Repression, Dissent, and the Onset of Civil War
| DOI | 10.1177/1065912912452485 |
| Date | 01 September 2013 |
| Published date | 01 September 2013 |
| Author | Joseph K. Young |
| Subject Matter | Articles |
5Political Research Quarterly XX(X)Young
Regular Article
Political Research Quarterly
66(3) 516 –532
Repression, Dissent, and the
© 2012 University of Utah
Reprints and permissions:
sagepub.com/journalsPermissions.nav
Onset of Civil War
DOI: 10.1177/1065912912452485
prq.sagepub.com
Joseph K. Young1
Abstract
The prevailing scholarly wisdom is that weak states, or resource-poor states, are the most prone to civil war. Yet many
weak states never experience civil war. Why then are some weak states prone to civil war while others are not? The
author offers a theory that explains how dissidents and states interact to jointly produce civil war. In sum, states that
repress their citizens are the most likely to kill citizens and to generate dissident violence. This insight resolves an
academic puzzle and when tested provides a model with better predictive ability than previous models.
Keywords
civil war onset, repression, dissent
Recent scholarship has established that the condition of
Since its inception in 1971, the king and other leaders of
being a weak state raises the probability of civil war
the state have exercised uncontested control throughout
(Fearon and Laitin 2003).1 Weakness is generally mea-
the territory. Even though it has an even lower level of
sured as lack of resources or low GDP.2 UN statistics
GDP and also has rough, mountainous terrain, Bhutan has
show that of the fifty countries in the world with the low-
avoided the civil war that El Salvador has experienced.
est GDP, nearly 60 percent experienced civil strife of
These are not isolated cases. For example, civil war
varying intensity and duration in the 1990s. While this is
has occurred in Rwanda, Cambodia, Angola, Afghanistan,
a high percentage, the experience of almost 40 percent of
Sierra Leone, and many other countries. What do these
these resource-poor states remains unexplained.3 These
countries all have in common? All of the above states are
statistics, coupled with the academic findings, are the
commonly termed weak states. Yet other weak states,
foundation for an interesting puzzle: why do some weak
such as Bhutan, Cameroon, Ecuador and Burkina Faso,
states experience civil wars whereas many others do not?
have avoided civil wars. Why?
Consider the experiences of El Salvador and Bhutan.
Although El Salvador and Bhutan are matched on
Each state is commonly considered weak, each has diffi-
some important civil-war-enhancing characteristics, such
cult terrain, neither was a democracy until recently, yet
as population and difficult terrain, an important differ-
one has seen a bloody, protracted civil war while the other
ence between the two is the Salvadoran military’s choice
is relatively tranquil.4 El Salvador is a weak state with a
to repress the population. I argue that states that lack
per capita national income that has hovered around $2,000
resources and societal support are at the most risk for the
since the 1970s. In the mid-1970s, the military leaders of
onset of civil war. State leaders make choices, such as
El Salvador responded to increased political mobilization
repression, that reduce support from society and increase
by dissidents with harsh repression. In 1975, for example,
the likelihood of active dissent. Similar to previous schol-
protesters against the Miss Universe Pageant were gunned
arship, I argue that states with low capacity are at the
down, killing fifteen (Wood 2003). The process began in
most risk of civil war. Contrary to previous work,
the early 1970s as dissident groups openly criticized and
however, I conceptualize state capacity as involving more
organized against the repressive military rule, and by 1979
than just resources; it also involves societal support.
a state of open civil war existed (Mason and Krane 1989).
The civil war lasted over a decade, claiming well over
1American University, Washington, DC, USA
fifty thousand lives (Lacina and Gleditsch 2005).
