8 Ethnicity as a generator of conflict

Stuart J. Kaufman

Ethnic identities have existed throughout recorded history. Even in ancient times,
ethnic groups such as the Hebrews, Babylonians and Egyptians were important political
actors (Smith 1986), just as contemporary Serbs and Kurds are. These different groups
typically have interests or goals that are competing in some way, and these differences
often lead to political or social conflicts. Most of these conflicts involve little or no
violence, instead being expressed through religious expression, economic competition,
social segregation, competition among ethnically based political parties, or other
peaceful means. Still, especially when the issue at stake is the political dominance of one
group over another, violent ethnic clashes do sometimes occur, leading sometimes to
riots, and in the worst cases to civil wars, mass expulsions of populations, and even
genocide.
Experts disagree about the extent to which ethnicity causes or generates conflict. One
group, the “instrumentalist” school of thought, sees ethnic identity as little more than a
tool used by elites to pursue competition over tangible goods such economic opportunity.
From this perspective, there is no such thing as “ethnic conflict” at all, and ethnicity
does not cause or generate conflict; it merely provides a framework or a label in which
other sorts of competition occur. Other scholars, in the “psychocultural” school of
thought (Ross 2007), argue that ethnic conflict is very real, and that conflicts – over the
status of the holy sites in Jerusalem, for example – often stem directly from the way
people define their ethnic identities, and are not primarily about the participants’ desire
for material goods. Thus ethnicity can be – though it does not necessarily have to be – a
generator of conflict.

What is ethnic conflict?

Discussions of ethnicity and ethnic conflict are notoriously imprecise, because people
Copyright © 2010. Routledge. All rights reserved.

disagree about what counts as an ethnic conflict. Are race relations between blacks and
whites in the United States an example of low-violence ethnic conflict, or is racial
conflict a different category altogether? If race is different, does the distinction extend
to Rwanda, where Hutus and Tutsis – both black – referred to their difference as one of
race? Are relations between Muslims and Hindus in India, or between Sunni and Shi’a
Arabs in Iraq, cases of ethnic conflict, or do they belong in different categories as
“religious,” “communal,” or “sectarian” conflicts?
For an anthropologist, what these cases all have in common is that the groups
involved are primarily ascriptive – that is, membership in the groups is typically assigned
at birth and is difficult to change. In theory, Indian Muslims can convert and become

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 92 Stuart J. Kaufman
Hindu, and Iraqi Sunnis can become Shi’a, but in practice few do, and the conversion
of those few is not always accepted by their new co-ethnics. Identities of this kind, then,
are “sticky,” hard to change even if they are not marked by the kind of obvious physical
differences that distinguish African-Americans from white Americans. Based on this
commonality, I will use the broader definition of ethnicity that encompasses all of these
kinds of ascriptive groups. According to Anthony Smith (1986), a group is an ethnic
group if its members share the following traits: a common name, a believed common
descent, elements of a shared culture (most often language or religion), common
historical memories, and attachment to a particular territory.
In the past, experts disagreed widely about where ethnicity comes from. Some,
focusing on the evidence that many ethnic identities seem to go back hundreds or
thousands of years, asserted that ethnicity was a “primordial” identity, and implied
that it was essentially unchangeable. They emphasized that groups often worked hard
to make their identity unchangeable, sometimes carving that identity onto their bodies
through tattoos or circumcision (Isaacs 1975). Even when they do not go that far,
however, people tend to stick to the identities – especially the language and religion –
they learn first from their parents. This view of ethnicity implies that ethnic conflict is
based on “ancient hatreds” that are impossible to eradicate and nearly impossible to
manage.
There is another, more complicated side to ethnic identity, however. Most people
have multiple identities that are either “nested” (as subgroups within larger groups) or
overlapping. The average Cuban-American is at the same time also an American
Hispanic or Latino, an American Catholic, an American, and a member of the
worldwide Catholic Church. Which identity is more important to her is likely to depend
on the situation: when listening to the Pope, she is likely to respond as a Catholic; when
watching the U.S. President, as an American; and when thinking about U.S. policy
toward Cuba, as a Cuban-American.
Furthermore, identities do sometimes change, with new ones emerging and old ones
disappearing, especially in times of crisis. For example, when the Soviet Union was
breaking apart in the early 1990s, Ukrainians and Russians in the Transnistria region
of Moldova came together as “russophones” – people who preferred to speak Russian
rather than Moldovan – to resist the assertiveness of the ethnic Moldovans (Kaufman
2001). On the other hand, the “Yugoslav” identity disappeared when the country of
Yugoslavia died in 1991, so people who formerly called themselves Yugoslavs had to
shift to another identity as Serbs, Croats, or members of some other group.
Noticing that people shift their identity – or at least the identity they use politically
– based on the situation, a second group of scholars emerged to argue that ethnic
identity is not “primordial” at all, but merely “instrumental” (Hardin 1995). From this
Copyright © 2010. Routledge. All rights reserved.

