War, Rivalry, and State Building in Latin America
Author(s): Cameron G. Thies
Source: American Journal of Political Science, Vol. 49, No. 3 (Jul., 2005), pp. 451-465
Published by: Midwest Political Science Association
Stable URL: http://www.jstor.org/stable/3647725
Accessed: 11-09-2017 12:50 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Midwest Political Science Association is collaborating with JSTOR to digitize, preserve and
extend access to American Journal of Political Science
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
War, Rivalry, and State Building in Latin America
Cameron G. Thies Louisiana State University
Scholars of Latin America have recently begun to apply the bellicist approach to state building to the region, the central claim
of which is that wars are a great stimulus to centralizing state power and building institutional capacity. This article argues
that current applications of these models of state building are too narrowly specified to be of much use in Latin America or
elsewhere in the developing world. Replacing the focus on interstate war with the more general phenomenon of interstate
rivalry, alongside the consideration of intrastate rivals, allows us to account for the impact of both external and internal
forces on the development of the state. I demonstrate the utility of this approach through several cross-sectional time-series
analyses that provide evidence that external and internal rivals affect the Latin American state in a manner consistent with
the general nature of bellicist theory.
cholars have recently begun to apply the bellicist
approach to state building to Latin America. The
central claim of this approach is that wars are a
great stimulus to centralizing state power and building
institutional capacity (Centeno 1997, 2002; L6pez-Alves
2000, 2001). However, the manner in which the approach
has been applied may be too literal. Since Latin American
state building occurred in a different historical context
than the early modern European experience upon
which bellicist or predatory theory is based, the literal
application has often been used as a foil to argue for
competing explanations for the relatively weak states that
populate the region today. I will argue that current applications
of Tilly's (1975, 1985, 1992) model of state building
are too narrowly specified to be of much use in Latin
America or elsewhere in the developing world. When
properly specified, these models are applicable to Latin
America.
I begin by reviewing the qualitative findings on the
relationship between war and state building in Latin
America. The scant scholarship on this area generally finds
that the bellicist approach developed by scholars like Tilly
is a useful grand theory "mirror" to hold up to the reality
of Latin America (Centeno and L6pez-Alves 2001). The
literature finds that war was not a stimulus to state building
in the region, but rather than completely rejecting the
bellicist approach, scholars have explained why war did
not operate in Latin America as it did in early modern
Europe. The comparative case study method or narrative
analysis is employed to show how war failed due to a
number of factors peculiar to Latin America as a region
and particular to states within that region. The common
argument is that internal violence overwhelmed the Latin
American state in the absence of countervailing pressure
that would typically be supplied by the fear of external violence.
Further, the preoccupation with the "enemy within"
also prevented states from engaging in external violence.
As a result, Latin American states are relatively weak entities
that exist in a precarious position relative to both
domestic society and other states in a region characterized
by a violent peace.
However, when the almost exclusive focus on external
war is replaced with the more general phenomenon
of interstate rivalry, and intrastate rivals are included simultaneously
in the model, we can account for both the
impact of external and internal forces on the development
of the state. Interstate rivals, whether operationalized
as "enduring" or "strategic" rivals, have a positive
effect on the state's extractive capacity. Intrastate rivals,
on the other hand, have the expected negative effect on
state building. It is also clear that a variety of other important
factors, such as the frequent recourse to external
debt have had significant effects on the state's ability to
extract from society. The conclusion is that a more generalized
version of the bellicist approach to state building
is applicable to Latin America. The same types of basic
Cameron G. Thies is assistant professor of political science, Louisiana State University, 240 Stubbs Hall, Baton Rouge, LA 70803-5433
(cthiesl@lsu.edu).
I would like to thank Moises Arce, Paul Hensel, Doug Lemke, Andrew Schlewitz, and the anonymous reviewers and editors of AJPS for
their helpful comments.
American Journal of Political Science, Vol. 49, No. 3, July 2005, Pp. 451-465
?2005 by the Midwest Political Science Association ISSN 0092-5853
451
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
452 CAMERON G. THIES
forces operate in the reg
during the period of early
tion. However, the results
not exactly the same, as La
not attained the capacity an
counterparts.
