Paradise is a Bazaar? Greed, Creed, and Governance in Civil War, 1989-99
| Author | INDRA DE SOYSA |
| Published date | 01 July 2002 |
| DOI | 10.1177/0022343302039004002 |
| Date | 01 July 2002 |
| Subject Matter | Articles |
395
Introduction
The collapse of the Cold War has intensi-
fied the search for appropriate theoretical
models for explaining a new environment of
civil violence. Some popular explanations
see creeping ecological and demographic
pressure as the causal mechanism behind
violent conflicts. Apparently increasing
environmental scarcity and Malthusian
© 2002 Journal of Peace Research,
vol. 39, no. 4, 2002, pp. 395–416
Sage Publications (London, Thousand Oaks,
CA and New Delhi)
[0022-3433(200207)39:4; 395–416; 025813]
Paradise Is a Bazaar? Greed, Creed, and
Governance in Civil War, 1989–99*
INDRA DE SOYSA
Department of Political & Cultural Change, Center for Development
Research (ZEF), University of Bonn
Some prominent recent studies of civil war argue that greed, not grievance, is the primary motivating
factor behind violence, basing their conclusions on a strong empirical association between primary com-
modity exports and civil war. This study contrasts alternative propositions that see need-, creed-, and
governance-based explanations that are intimately related to the question of primary commodity depen-
dence and conflict. Maximum likelihood analysis on approximately 138 countries over the entire post-
Cold War period shows little support for neo-Malthusian claims. Abundant mineral wealth makes
countries highly unstable, whereas scarcity of renewable resources is largely unrelated to civil conflict.
A positive effect of population density on conflict does not seem to be conditioned by renewable
resource scarcity. Ethnicity is related to conflict when society is moderately homogenous; a highly plural
society faces less risk. Very slight political liberalization leads to conflict, but larger increases reduce the
danger considerably, supporting the view that conflict is driven by opportunistic behaviour rather than
by grievance. Increases in homogeneity among Islamic and Catholic populations make them riskier.
Perhaps institutional factors relating to separation of church and state rather than competing creeds
explain culture conflicts. Larger shares of both Christians and Muslims within countries make them
safer, contrary to claims of natural antagonism between the two. Governance, proxied by the ratio of
total trade to GDP, predicts peace strongly, an under-theorized area within the study of civil war. Trade’s
relationship to peace is robust to specification and sample size, supporting the findings of the State
Failure Project. Greater attention should perhaps be paid to formal and informal institutional factors
that create the synergy between private and public spaces for overcoming collective action problems of
maintaining peace.
* I thank John Oneal, Erich Weede, Maximo Torrero, Nils
Petter Gleditsch, Mansoob Murshed, and three anonymous
referees for helpful comments and suggestions. I am grate-
ful for comments received at the World Bank/Princeton
University workshop (March 2000), GECHS workshop at
UC Irvine (March 2000), UNU/WIDER in Helsinki
(October 2000), and from colleagues at ZEF. The generous
support of the Research Council of Norway for the Oslo
Project Office of the Global Environmental Change and
Human Security (GECHS) program is gratefully acknow-
ledged. Kerstin Henke and Maria Lensu provided invalu-
able research assistance. The data are posted at http:
//www.prio.no/jpr/datasets.asp. I am the sole miscreant of
all errors.
02desoysa (ds) 20/6/02 8:35 am Page 395
at SAGE Publications on December 7, 2012jpr.sagepub.comDownloaded from
pressures have created a new age of insecur-
ity that drives people to fight for survival
(Homer-Dixon, 1999; Renner, 1996).
Others argue that conflict is caused by
‘greed’ rather than ‘grievance’, and that a
relative abundance of natural resources sup-
plies the motivation for organizing violence.
Resources offer lootable income over which
to fight, making costly strategies of violence
viable – a few can ‘do well out of war’
(Collier, 2000; de Soysa, 2000). Some focus
on societal grievances based on ethnic and
civilizational divides. Cultural difference
and ascription create natural battle lines
that result in violence over primordial
hatred. Others view culture as being instru-
mentalized by ethnic entrepreneurs to gain
access to political and economic resources
(Gurr & Harff, 1994). This study will
systematically assess the relative contri-
butions of these general propositions, uti-
lizing new models of civil conflict generated
by research at the World Bank (Collier &
Hoeffler, 1998). This study introduces
governance and social capital arguments
into the models on sound theoretical and
methodological grounds to test the robust-
ness of the new explanations.
Despite an explosion of studies on the
causes of civil conflict, there are few system-
atic tests of competing propositions that
utilize objective criteria in quasi-
experimental ways. Naturally, a complex of
factors are involved in the generation of con-
flict, but it is still rare that studies systemati-
cally gauge the relative weight of competing
propositions. This study will gauge the
importance of proxies that capture some
aspect of theoretical linkages in an inte-
grated model. The results may then speak to
contrasting propositions and the larger
debate as to whether conflict is a function of
selfless opposition to tyranny (the search for
justice), or whether tyranny derives from the
vanities of a few men (the search for private
gain), a problem that has occupied social
research since at least the genesis of the
science of politics.1
The study of the causes of conflict is
fraught with many difficulties. One problem
facing systematic examination of the causes
of violent conflict is identifying the appro-
priate model and variables to hold constant
when exploring the impact of variables of
interest. All too often, research is pursued
based on the dependent variable, which is
that conflict already exists to arouse the inter-
est of research, leading to misleading con-
clusions about causes because the lack of
variance in case-study based research design
leads to overdetermination of some variables
over others (Collier & Mahoney, 1996; King,
Keohane & Verba, 1994). Moreover, scholars
are likely also to confuse what they want to
explain (underlying causes of violent con-
flict) by conflating contests within the
political arena, which may create a dominant
discourse, with acts of violence. The
violence, however, might in fact be quite
independent of the contests that drive the
dominant discourse (Varshney, 2001).
The discourse dominating zones of con-
flict is heavily laced with stories of grievance,
and objective factors are likely to be well
masked, particularly to academics and jour-
nalists. In these situations, one can find just
about any narrative of grievance to provide
the basis of the causal story. The focus on
difference, which is a natural function of the
‘enemy-image’, is an especial feature of
political contests and violence. In such situ-
ations, cultural differences in particular may
come to be overemphasized in games of ‘us
versus them’, even though the participants in
disputes rarely speak with one voice. The dis-
course of grievance is often unrelated to
objective truth, which makes the discourse
journal of PEACE RESEARCH volume 39 / number 4 / july 2002
396
1Political scientists and sociologists debate whether rebel-
lion is motivated by relative deprivation, or whether by self-
serving, rational action (see Weede, 1998 for a review), a
debate that relates intimately to the greed versus grievance
juxtaposition.
02desoysa (ds) 20/6/02 8:35 am Page 396
at SAGE Publications on December 7, 2012jpr.sagepub.comDownloaded from
To continue reading
Request your trial