perspective, people follow “ethnic” leaders when it is in their interests to do so, and
leaders try to create ethnic solidarity when it works for them. This view of ethnic
identity implies that ethnic conflict can be blamed primarily on selfish leaders who
mislead their followers in pursuit of their own power. The conflicts themselves, these
scholars argue, are typically not really “ethnic” at all – in many cases, clashes are
motivated by economic or criminal disputes, but are later reinterpreted as having been
ethnically motivated for political purposes (Brass 1997).
A third point of view about ethnic identity mixes the other two views by emphasizing
the degree to which people create their identities. Expressed in book titles such as The
Invention of Tradition (Ranger 1992), this view points out that ethnic identities are

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 Ethnicity as a generator of conflict 93
“socially constructed.” They are not “natural” in the sense that a simple primordialist
view would assume; even racial distinctions are just a matter of custom. For example,
most African Americans accept the label “black,” but in South Africa, most of them
would be classified as “colored” – of mixed race – rather than as the darker, purely
African “blacks.” Most Americans would not notice the difference, but in Apartheid-
era South Africa the difference would have shaped every aspect of people’s lives.
Furthermore, constructivists pointed out, the source of these customs was “invented
traditions”: writers or scholars who created what Anthony Smith calls a “myth-symbol
complex.” This myth-symbol complex establishes the “accepted” history of the group
and the criteria for distinguishing who is a member; identifies heroes and enemies; and
glorifies symbols of the group’s identity. In most cases, these mythologies “mythicize”
real history, taking real events but redefining them as the morally defining experiences
of their people. In many cases, these events are what Vamik Volkan (1997) has called
“chosen traumas,” such as the Holocaust for Jews or the 1389 battle of Kosovo Field
for Serbs. In some cases, however, histories and myths are invented from whole cloth to
create new identities.
These constructivist insights can be viewed as a way to settle the argument between
primordialists and instrumentalists, because constructivist ideas explain both the
insights and the problems of the other two views. For example, most Serbs honestly
believe that their identity is primordial, forged in the fires of battle against the Turks at
Kosovo in 1389, so their perception is that their conflicts with Muslims are the result of
primordial “ancient hatreds.” In fact, though, that view of history was the result of late
nineteenth-century Serbian politics and educational policy (Snyder 2000); before then,
most Serbs did not think of themselves as Serbs at all. Similarly, Serbian politicians like
Slobodan Milosevic did indeed use Serbian ethnic identity instrumentally to pursue
their own power in the 1990s, but that identity “worked” politically only because it had
been socially constructed before. Any old identity will not do.
Another question is how to tell whether a particular conflict is an ethnic conflict.
Most African countries are multiethnic, for example, but African civil wars often
involve warlords competing for control over resources such as diamond mines, so
ethnicity has little to do with who is on which side. A conflict is ethnic only if the sides
involved are distinguished primarily on the basis of ethnicity. Often one or both sides
in an ethnic conflict will be a coalition of ethnic groups, rather than a single one, but the
conflict is still ethnic because the people involved choose sides on the basis of their
ethnic group membership, rather than other considerations such as economic interests.

An overview of ethnic conflicts

Copyright © 2010. Routledge. All rights reserved.

Ethnic groups and ethnic conflicts are everywhere. One comprehensive survey found a
total of 275 ethnic or communal groups in 116 countries around the world that were
socially disadvantaged in some way – “minorities at risk.” Put together, the groups
included more than 1 billion people, or about 17.4 per cent of the world’s population
(Gurr 2000, pp. 9–10). Of the fifty biggest countries in the world by population, only
four – Poland, Tanzania, Nepal, and North Korea – did not have at least one “minority
at risk” (and Tanzania has many ethnic groups: they were merely judged not to be “at
risk”). Some of these groups are very small, in mostly homogeneous countries:
Australia’s lone “minority at risk,” the Aborigines, are only about 1 per cent of the
country’s population; while Japan’s only minority, the Koreans, are only one-half of