Bellicist Approac
Building in Lati
Latin Americanists are not the first to look at the impact
of interstate war and a threatening security environment
on the modern developing state. Cohen, Brown, and
Organski (1981), Kirby and Ward (1991), Jaggers (1992),
Stubbs (1999), and Bates (2001) all highlight the striking
similarities that contemporary developing states share
with their early modern European counterparts, including
the impact of war and external threats on state building
efforts. Others argue that the lack of external threats
and war have led to relatively weak states (Desch 1996;
Herbst 1990, 1996/97, 2000; Lustick 1997), but the logic
is the same-strong external threats, including but not
limited to war, produce states with stronger institutional
capacities to extract from society.
The logic of the bellicist approach is most clearly articulated
by Tilly (1985, 182) who maintains that rulers
engage in four activities in the process of building the
state. The first activity is war making through which rivals
located outside of the territory they are attempting
to control are neutralized or eliminated. The second activity
is state making, which involves the elimination or
pacification of potential rivals to their rule from within
their territorial base. The third activity is the protection
of those actors that support their rule. Finally, in order
to accomplish the other activities, the state must engage
in the extraction of resources from the population they
are attempting to control. Tilly (1985, 180) argued that
of these activities, war making was the main stimulus to
increases in the level of taxation and debt, hence it has
become the main focus for scholars attempting to apply
this model outside of Europe.
As Centeno notes, "on the most basic level, the functions
of a state include the provision and administration
of public goods and the control of both internal and external
violence" (2002, 2), which seems to place his work
on the Latin American state squarely within the bellicist
tradition. However, as he later makes clear, he does not
view the state as an independent actor, which severs the
theoretical tie to Tilly's work and others in the predatory
theory tradition (e.g., Levy 1981; North 1981). According
to Centeno,
states are not actors in and of themselves. They
are shells-potentially powerful shells-but nevertheless
hollow at the core. The machine of the
state needs a "driver" able to use the stimulus provided
by war to expand its reach and power. Without
such a driver, whether it be state personnel,
a dominant class, or even a charismatic individual,
the political and military shell of the state has
no direction. Without this direction, wars do not
present opportunities for growth, but are mere
challenges to survival. (2002, 166)
Centeno believes that wars only make states when a politically
or militarily dominant institution and a social class
are united in the view that war is the best way to maintain
their position and privilege. Centeno (2002, 142) argues
that geographical, social, and racial divisions in Latin
America prevented the emergence of a unified, centralized
state able to take advantage of the stimulus provided by
war. While Latin American elites shared a common fear of
the "enemy below"-nonwhite subalterns, they were divided
among themselves as liberals versus conservatives,
and criollos versus peninsulares, among other cleavages.
Ultimately, these divisions meant that the state "was never
able to impose the internal unity required for the extraction
process, even in the face of military threats" (Centeno
2002, 138-39). The existence of "rival claimants" to authority
among elites also diminished the state's ability to
maintain a monopoly over the use of violence, resulting
in an "almost predatory quality" of life in many countries
because the state was unable to provide protection to its
citizens (Centeno 2002, 7-8).
Centeno (2002, 46) argues that internal conflict continues
to reflect both the absence of international competitors
and the lack of centralization of power domestically.
He suggests that it is possible that higher levels
of external conflict might have served to mitigate internal
divisions, yet there were just too few wars in Latin
America that occurred at the wrong time to have this
type of impact. Centeno (2002, 9) supports this position
by demonstrating that Latin America has participated in
relatively few interstate wars since the early nineteenth
century compared to the rest of the world. Further, the
borders of Latin American countries have remained relatively
stable, and no state has disappeared due to conquest.
Finally, while civil wars were sometimes defined territorially,
they typically revolved around competing claims to
the state (Centeno 2002, 128). The combination of high
levels of internal violence and low levels of external violence
produced a "violent form of peace" (Centeno 2002,
35). This finding is consistent with previous work on regional
order in Latin America that characterizes it as a
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY AND STATE BUILDING 453
"violent peace" (Mares 2001) or "zone of n
(Kacowicz 1998).
The problem with the bellicist model, a
Centeno, is that Latin American states have
ited" as opposed to "total" wars (2002, 2
wars are short in duration, encompass a sm
area, occur between states over frontier or
sues, are fought by professional as opposed
soldiers, and tend to have little impact on th
izen. The net result is that these wars do n
large-scale mobilization of society, nor do t
usual effect on government's ability to ext
Total war, on the other hand, enhances the
to extract resources, enables the state to cen
shifts loyalties and emotional attachments to
represents the nation, and transforms the
a "citizen." Latin America's experience with
has produced states built on "blood and deb
to "blood and iron."