Corresponding Author:
Like El Salvador, Bhutan is considered a weak state. It
Joseph K. Young, School of Public Affairs, American University, 4400
is a landlocked country with a per capita GDP that fluctu-
Massachusetts Ave NW, Washington, DC 20016, USA.
ated between $400 and $1,000 between 1988 and 2005.5
Email: jyoung@american.edu
Young
517
state, I describe how the actions and resources of the state
affect dissident choices. I then integrate the state and dis-
sidents and show how their interaction affects the likeli-
hood of producing or avoiding civil war. After outlining
the theory and deriving hypotheses, I discuss research
design. Previous work in the quantitative study of civil war
has relied on single equation regressions with a set of con-
trol variables. Instead, I estimate a series of equations using
a two-stage maximum likelihood procedure and offer a
way to empirically evaluate the hypotheses while incorpo-
rating the endogenous relationships among some variables.
This procedure more accurately models the relationship
among the variables of interest and closely ties the theo-
retical expectations to the statistical estimation. After dis-
cussing the sample and variables used in the study, I
Figure 1. Repression levels prior to civil war.
evaluate the results of the estimations. In the conclusion, I
offer some extensions of the theory to other types of politi-
cal violence and discuss some prospects for future research.
To illustrate this process, Figure 1 examines repres-
sion prior to civil war onset across a large sample of
Why Civil War?
countries.6 Figure 1 shows average repression levels in
countries prior to experiencing civil war versus average
Scholars have generally explained the causes of civil war
levels of countries that do not have this type of conflict.
using a grievance or opportunity approach.7 In the griev-
As the figure shows, average levels of repression are
ance story, there are certain background conditions or
always higher in countries that will have civil war onsets,
actions by the state that push members of society into
and this process builds about five years prior to the civil
active violent opposition. Havard et al. (2001), for exam-
war. Between years 6 and 5, for example, repression
ple, find that ethnic heterogeneity and inconsistent
increases 8.8 percent. All of the years following show
regimes, or regimes that have elements of openness and
similar increases except between years 4 and 3, which
repression, increase the probability that civil war will
show a slight decrease (2 percent). Overall, the trend is
occur. In contrast, Collier and Hoeffler (2001) suggest
consistently positive, representing a 26 percent increase
that opportunities for rebellion increase the likelihood
in the level of repression from year 6 to year 1.
that civil war will develop. For adherents to this approach,
Extant work has focused primarily on state weakness
where rebels can extort or extract resources, their oppor-
and failed to incorporate strategic interaction between
tunity costs for violent rebellion are low.
the state and dissidents. In contrast, I offer a model that
Another opportunity-based approach that has gener-
explicitly identifies the process of violence between states
ated much debate suggests that civil war is most likely
and dissidents that links weak states to civil war. This
when the conditions that favor insurgency are present in
model clarifies the micro-foundational motives of the cen-
a particular state (Fearon and Laitin 2003). These condi-
tral actors, the state and the dissidents. State repression
tions affect the relative strength of the insurgency and
both leads to rising violence levels (illustrated in Figure 1)
include such characteristics as a financially weak state,
and pushes civilians into active dissent. Once dissidents
rough terrain, large populations, and the newness of a
are mobilized and challenging the state, civil war becomes
state.8 In short, Fearon and Laitin (2003) conclude that
more likely. The answer to the puzzle? High levels of
structural conditions such as mountains and large popu-
repression occur in some resource-poor states; these states
lations increase the strength of an insurgency, which
are most likely to experience civil war. In sum, I argue that
increases the probability of civil war. Financial weakness,
the likelihood of a state experiencing civil war is a func-
as proxied by GDP, is critical as it affects the state’s ability
tion of state–dissident interaction; this is an important
to perform counterinsurgency, use discriminate violence,
claim that, though noncontroversial on its face, has been
and provide local policing. For Fearon and Laitin, finan-
ignored in the literature.
cially weak states, or resource-poor states, are the most
I begin with a discussion of the prevailing wisdom
prone to civil war as they are unable to perform the tasks
regarding civil war: civil war is most likely to occur within
that reduce the effectiveness of potential insurgent adver-
states that have ripe conditions for insurgency. I then dis-
saries. Fearon and Laitin conceptualize state weakness as
cuss the preferences of state actors and offer an explana-
one-dimensional, meaning that the resources available to
tion for why some states use violence. After outlining the
the state determine its capacity. Fearon and Laitin argue
518
Political Research Quarterly 66(3)
that...
To continue reading
Request your trial