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 94 Stuart J. Kaufman
one per cent. Some of the groups are very large and important, however: Malaysia’s
Chinese minority is 27 per cent of the population, and India’s oft mistreated Muslims
are 11 per cent of India’s population. Overall, it is accurate to say that most countries
in the world are ethnically diverse, and ethnic relations yield some degree of conflict in
most of them.
Most of the time, the existence of minority groups does not lead to violence or even
to serious conflict. In 1995, most of the “minorities at risk” (58 per cent) were either
politically inactive or mobilized only for routine politics. Another 15 per cent were a bit
more volatile, engaging in demonstrations, rioting, or both. Still, violent ethnic conflicts
were unfortunately plentiful: forty-nine (18 per cent) of ethnic groups were engaged in
“small-scale rebellion” in 1995, and another twenty-two (8 per cent) were fighting a
“large-scale rebellion” (Gurr 2000, p. 28). These numbers, however, were just about the
worst ever: the long-term trend is that the number of violent ethnic conflicts increased
fairly steadily from the end of World War II until the mid-1990s, but then it started to
drop. A separate survey for 2003 lists only ten “intermediate armed conflicts” or “wars”
that were more or less ethnic conflicts. Those conflicts were: the Karen insurgency in
Burma, Hutu-Tutsi conflict in Burundi, the Kashmir insurgency in India, Palestinian
resistance against Israel, the Muslim rebellion in the southern Philippines, the Chechnya
conflict in Russia, the Tamil separatist conflict in Sri Lanka, two separate wars in Sudan
(one against southern Christians, another in Darfur), and the Kurdish insurgency in
Turkey (Wallensteen 2004).
What are these violent conflicts about? The simplest answer is political power in a
disputed territory. Most of the conflicts involve a regional minority that wants to
separate and form its own state, or at least its own autonomous region. The conflicts in
Burma (Karens), India (Kashmir), Palestine (versus Israel), Philippines (Muslims),
Russia (Chechnya), Sri Lanka (Tamils), and Turkey (Kurds) are more or less of this
type. In other cases, the insurgent ethnic group wants to take over government of the
whole country: thus Burundi’s majority Hutus wish to take power from the minority
Tutsi government. Often the goals and stakes are unclear, as rebels may disagree with
each other. For example, some Palestinians want to establish their own state alongside
Israel but others are fighting to replace Israel with a Palestinian state.
The role of ethnicity itself in generating these conflicts – both the violent and the
non-violent ones – remains the subject of dispute. The remainder of this chapter
explores these issues.

Ethnicity as generator of non-violent conflict

It is misleading to say that ethnicity itself is the cause of any conflict, violent or not. It
Copyright © 2010. Routledge. All rights reserved.

is never true that two individuals or groups come into conflict merely because A. has
one ethnic identity and B. has another. Ethnically defined street gangs, for example,
may claim that they attack individuals of other groups merely because of ethnic
difference, but this is not true. Most often, gang members attack because they are
“defending their turf” – because they are gang members fighting turf wars, not merely
because they are members of different ethnic groups.
That said, ethnic identities can generate conflict by associating different groups with
different interests. Thus many ethnic groups are distinguished from each other because
their native languages are different, so they tend to disagree over language policy. If one
group’s mother tongue is the official language of their country, members of that group

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 Ethnicity as a generator of conflict 95
will probably find it easier than nonnative speakers to get certain benefits – for example,
they are more likely to do well on university entrance exams or civil service tests. In one
longstanding example of such a dispute, members of Sri Lanka’s Sinhalese majority
have long championed a “Sinhala only” language policy that disadvantages the Tamil-
speaking minority. Similarly, a group that includes substantial numbers of relatively
recent immigrants (such as American Latinos) is likely to benefit from liberal
immigration policies while native groups may feel disadvantaged by such policies. One
way of thinking about this process (Hale 2008) suggests that the role of ethnic identity
is to reduce people’s uncertainty by clarifying who is the “us” whose interests (in this
case, language interests) will be pursued.
When ethnic groups are distinguished by religion or sect, several other types of
problems can occur. Those who deeply believe that theirs is the one true faith are likely
to desire laws that restrict the practice or spread of rival faiths, discriminating against
the adherents of those rival faiths. They may restrict the availability of ritually banned
foods, offending those who wish to eat those foods. They may give their faith official
government status, devaluing the status of believers in other religions. They may also
push for religiously motivated laws offensive to practitioners of other faiths. Finally,
there may be conflict on all of these issues within religious groups between hard-liners
who wish to entrench their faith in law and moderates who are more concerned with
accommodating minority groups and their own less pious coreligionists. Such issues are
especially common in Muslim-majority countries, many of which designate Islam as the
official religion (Fox 2007), but there are exceptions. In Muslim-majority Uzbekistan,
the government discriminates against the Muslim faithful, associating religious piety
with support for the terrorist Islamic Movement of Uzbekistan. Russia under Putin, on
the other hand, sparked disputes by banning certain religious organizations and
restricting others’ growth while supporting the Russian Orthodox church.
Most common of all is probably conflict between ethnic groups over economic goods.
Sometimes these conflicts are interpreted as being “merely” economic disputes and not
ethnic at all, but this view is misleading. In ethnically diverse societies, economic issues
are almost always at the same time ethnic issues. People tend to have more contacts
with people within their ethnic group than outside it, and people tend to use the resulting
intra-ethnic networks of personal ties when making economic decisions – whom to buy
goods from, whom to hire, where to go to look for work, and so on. In the United
States, the resulting dynamics may create “institutional racism” – the tendency for
white employers to favor white job applicants not because of animosity toward minority
groups, but merely because their personal contacts are mostly white. Another way
economic and ethnic interests may align is when particular ethnic groups come to
specialize in particular lines of work, resulting in the emergence of “middleman
Copyright © 2010. Routledge. All rights reserved.

