest rival, with each prepared to
deploy sufficient armaments to provoke a global war aimed at preventing or
facilitating a change of world leader.
10 Hegemony
a new set of products is introduced that sells well, the market expands,
and related employment swells, allowing for expansion of worker/
‘consumer’ spending. The market eventually becomes saturated, sales
drop, income contracts, and workers are laid off. . . . But eventually, the
excess inventory is sold out, production resumes, and renewed growth
is possible.
(1995: 404)
According to the French physician Clement Juglar, this process may oc-
cur approximately every 7–11 years at the point when most production
machinery is coming to end of its working life and firms are synchroni-
cally beginning to order replacements (1966). Examples of Juglar recessions
include the American economic downturns of the mid-1970s, early 1980s
and early 1990s. Simon Kuznets also traced cyclical patterns in a number
of diverse economies in Japan, Belgium and Argentina (1971). These are
between 20 and 25 years in length and have been explained in a number of
ways, including migration and generational fluctuations in housing demand.
Perhaps the most influential of the cycle economists, however, was Nikolai
Kondratieff. He noted oscillations in price indices over a period of 40 and
60 years between 1795 and 1925 that he believed may rest on the periodic
need to incorporate new technology through large-scale capital investment
(1984). This tends to take place, he maintains, initially in key industrial sec-
tors ((1) cotton, textiles and iron (1780–1817), (2) railways, steam and steel
(1840–75), (3) electricity, industrial chemistry and the internal combustion
engine (1890–1920)) before becoming diffused throughout the economy
(Rostow 1975). Once this occurs, overall investment slowly moves away
from production and towards speculation, thereby inhibiting demand and
gradually depressing the economy as a whole (progressively compressing,
in other words, the peaks and troughs of intracyclical Juglar and Kuznets
patterns) (Berry 1991). According to Joseph Schumpeter, who revised Kon-
dratieff ’s work to place a greater emphasis on innovation rather than invest-
ment, the effects of this type of cycle can be seen in the recessions of 1825,
1873 and 1930 (1934).
The relationship between these cycles and conflict is complex and has
long been subject to extensive debate. Writers such as Johan Åkerman have
suggested that, as Kondratieff booms appear to culminate in periods of mili-
tary confrontation, it is the frequency of warfare that explains the apparent
Hegemony 163
periodicity of economic change (cited in Eklund 1980). This is because wars
tend to increase short-term spending and demand while shifting production
away from non-military goods, thereby forcing prices up rapidly and prompt-
ing a Kondratieff upswing. Once demand lessens following the conclusion
of the war, limitations to production (commonly exacerbated by shortages
of labour, foreign exchange, equipment and so on), push prices down, de-
press the economy and generate a Kondratieff downswing (Bernstein 1940:
524–7). Indeed, as Table 10.1 shows, ‘there can be little doubt that wars
and their associated economic disruptions do lead to price level increases’,
at least during those in which the United States and Great Britain have been
involved (Thompson and Zuk 1982: 633).
Kondratieff himself, however, held the view that cyclical economic pat-
terns may cause wars. He suggested that, because periods of extended com-
mercial enlargement tend to destabilise political systems by accentuating the
international competition for raw materials and markets and by intensifying
domestic struggles over wealth distribution, ‘the most disastrous and ex-
tensive wars and revolutions occur’ during the upswing section of the cycle
(1935: 111). Quincy Wright, working during the Second World War, broad-
ly agrees. He argues that economic cycles, synchronous with Kondratieff ’s
findings, may, in part, be a cause of warfare concentrations ‘in approximately
fifty-year oscillations’ with especially intense periods of conflict occurring
alternately (1965: 227).