Andreski (1980) offers a complementary analysis of
the relationship between the external and internal uses of
military force. In fact, he argues there is a fundamental incompatibility
between internal and external uses of force,
such that increased frequency of the internal use of force
diminishes the capability of the military to wage war externally.
The frequent deployment of the military to dispense
with rival claimants to centralized rule in Latin America
may have limited the military's ability to properly prosecute
interstate wars. This is because time, equipment,
and organization directed toward the pursuit of internal
rivals detract from the pursuit of external rivals. More importantly,
the coercive use of the military within the state
detracts from national solidarity and public readiness to
support the military and the state more generally through
either symbolic or material means. Andreski's analysis
also points toward the deleterious effect that simultaneous
deployment of forces both internally and externally
must have on the state. The logical result is equivalent to
Centeno's argument that internal rivals overwhelmed the
state and prevented it from augmenting its extractive capacity
to more properly prosecute external wars.
L6pez-Alves (2000, 2001) similarly argues that holding
Latin America up to the "mirror" ofTilly's (1992) work
provides a useful lens through which to view the region's
experience, though it does not capture a complete understanding.
In his judgment, Latin America (like Europe)
falls under Tilly's capitalized coercion model, in which
the state used both coercion and capital to centralize their
control.
On both shores, states spent money in reconstructing
the economy, buying the loyalty of local
lords and army officers, establishing bureaucracies
(including a system of taxation), and trying to
secure the support of the economic elite. Part of
the state budget also went to placate local revolts,
and, on occasion, smaller sums were distributed
among the populace. (L6pez-Alves 2001, 160)
The difference is that Latin American states did not
rely on a central army to subdue the upper classes or
enforce taxation, instead relying on foreign loans and
customs duties for revenue and suffering the ill effects
of internal political rivals. Rouquie (1987) offers a similar
argument yet he rejects the application of the capitalized
coercion model to Latin America because the
region's economies were so externally oriented and penetrated.
L6pez-Alves (2001, 164-70) also stresses the effect
of guerrilla warfare in delaying the consolidation of the
state, and generally diminishing the state's extractive capacity
as the government often used tax exemptions to
placate local and regional rebellions. Overall, Centeno's
conclusion "that while war may have played a significant
role in the development of some European states, its explanatory
power wanes on crossing the Atlantic" (2002,
20) seems to represent the dominant view of the region.
Current scholarship thus starts with the bellicist approach
inspired by Tilly's work on Europe, draws some
lessons from it, then moves on to explanations grounded
in regional experience. However, the approach adopted in
this article is to begin with Tilly's work, but enhance its
generalizability outside of the European experience. After
all, states are states, no matter what time period or geographical
region they inhabit. States clearly vary in their
institutional capacity, but that does not mean that they
are incomparable, simply that some underlying factors
are likely responsible for that variation. In our search for
a general approach to state building, we can begin with
Tilly's contention that the state must engage in war making
to deal with external rivals and state making to deal
with internal rivals.
As Centeno notes, "it is not necessarily war itself, but
the threat of war that often produces the positive statebuilding
consequences" (2002, 266). This language suggests
that external threats in the form of interstate rivals
may produce a similar effect on the state as actual war. Interstate
rivalries contain varying levels of threat, and the
threat is of extended duration, which may act to similarly
empower the state to extract from society. But how do the
political processes of war and rivalry affect the general
state-building process?
Historical institutionalism has previously afforded
analytical leverage in understanding the effects of these
types of political processes on state building (e.g., Ertman
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
454 CAMERON G. THIES
1997; Waldner 1999) an
the Latin American con
nature of state institu
Mahoney 2001). Histori
stitutional arrangemen
are disrupted by some
2001 la). Stability is the
that privilege certain a
military elites, who wo
and benefits. Change i
vested interests of such
stemming from adopti
ments. In fact, change
quent shocks to the ins
In the study of state-b
thought to disrupt the
surrounding the state's
society. Wars produce
effect" that moves the
a new and higher equ
pansion of the state's f
(Campbell 1993; Peaco
Thompson 1985; Russ
However, if Centeno's
limited wars fought in
produce the familiar ra
rupting the institution
effect was evident onl
levels of public debt. T
suggests that when do
ditions make it difficu
enues, states will turn
to prosecute and pay f
sion that "war, state ap
advanced in tight cad
Europe is incomplete
according to Centeno's
produced increased bo
effect on taxation an
state fiscal apparatus.