minorities” who are prominent in retail trade in some areas (Horowitz 1985). Finally,
when ethnic groups are concentrated in particular regions, economic competition
between regions comes to be defined in ethnic terms.
When the group cleavage involves racial difference, it almost always also involves a
history of racial discrimination, inevitably yielding tensions and a wide range of
approaches to dealing with them. Rwanda, after the 1994 genocide, has tried to ban any
official consideration or even mention of the formerly central Hutu–Tutsi divide as a
way of managing that dispute – and of obscuring the fact that most government leaders
after the 1990–94 civil war have been members of the Tutsi minority. South Africa
focused on a transition to political democracy as the group that had been most

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 96 Stuart J. Kaufman
discriminated against, the blacks, formed the majority of the population. The United
States, to overcome the legacy of its centuries of racial discrimination, created a policy
of “affirmative action” giving special benefits to members of previously repressed
groups, especially African-Americans – and sparking continuing controversy and
resistance by some who are disfavored by those policies.
What makes these different kinds of groups – and group conflicts – similar is that the
ethnicity comes to define people’s identity, generating conflict over issues that go
beyond the specific cleavages that separate the groups. In Northern Ireland, for
example, the main line of cleavage is between Catholics and Protestants, but the issues
are not religious per se. Rather, the issue is one of national loyalty – Catholics
(“nationalists”) wish Northern Ireland to become part of the Republic of Ireland, while
Protestants (“unionists”) wish to maintain its union with Great Britain. Even after the
1998 Good Friday Agreement settled the violent conflict between these groups, tensions
continue over issues like the right of Protestant “Orange Order” groups to conduct
marches through Catholic neighborhoods (see Ross 2007).
The example of Orange Order parades illustrates another key fact about ethnic
conflict: sometimes the issues at stake are not tangible interests at all, but purely
symbolic ones. In 1950s Warri, Nigeria, ethnic Itsekiri and Urhobo clashed violently
over the issue of whether the traditional Itsekiri ruler should be given a title implying he
“was paramount ruler over the entire Province” (Horowitz 1985). Other symbolic
ethnic disputes involve the rules governing the wearing of Islamic head scarves in
France and Turkey, and rules governing archeological digs in Jerusalem (Ross 2007).
In Bendery, Moldova, in 1990, clashes were sparked by ethnic Moldovan efforts to
raise their flag in that ethnically Russian, and soon to be separatist, city (Kaufman
2001, p. 141).
Why are symbolic issues like these often contested so fiercely? Psychocultural
theorists point out that in psychological experiments, people randomly assigned to
groups tend to evaluate their own group more highly than other groups, even when
they are told they are not competing; and they tend to prefer to maximize the difference
between their group’s profits and those of another group even if there is an alternative
that would give their group a bigger profit. Donald Horowitz (1985) – in explaining the
riot over the Itsekiri chief’s title and other similar events – argues that these findings
explain ethnic conflict well: what is at stake is not just absolute benefits but group
self-esteem, or, in his terms, group worth and legitimacy.
Symbolic politics theory (Kaufman 2001) provides a similar explanation of such
events. Symbolic politics theory begins with the fact, noted above, that an ethnic
identity is defined by a “myth-symbol complex” that sets out not only who is in the
group, but also who the group’s heroes and villains are, what its history is, and what it
Copyright © 2010. Routledge. All rights reserved.