This prompted Ludwig Dehio to construct a centenary cycle of global
conflagration (1962). France, he argues, initiated the first challenge for
supremacy in 1494 by attacking the Venetians and beginning a prolonged
struggle over Italy. This benefited Spain under Charles V (who inherited the
Table 10.1 Selected British and American wars and wholesale price index
fluctuations (1913 = 100)
War War peak Preceding 5-year mean
Great Britain
Seven Years 93 78
American Independence 110 96
Napoleonic 201 157
Crimean 118 92
First World War 228 94
United States
Mexican 88 80
Civil War 189 97
Spanish 69 70
First World War 188 102
Second World War 152 115
Source: Adapted from Thompson and Zuk (1982: 633).
164 Hegemony
Habsburg throne in 1515), which, led by Charles’ son Philip II, succeeded
in obstructing France’s ascendancy by defeating Portugal in 1580 and as-
suming control of Venice’s trade routes (de Oliveira Marques 1972). Spain’s
hegemony was, however, immediately qualified by England’s destruction of
her navy in 1588, leading to a long period of gradual decline and ending
in a conclusive land battle at Rocroi in 1643. Her place was slowly taken
by a resurgent France following Louis XIV’s coronation in 1661. Again,
England led the rest of Europe’s resistance and, by combining forces with
the Netherlands, eventually halted French ascendancy in 1692, before suc-
cessfully preventing her from returning to power by winning the War of the
Spanish Succession in 1713. This ushered in another century of stalemate
before France again rose to hegemony in 1792. Abruptly deposed in 1805 at
Trafalgar by Britain and then crushed in Moscow in 1812, France’s authority
was replaced by a century of Anglo-Russian competition, which ultimately
permitted the expansion of German power – a process curtailed in 1918
(again by Britain operating in concert with a number of other Powers). In di-
vergence from the previous pattern, however, the victorious coalition failed
to take control of the international system and could not prevent a second
German challenge from leading to another, even more destructive period of
concentrated war (North and Thomas 1973).
Dehio’s ‘four-crested wave of regional power concentration and domina-
tion, with its delineation of peaks and troughs and waves of varying shapes’
has been highly influential (Thompson 1992: 135). Arnold Toynbee, for
instance, also structures his account of the last 500 years around the drive
to achieve hegemony. His 115-year cycle of ‘general war’ (1494–1525,
1568–1609, 1672–1713, 1792–1815 and 1914–18) corresponds broadly
with Dehio’s ‘drives for world domination’ (Toynbee 1954). Like Dehio,
he regards each cycle as initiated by an ambitious continental Power (the
Spanish Habsburgs, the French twice and Germany). For Toynbee, however,
this is not driven by relative expansions in military strength. Instead, hege-
monic patterns of conflict and violence are a reflection of more fundamental
cultural and political changes: notably, the generational transference of war
fatigue as a constraint upon martial bellicosity – elements also important to
Pitirim Sorokin’s historiographic work (1957) and Lewis Richardson’s study
of international crises (1960b). ‘The survivors of a generation that has been
of military age during a bout of war will’, Toynbee writes, ‘be shy, for the rest
of their lives, of bringing a repetition of this tragic experience either upon
themselves or upon their children’. Therefore, he continues,
two key explanatory variables, relative power and the degree of satisfac-
tion with the international order (or status quo) . . . [is] the primary
determinant of war and peace. States that have insufficient capabilities,
no matter how dissatisfied with the status quo, will be fundamentally
unable to challenge the dominant power. States that are powerful but
satisfied will have little motivation to challenge the dominant state for
its preeminent position and the accompanying ability to shape the inter-
national order. Only the powerful and dissatisfied pose a threat.
(DiCicco and Levy 1999: 682)
Ottoman Empire
United States
France
Netherlands
Sweden
Britain Russia/Soviet Union Prussia/Germany China
0
Y
–2
–4
–6
–8
0 50 100 150 200 250 300 350 400 450 500
Year
Figure 10.2 Randomly generated stationary data. Source: Redrawn from Beck (1991:
460).