Interstate rivalries, on
more of a "slow-mov
Historical institutional
lengthy, soc
nature. Wh
legacies may
consequence
an increme
Skocpol (200
types of pro
ticed by sc
processes th
duce increm
time before
change (Bau
This is one o
point for in
versial (Ben
that produc
makers are
Thies (2001
produced, m
Chilean riva
ing process
elites in both
alism that to
Argentine st
spite the fac
of this proc
by the popu
taught in th
planning, a
racy (if at a
war, the mil
strumental i
among the p
"protection
justify incr
the two cou
in Latin Am
tally adjust
periods of ti
1987). In ad
protection r
terstate riva
inability to
wartime.
External rivals should not be considered without considering
internal rivals as well, since both are important
to Tilly's model and predatory theory in general (Thies
2004). While Centeno and L6pez-Alves suggest that internal
rivals overwhelmed the Latin American state, both
suggest that these rivals were able to do so in the context
of a lack of external rivals. Therefore, Centeno and L6pezAlves's
work does fit within a generalized version ofTilly's
1A variety of explanations for the ratchet effect have been proposed
in this literature (see Campbell 1993 for a review). First, the state
may be able to increase the tax burden in a war context because citizens
prefer greater protection to foreign domination. Second, the
threat of disinvestment by owners of capital in the face of higher tax
rates may also diminish during wartime, as owners of capital benefit
directly and indirectly from the infusion of increased government
spending into the economy. Third, rulers may simply increase the
tax burden during wartime to ensure their survival, regardless of
the impact on citizens or investors.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY AND STATE BUILDING 455
model. We should therefore simultaneously
fects of both external and internal rivals on
capacity of the state. External rivals should
to augment its extractive capacity, while in
should have the opposite effect in Latin Am
Data and Method
The current scholarship on the connection between war
and state building in Latin America focuses primarily
on South America. Centeno (2002) examines Argentina,
Bolivia, Brazil, Chile, Colombia, Ecuador, Paraguay, Peru,
Uruguay, Venezuela, and Mexico, which this article also
selects as cases for the sake of comparability.2 However, I
conduct a large-N, statistical analysis, as opposed to the
qualitative (and sometimes small-N) analyses of Centeno
and L6pez-Alvez. Centeno states that "there is little question
that we may speak of a probabilistic connection between
war and state development" (2002, 19), seemingly
suggesting that statistical analysis of this relationship is
appropriate. Yet, Centeno (2002, 116; footnote 37) also
states that he studiously avoided statistical analysis because
the data would provide deceptive cross-national
comparisons and coding particular years for war would be
too subjective. In fact, Centeno cautions against grafting
"the kind of modeling borrowed from mainstream social
science onto comparative historical studies. Finding a single
path through historical evidence is nearly impossible"
(2002, 278). Historical institutionalists are rather eclectic
in this regard, with some scholars creating general models
(Collier and Collier 1991; Mahoney 2001) and others
emphasizing the unintended or unanticipated aspects of
history (Pierson 1996). However, all scholars would agree
that macrosocial structures and processes are complicated
phenomenon to understand and explain, hence we should
exercise appropriate caution when interpreting any type
of evidence for our theoretical propositions. This article
suggests that qualitative and quantitative approaches can
provide complementary evidence on the state building
process in Latin America.
The economic data used here comes from the
Oxford Latin American Economic History Database
housed at the Latin American Centre at Oxford University,
which is an updated version of Thorp (1998).3 It is
the most comprehensive source of economic data on Latin
America available for the entire twentieth century (1900-
2000). Unfortunately, comprehensive, reliable data for the
nineteenth century, which would have allowed for a more
comprehensive study of state building, is not available. We
should bear in mind when interpreting the evidence that
these states are further along in their consolidation in the
period under study than Tilly's early modern European
states. However, given that state building is an ongoing
process, the period under study is still likely to shed light
on the factors that affect state building efforts.
The dependent variable of interest is the tax ratio,
which is the state's tax revenue as a percentage of gross
domestic product (GDP). The tax ratio is the standard
measure of the state's extractive capacity (e.g., Campbell
1993; Centeno 2002; Fauvelle-Aymar 1999). Taxes and the
bureaucracies that support regularized extraction are the
stuff out of which the modern state is built. Organski and
Kugler argue "taxes are exact indicators of governmental
presence" (1980, 74). The inability to extract tax revenue
from society is a key indication of the state's incapacity
to obtain and maintain national unity, legitimacy, and
control.