means to be a group member. From this perspective, Orange Order parades are so
contentious in Northern Ireland because they are meant to commemorate William of
Orange’s Protestant victory at the 1690 battle of the Boyne – which is, of course, seen
as a great defeat in the Catholic myth-symbol complex, and so its commemoration in
Catholic neighborhoods is seen as an insult. The insult, of course, is part of the point of
the exercise, as again the issues are group worth and legitimacy: the right to march
through Catholic neighborhoods was for many decades symbolic of Protestant political
dominance, and higher social status, in Northern Ireland.
The tendency of political conflicts to line up with ethnic divisions often causes
political parties in ethnically diverse countries to become associated with particular

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 Ethnicity as a generator of conflict 97
ethnic groups. Belgium, for example, began with the typical European range of
ideologically based parties such as Christian Democrats and social democrats, but all
of these parties later split on linguistic lines between French-speakers and Flemish-
speakers. The same process occurred in many ethnically diverse Caribbean countries as
well (Horowitz 1985). Malaysia exemplifies a different model, in which the parties were
from the beginning ethnically-oriented, with the United National Malay Organization
and the Malaysian Chinese Association being for many years two of the leading parties.
Nigeria at its birth followed the Malaysian pattern, but later changes in Nigerian
election law required presidential candidates to gain political support across ethnic and
regional lines, leading to the rise of more ethnically diverse parties.

Ethnicity as a generator of violent conflict

As noted above, ethnic conflicts are usually managed peacefully. Sometimes, however,
ethnic diversity does lead to ethnic violence. In the statistics about ethnic conflicts
quoted above, the violent conflicts fell into two broad categories: riots, and armed
conflicts or civil wars.

Ethnic riots
Deadly ethnic riots have occurred all over the world, but how and why they occur
seems puzzling. Particularly puzzling is why rioters tend to be very careful in attacking
only members of the target ethnic group, while at the same time making no distinction
between men, women and children of that group, and indulging in unspeakable brutality
in how they are killed, with rape, torture and mutilation not uncommon. After the
killing is done, there is usually no remorse on the part of the killers: “they had it coming”
is the attitude typically expressed by rioting communities all over the globe (Horowitz
2000).
One comprehensive survey, which takes a social psychological approach, finds three
main factors that lead to deadly ethnic riots (Horowitz 2000). First, there needs to be a
hostile ongoing relationship between the groups – tensions of long standing to motivate
the killing. Second, there needs to be authoritative social support: potential rioters need
to be assured by public statements from community leaders in their group that the
leaders agree killing members of the other group is justified. At the same time, this
support usually extends to the security forces: riots usually become large only if the
police are sympathetic, or at least do not make determined efforts to stop the killing.
Finally, there needs to be a stimulus, some event – usually implying some sort of
threat – that provokes fear, rage, or hatred in the rioting group. For example, a report
Copyright © 2010. Routledge. All rights reserved.

(true or not) of a violent attack by one of “them” against one of “us” might spark a
widespread cry to “teach them a lesson.” Alternatively, a political change – even a
potential one – might provoke a similar outburst. In 1958, for example, Sri Lankan
Prime Minister S. W. R. D. Bandaranaike, a Sinhalese, signed a power-sharing deal
with the leader of his country’s Tamil minority, but quickly backed away under political
pressure. After the deal was abrogated, ordinary Sinhalese vented their wrath at the
very idea of such power-sharing by attacking innocent Tamils in a large-scale riot.
Another approach to explaining ethnic riots focuses not on psychology but on social
organization. In India, for example, hostile relations between the Hindu and Muslim
communities are common, but most of the riot violence is concentrated in just a few

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 98 Stuart J. Kaufman
cities. Why is that? The riot-prone cities, in turns out, have “institutionalized riot
systems”: community activists and extremist organizations that benefit from keeping
tensions high, politicians who benefit from occasional violence, and criminals and thugs
who can profit from it (Brass 1997). On the other hand, Indian cities with little or no
riot violence have community organizations (business groups, labor unions, etc.) that
cross communal lines, bringing Hindus and Muslims together instead of driving them
apart (Varshney 2002).