Hegemony 177
needs’ (1980: 63), then it may be possible to prevent – or at least – delay
the next phase of world war. After all, it may be that the ‘great wars of the
system were not a consequence of its normal operation’ and thus the current
international order is not ‘a system which will inevitably generate hegemonic
wars’ (Holsti 1985: 682, 684).
Kugler and Lemke point to four such preventative measures. First, as China
will soon be in a position to challenge the United States’ hegemony, efforts
must be made to ensure that she is satisfied with, and therefore a supporter
of, the status quo. Second, because the historical record suggests that wars
are fought between hegemonic contenders despite the costs being extremely
high, efforts to restrain the proliferation of weapons of mass destruction
should be redoubled. Third, the North Atlantic Treaty Organisation should
be enlarged to include eastern European states in the short term and then
eventually Russia and China. This would expand the number of satisfied
states within the international system and produce a bloc too powerful to
be challenged by a rising contender. Fourth, in order to prevent regional
hierarchies from leading to major wars, greater multilateralism should be
promoted as a means of generating consensual support over the status quo
and predicting possible flashpoints in the future (2000: 158–9).
Conclusion
This chapter has presented a consideration of the big picture – the centennial
rise and fall of hegemonic Powers since the emergence of the modern world
system in the fifteenth century. It has looked at the relationship between
hegemony and the incidence of conflicts in general and large-scale warfare
in particular. It has reviewed the ways in which changes in economic and
military power have influenced the course of international relations over the
last five centuries. Here, conflicts are caused either by incumbent hegemons
protecting their standing in a preventative war against an ascendant Power
or by rising challengers initiating a confrontation in order to achieve the
status of world leader. As the benefits of such a position are vast, each is
predicted to be prepared to suffer considerable losses. Resultant conflicts
are therefore typically pursued on a colossal scale; a fact that renders them
comparatively uncommon. Indeed, in plotting the incidence of these rare
conflagrations, analysts, drawing on economists’ work on the periodisation
of boom and recession, frequently propose that a pattern of cyclical perio-
dicity is apparent. Here, fluctuating levels of naval capability are often seen
as especially important in determining the succession of hegemonic powers
and the accompanying occurrence of severe periods of conflict.
Associated with the study of these shifts in coercive strength are expla-
nations of warfare based on power transitions – broadly the focus of this
chapter’s second section. These generally concentrate on linear patterns of
capability assessed by variables such as military strength, differential econom-
ic growth rates and satisfaction with the regional or global status quo. Many
178 Hegemony
predict that war is most likely to begin when the relative might of two or
more disputatious Powers is approximately equal. It is at the point of power
parity that a rising challenger might expect to secure victory. An alternative
way of analysing a similar process is to assume that the dominant Power
or Powers will act to arrest the challenger’s ascendancy before it becomes
an imminent threat. These approaches and long-cycle theories both tend
to focus on the determinative properties of states’ internal characteristics.
As such, they offer something of a challenge to the assumptions of classical
realism, which hold that global conflict tends to be a consequence of three
inter-related features of the system itself – the inherently uneven and chaotic
nature of geopolitics, the changing objectives of national interest and states’
constantly shifting relative power to project their military force abroad.
Nonetheless, there is little agreement among writers in this field over both
the precise timings of shifts in the international balance of power and the
identity of the variables that drive such fluctuations in state capabilities. This
widespread divergence of opinion is due, in part, to questions over the very
existence of cyclical periodicity. It has been noted, for instance, that attempt-
ing to draw inferences from the historical record is inevitably epistemologi-
cally weak – the fact that something can be demonstrated to have happened
regularly in the past is not, by itself, an assurance that it will continue to
happen in the future. Such reservations notwithstanding, however, the issues
at stake in the investigation of the causes of global warfare are of such mag-
nitude as to ensure that these approaches remain an important area in the
study of international politics in general and conflict in particular. Indeed, it
is particularly pertinent today when globalised changes to the world order
now mean that ‘few, if any, issue areas . . . remain effectively closed off from
system dynamics’ (Marshall 1999: 7).