I first test Centeno's arguments about the effects of
interstate war and civil war on state building efforts. Although
he makes it clear that he thinks the coding of wars is
too subjective to provide meaningful cross-national comparisons,
Centeno (2002, 44-46) actually generates his
own lists of interstate wars and civil wars, which I use to
test his argument. Centeno's classifications of interstate
war and civil war are similar, though not identical, to
the Correlates of War classifications (see Tables 1 and 2).
Given the general expectation that a "ratchet effect" on
tax revenues is the result of a political shock like war, I
allow interstate and civil wars to have a 10-year impact on
tax revenues following convention in the literature (e.g.,
Goertz and Diehl 1995). Coding interstate wars solely for
the wartime years is likely to miss the impact this type of
shock would have on state institutional arrangements surrounding
extraction. Following Andreski's (1980) logic, I
also include a variable to test for the interaction of interstate
and civil wars. If Andreski is correct, internal rivals
2Centeno (2002, 1; footnote 1) excludes Central America for several
reasons: it is geographically separate from the other countries
under study; only so many countries can be studied in a particular
scholarly endeavor; and these countries may represent an exception
to his arguments. However, these reasons may simply serve to highlight
selection bias in the choice of cases. For example, Mexico is not
geographically united with the rest of these South American countries,
yet it is included. Certainly, as the number of cases increases,
the tractability of a comparative case study approach declines, yet
this suggests that a large-N, statistical analysis may be warranted.
Finally, excluding cases because they may not conform to one's a
priori theoretical expectations is problematic. L6pez-Alvez (2000,
7-14) restricts his analysis to five Latin American states, but for
stronger theoretical and methodological reasons. I restrict the current
analysis to Centeno's 11 cases in order to conduct a fair assessment
of his work; however, future research should examine the
Central American cases.
3The data can be accessed at the following web site: http://oxlad.qeh.
ox.ac.uk/index.php.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
456 CAMERON G. THIES
TABLE 1 Latin Americ
War States Dates Centeno COW
World War I Brazil 1917-1918 Yes No
Chaco War Bolivia, Paraguay 1932-1935 Yes Yes
Leticia Colombia, Peru 1932-1933 Yes No
World War II Brazil 1944-1945 Yes Yes
Border Dispute Ecuador, Peru 1941 Yes No
Korean War Colombia 1951-1953 No Yes
Border Dispute Ecuador, Peru 1981 Yes No
Falklands War Argentina 1982 Yes Yes
Border Dispute Ecuador, Peru 1995 Yes No
Source: Centeno (2002, 44) and Correlates of War Project.
might take advantage of the inability of the state to effectively
deploy military force both internally and externally,
leading to a decline in the tax ratio. Unlike Central America,
there are no cases of third-party state intervention in
South American civil wars in this time period.
TABLE 2 Latin American Civil Wars in the
Twentieth Century
State Dates Centeno COW
Mexico "Indian 1880-1900 Yes No
Campaigns"
Colombia "Thousand 1899-1903 Yes Yes
Days"
Venezuela 1898-1900 Yes No
Venezuela 1901-1903 No Yes
Secession of Panama 1903 Yes No
(Colombia)
Uruguay 1904 Yes Yes
Mexican Revolution 1910-1920 Yes Yes
Ecuador 1911-1912 Yes No
Paraguay 1911-1912 Yes Yes
Ecuador 1922-1925 Yes No
Mexico 1923-1924 No Yes
Mexico "Cristero 1926-1930 Yes Yes
Rebellion"
Brazil 1932 No Yes
Paraguay 1947 Yes Yes
Colombia "La Violencia" 1948-1962 Yes Yes
Bolivia Revolution 1952 Yes Yes
Argentina Anti-Peron 1955 Yes Yes
Chile 1973 No Yes
Peru "El Sendero" 1982-1992 Yes Yes
Colombia 1984-Present Yes Yes
Source: Centeno (2002, 45-46) and Correlates of War Project.