Ethnic civil wars

Explanations of ethnic civil wars divide along similar lines: social psychology
approaches, social mobilization approaches, and instrumentalist approaches.
Instrumentalist approaches start with what creates the opportunity for rebels to act:
weak governments, large populations and inaccessible terrain create the opening
extremists need to act (Fearon and Laitin 2003). Also important in most instrumentalist
arguments are extremist leaders seeking to grab or hold on to power, who stir up ethnic
disagreements and provoke violence to create a “rally around the flag” effect uniting
their group around their own leadership (Gagnon 1995). Extremist media also plays a
key role in this view, as these extreme outlets seek popularity by appealing to group
loyalties, presenting the news in terms of “us” against “them” (Snyder 2000). These two
factors work together: extremist leaders provide heroes for the extremist media to
promote, while one-sided media portrayals seem to validate the extremist leaders’
claims that their group must unite against the “enemy.” In most of these accounts,
security fears play a crucial role: the argument by extremist politicians and media
outlets that one’s own group is in danger is what makes their appeals seem credible.
Some instrumentalists go a step further and claim that civil wars involving these
issues are not ethnic at all, but merely about political power or economic benefits. These
scholars argue that the statistical link between ethnic diversity and civil war is weak,
and that the main causes of civil war are poverty, weak governments, and other factors
that make it easy to start a guerrilla campaign (e.g., Fearon and Laitin 2003). It is also
true, however, that while economic grievances are always present, in ethnic conflicts
they are expressed in ethnic terms. In Mindanao in the southern Philippines, for
example, the poor – Christians as well as Muslims – are all disadvantaged by inadequate
government spending on education and infrastructure. But the communist New
People’s Army, which tries to exploit such rich–poor distinctions to gain support, has
had little luck in Muslim areas. Rather, Muslims there respond to specifically Muslim
rebel groups who emphasize the differences between Muslims and Christians, not
between rich and poor (McKenna 1998). In other cases, it is not the poor ethnic group
Copyright © 2010. Routledge. All rights reserved.

but the rich one that rebels: in Yugoslavia, for example, it was the relatively prosperous
Slovenes and Croats who first tried to secede, because they felt they were being held
back by the more “backward” ethnic groups in the rest of the country. In these cases,
the contest for power and wealth takes a peculiarly ethnic form.
Like instrumentalist approaches, social mobilization approaches consider the roles
of leaders, but they are also interested in how ethnic groups mobilize – that is, how do
members of the group get together the people and resources needed for collective
action? The answer, these theorists point out, is that people use social organizations
and networks that already exist, like political parties and labor unions. Successful
mobilization efforts find “brokers,” people who can link different groups and networks

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 Ethnicity as a generator of conflict 99
together to help them cooperate in a single movement (McAdam et al. 2001). This
provides one answer to the question: why do people mobilize as ethnic groups instead,
for example, of organizing as economic interest groups? It is because people’s social
networks tend to be mostly within their ethnic group; barriers of language or religion
typically separate them from members of other groups.
Social psychological approaches focus on a different puzzle: why do followers follow
these extremist leaders? Even if people mobilize as ethnic groups to look out for their
interests, why do they follow extremist leaders who want violence, instead of following
moderate leaders who will work for peace? Symbolic politics theory suggests that when
the group’s myth-symbol complex points to the other group as an enemy, its members
will be predisposed to be hostile to the other group. Politicians will then be able to
appeal to symbols of past hostility – such as Slobodan Milosevic referring to the battle
of Kosovo Field – to rouse people’s emotions against the enemy that symbol brings to
mind (Muslims, in the case of Kosovo). If the group is at the same time convinced that
they are in danger of extinction – of being wiped out as a group – they can be persuaded
to back extreme measures that are justified as “self-defense” (Kaufman 2001).
One point on which the different approaches agree is that even when groups are
differentiated by religion, violent conflicts are rarely religious in the sense of one group
trying to impose its religion on another. For example, even though Sri Lanka’s Tamils
are Hindu while the majority Sinhalese are Buddhist, neither group wants the other to
convert. Rather, the rebel Liberation Tigers of Tamil Eelam want to establish their own
state (Tamil Eelam) in northern and eastern Sri Lanka, while the Sinhalese-dominated
government wants to prevent that outcome. The Kashmir, Chechen, Palestinian, and
Philippine Muslim conflicts have a similar flavor. The biggest exception is Sudan, where
the main grievance of the Christian and animist southerners was the attempt by the
Sudanese government to impose Islamic law on the whole country, including them.
To see how these complex processes play out in practice, let us consider in more
detail the example of one prominent case of ethnic warfare, the 1990s fighting in the
former Yugoslavia. Yugoslavia, formed in the aftermath of World War I, was a multi-
ethnic state with no majority group. The three largest groups all spoke the same
language, Serbo-Croatian, but differed in their religious tradition among Serbs
(Orthodox Christians), Croats (Catholics), and Bosnian Muslims. The fourth-largest
group, the Slovenes, are Catholics but speak a different (though related) language; the
next largest, the Albanians, are Muslims who speak a wholly unrelated language.
Before World War II, Yugoslavia was ruled by a Serbian king and dominated by
Serbian politicians. During World War II, the Germans conquered the country and
placed Croatian fascists, the Ustashe, in power in the regions of Croatia and Bosnia
and Herzegovina, where they engaged in genocidal violence against the Serbs. As the
Copyright © 2010. Routledge. All rights reserved.