Most writers are, for instance, agreed that the weight of the past does not
make another world war inevitable at a given point in the future. Globalisa-
tion, the unprecedented scale of the last great conflict and the uniquely deter-
rent properties of nuclear weapons may all combine to make the hegemony
of the United States different from earlier periods of leadership. To prevent
a new hegemon from emerging amid the customary bloodshed, it may, for
instance, be necessary to make certain that the next great challenger remains
content with the status quo, that mass destruction armaments do not fall into
the hands of the dissatisfied and that more egalitarian forms of multilateral-
ism are promoted. After all, to fail in these regards and to permit the world
to pass through another calamitous period of hegemonic change could have
uniquely fateful consequences. As George Modelski and Patrick Morgan put
it, should a systemic war ‘occur again amidst a plethora of nuclear weapons
it will mean destruction on a scale endangering the existence of civilisation
and the human species’ (1985: 394).
11 Conclusion
This chapter will present a review of the overarching themes of the book.
Extending the work of Paul Rogers and Oliver Ramsbotham (1999), it fo-
cuses on seven overlapping centres of debate. These concern, first, the trans-
formation of conflictive situations. Here, important problems relate to the
definition of conflict and the role of the researcher. Second, an attempt to
connect analyses of conflict and violence with broader debates across dis-
ciplinary boundaries generates a number of issues of coherence within the
literature. These relate to matters of both principle and empirical measure-
ments. Third, the implied or explicit association between studies of conflict
and the achievement of peace raises important questions regarding the na-
ture of change and the utility of violence. Fourth, the use of multi-levels of
analysis presents debates over the applicability of theoretical models and the
relationship between the domestic and international social spheres. Fifth, an
aspiration to incorporate approaches from various cultural traditions into
the analysis of conflict and violence has valuable implications for both re-
search and policy. Sixth, a tension between empirical and normative elements
of research provokes notable debates over the relationship between values
and analysis, particularly in the areas of methodology and epistemology. Fi-
nally, the issue of the practical or policy implications of research prompts
a number of significant ethical considerations, not least in what precisely
theory should seek to cover, how academics should relate to decision-makers
and the degree to which researchers become complicit in policy outcomes.
Each of these areas will be looked at in turn with the aim of reflecting on the
main findings of the book.
The result is that the study of ‘armed conflict and war remains fragmented
between disciplinary boundaries, which produce conflicting and often mutu-
ally exclusive theories’ (Porto 2002: 1).
Similarly, at the empirical level, a proliferation of datasets proceeding
from varying definitional bases, using competing indicators and adhering
to contrasting coding conventions has generated considerable confusion
over even rudimentary measurements such as conflict frequencies, battle-
field intensities and mortality rates. The study of international warfare is,
for instance, a field largely made up of discrete topics. As Dina Zinnes has
commented, everybody ‘loves their variable; nobody [i]s willing to stand up
and say, “I give up my variable” ’ (quoted in Brecher et al. 2000: 37). At the
level of civil war, similar problems exist. In reviewing the current literature,
Nicholas Sambanis points to nine major categories of concern: ‘poverty
and slow economic growth’, ‘ethnic diversity and polarization’, ‘natural
resources’, ‘ethnic diversity’, ‘geographical dispersion, rough terrain, and
security dilemmas’, ‘democracy: level and change’, ‘ethnic vs. revolution-
ary wars’, ‘external intervention in civil wars’ and ‘post-war peacebuilding
and war recurrence’ (2002). The presence of such an ‘enormous diversity of
theoretical, methodological, and epistemological perspectives’ on the study
of conflict, violence and peace acutely ‘complicates the task of providing a
concise assessment of the field’ (Levy 1998: 140). This is a major problem
for, as Muñoz notes, ‘one of the greatest obstacles’ in converting these per-
spectives into a ‘commonly recognised theoretical field [is the absence of a]
. . . system of organising and articulating the information at our disposal on
the subject’ (2005: 2). Clearly,