war ended, communist partisan leader Josip Broz Tito took power in Yugoslavia,
massacring the Ustashe and re-creating Yugoslavia as a nominal federation of six
republics: Serbia, Croatia, Bosnia and Herzegovina, Slovenia, Macedonia, and
Montenegro (Kaufman 2001).
When Tito died in 1980, the loss of his charismatic authority severely weakened
Yugoslavia’s government. The increasingly powerful republic governments more and
more allowed the kind of mutually hostile mythmaking Tito had tried to stamp out.
For example, nationalist Serbs began talking about the menace of the Albanian
minority in the symbolically important Kosovo region while labeling any Croatian
disagreement as evidence of resurgent Ustashe fascism. As symbolists would note,

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 100 Stuart J. Kaufman
ethnic myths and fears were growing. The leader of Serbia’s League of Communists,
Slobodan Milosevic, noticed the power of this nationalist sentiment and in the late
1980s became its spokesman, repressing the Albanians and attempting to impose
Serbian control on the whole of Yugoslavia (Gagnon 1995). In response to this Serbian
threat, voters in Slovenia, Croatia and Bosnia and Herzegovina turned to supporting
their own nationalist leaders – with the Croatian nationalists reviving the national
symbols last used by the Ustashe fascists, raising alarm among Serbs and making
Milosevic’s appeals ever more plausible.
Yugoslavia was dying. Slovenia moved first, declaring independence on June 25,
1991. The Croats quickly followed, sparking a month-long war in which the Yugoslav
army conquered areas in Croatia inhabited by Croatia’s Serbian minority.
The agony of Bosnia and Herzegovina was to be longer. Home to a mixture of
Bosnian Muslims, Serbs, and Croats, Bosnia and Herzegovina was torn three ways.
Serbs wanted to remain in Yugoslavia but, fearing Serbian domination, the Muslims
wanted to secede and form an independent Bosnian state, while Croats wanted their
areas (especially western Hercegovina) to join Croatia. In 1992 a coalition of Muslims
and Croats therefore declared Bosnian independence, sparking a three-sided civil war
in which Serbia and Croatia – trying to take over chunks of Bosnian territory – provided
military assistance to their co-ethnics in Bosnia and Herzegovina, while the Muslims
were the principal victims. Under the slogan “Only unity saves the Serbs,” Serbs
exaggerated the disadvantages of separation from Serbia into a threat to their national
existence, and used this invented threat to justify – and invent the term – “ethnic
cleansing”: the Serbs’ program of massacring enough of their ethnic enemies to force
the rest to flee any territory they claimed. Finally, in 1995, a Croatian military
counteroffensive backed by NATO air power prompted the Serb side to agree to stop
the fighting.

Conclusion
Ethnicity generates conflict in a number of different ways. When ethnic groups are
differentiated by language, then disputes about the use of language, especially in
government and education, tend to line up across ethnic divides. When ethnic groups
are differentiated by religion, disputes over the role of religion and the influence of
religious values on public policy tend to arise. Regardless of what differentiates groups,
economic interests – and disputes – tend to pit ethnic groups against each other due to
the importance of social networks in causing members of ethnic groups to favor their
own economically. In addition to disputes over tangible interests, ethnic politics also
often turns into contests for status or group worth, so groups may seek political
Copyright © 2010. Routledge. All rights reserved.

dominance as a way of expressing their desire for high social status. When this sort of
seeking for group dominance becomes especially pronounced, and especially when
groups’ myth-symbol complex encourage hostility toward other groups, peaceful ethnic
disputes can escalate into violent conflict.

Further reading
Brass, Paul R. (1997), Theft of an Idol: Text and Context in the Representation of Collective
Violence (Princeton NJ: Princeton University Press).

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 Ethnicity as a generator of conflict 101
Horowitz, Donald L. (1985), Ethnic Groups in Conflict (Berkeley CA: University of California
Press).
Kaufman, Stuart J. (2001), Modern Hatreds: The Symbolic Politics of Ethnic War (Ithaca NY:
Cornell University Press).
Laitin, David D. (2007), Nations, States and Violence (Oxford: Oxford University Press).
Ross, Marc Howard (2007), Cultural Contestation in Ethnic Conflict (Cambridge: Cambridge
University Press).

References
Brass, Paul R. (1997), Theft of an Idol: Text and Context in the Representation of Collective
Violence (Princeton NJ: Princeton University Press).
Cohen, Lenard J. (1995), Broken Bonds: Yugoslavia’s Disintegration and Balkan Politics in
Transition (Boulder CO: Westview Press).
Fearon, James D., and David D. Laitin (1996), “Explaining Interethnic Cooperation,” American
Political Science Review 90, No. 4 (December), pp. 715–35.
—(2003), “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97, No. 1
(February), pp. 75–90.
Gagnon, V. P. (1995), “Ethnic Nationalism and International Conflict: The Case of Serbia,”
International Security 19, No. 3 (winter), pp. 130–66.
Gurr, Ted Robert (2000), Peoples versus States (Washington DC: United States Institute of
Peace).
Hardin, Russell (1995), One for All: The Logic of Group Conflict (Princeton NJ: Princeton
University Press).
Hobsbawm, Eric, and Terence Ranger (1992), eds, The Invention of Tradition (Cambridge:
Cambridge University Press).
Horowitz, Donald L. (1985), Ethnic Groups in Conflict (Berkeley CA: University of California
Press).
—(2000), The Deadly Ethnic Riot (Berkeley CA: University of California Press).
Isaacs, Harold R. (1975), The Idols of the Tribe: Group Identity and Political Change (New York:
Harper & Row).
Kaufman, Stuart J. (2001), Modern Hatreds: The Symbolic Politics of Ethnic War (Ithaca NY:
Cornell University Press).
—(2006), “Symbolic Politics or Rational Choice? Testing Theories of Extreme Ethnic Violence,”
International Security 30, No. 4 (spring), pp. 45–86.
Kaufmann, Chaim (1996), “Possible and Impossible Solutions to Ethnic Civil Wars,” International
Security, 20, No. 4 (spring), pp. 136–75.
Lijphart, Arend (1985), Power Sharing in South Africa, Policy Papers in International Affairs 24
(Berkeley CA: Institute of International Studies, University of California).
McAdam, Doug, Sidney Tarrow, and Charles Tilly (2001), Dynamics of Contention (Cambridge:
Cambridge University Press).
Copyright © 2010. Routledge. All rights reserved.

McKenna, Thomas M. (1998), Muslim Rulers and Rebels: Everyday Politics and Armed Separatism
in the Southern Philippines (Berkeley CA: University of California).
Rieff, David (1996), Slaughterhouse: Bosnia and the Failure of the West (New York: Simon &
Schuster).
Ross, Marc Howard (2007). Cultural Contestation in Ethnic Conflict (Cambridge: Cambridge
University Press).
Saideman, Stephen M. (2001), The Ties that Divide: Ethnic Politics, Foreign Policy and
International Conflict (New York: Columbia University Press).
Smith, Anthony D. (1986), The Ethnic Origins of Nations (New York: Blackwell).
Snyder, Jack (2000), From Voting to Violence: Democratization and Nationalist Conflict (New
York: Norton).

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.
 102 Stuart J. Kaufman
Varshney, Ashutosh (2002), Ethnic Conflict and Civic Life: Hindus and Muslims in India (New
Haven CT: Yale University Press).
Volkan, Vamik (1997), Bloodlines: From Ethnic Pride to Ethnic Terrorism (New York: Farrar
Straus & Giroux).
Wallensteen, Peter (2002). Understanding Conflict Resolution: War, Peace and the Global System
(London: Sage).
Wallensteen, P., and M. Sollenberg (1999), “Armed Conflict, 1989–1998,” Journal of Peace
Research 36, No. 5, pp. 593–606.
Zartman, William I. (1985), Ripe for Resolution: Conflict and Intervention in Africa (New York:
Oxford University Press).
Copyright © 2010. Routledge. All rights reserved.

Routledge Handbook of Ethnic Conflict, edited by Karl Cordell, and Stefan Wolff, Routledge, 2010. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/londonschoolecons/detail.action?docID=957760.
Created from londonschoolecons on 2023-12-27 22:15:08.