I also examine the impact of int
state building, which may present m
slow-moving effect on state instit
ertz conceptualize enduring rivalry
tween two states in which both use,
ity, military threats and force as
both sides formulate foreign polic
(2000, 4). These types of rivalry invo
tarized activities, including both th
of force deployed to resolve foreign
extended periods of time. Operation
valry is a dyad that experiences at l
time period of at least 20 years. Die
44-45) also classify two other types
proto-rivalry (at least three MIDs in
and isolated rivalry (two or fewer
years).4 The Latin American endurin
for the twentieth century are foun
the other conceptualizations of inte
here.
Thompson's (2001) alternative measure of interstate
rivalry, known as strategic rivalry, is included as a comparison
to the Diehl and Goertz measure of enduring
rivalry. Strategic rivalries are conceived as situations in
which states view each other as "(a) competitors, (b) the
source of actual or latent threats that pose some possibility
of becoming militarized, and (c) enemies" (Thompson
2001, 560). Thompson's operationalization of the concept
also differs greatly from the Diehl and Goertz measure,
since there is no reliance on the MID data to mark the beginning
and ending of a rivalry, nor on absolute numbers
of MIDs required to establish a rivalry's existence. Instead,
4I have updated the Diehl and Goertz (2000) classifications of these
types of rivalry through 2000 based on the newly released MID 3.0
data (Ghosn and Palmer 2003).
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY, AND STATE BUILDING 457
TABLE 3 Latin American External Rivalries in t
Dyad Enduring Proto Strategic Interstate I Interstate II
Argentina-Brazil 1817-1985
Argentina-Chile 1873-1909 1843-1991 1873-1984 1897-1984
Argentina-Chile 1952-1984
Argentina-Germany 1939-1945
Argentina-UK 1976-1983 1965-present
Bolivia-Chile 1836-present 1857-1904 1927-1938
Bolivia-Paraguay 1918-1938 1887-1938 1886-1938
Bolivia-Peru 1825-1932
Brazil-Germany 1942-1945
Brazil-UK 1849-1965
Chile-Peru 1911-1921 1832-1929 1871-1929
Chile-Peru 1976-1977
Colombia-Ecuador 1831-1919
Colombia-Germany 1943-1944
Colombia-Nicaragua 1979-1992
Colombia-Peru 1899-1913 1824-1935 1899-1934
Colombia-Venezuela 1982-2000 1831-present
Ecuador-Peru 1891-1955 1830-1998 1891-present 1911-present
Ecuador-Peru 1977-1998
Ecuador-USA 1952-1981 1972-present
Mexico-Germany 1939-1942
Mexico-USA 1911-1920 1836-1923 1859-1927
Peru-USA 1955-1992 1992-present
Uruguay-Germany 1939-1945
Venezuela-Guyana 1966-1982 1966-present
Venezuela-Trinidad 1996-1999
Source: Diehl and Goertz (2000) and Thompson (2001, 570-73).
Thompson relies on the perceptions of foreign policy
makers in the affected states as reported by historians.
Given the controversy in the literature about when
to date the start of a rivalry, I also employ two other
operationalizations of the rivalry concept that affect the
start of the rivalry as a check on the robustness of my
results. MID-based approached to rivalry, such as Diehl
and Goertz's enduring rivalries, are frequently criticized
because the beginning and ending dates of the rivalry are
only known by the researcher retroactively. However, as
Thompson (1995, 2001) has argued, if decision makers
are making policy choices based on a rivalry, then it is
important that they realize they are participating in a rivalry.
Thompson (2001) developed his aforementioned
measure of strategic rivalry precisely to address this problem.
Bennett pursued another approach to improve the
dating of MID-based approaches to rivalry based on the
issues under contention. Bennett's (1996, 1997a) first operationalization
of rivalry, which Thompson (2001) refers
to as "Interstate I rivalry," requir
engage in at least five MIDs durin
25 years concerning the same issu
Bennett's (1997b, 1998) second ope
valry, "Interstate II rivalry," requir
20-year period with no more than a
disputes. In both operationalizatio
ending dates for the rivalry are not
as in the Diehl and Goertz "endur
despite the same reliance on MIDs.
This article will test the effects of t
state rivalries as measures of signifi
Given that interstate rivalries, rega
tional definition, are by their natur
titions that extend over long period
expect to see increased extractive ac
the state to deal with them. Howeve
isolated rivalries from the Diehl a
may lack the level of sustained seve
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
458 CAMERON G. THIES
the state to augment its ex
Centeno's arguments abou
expect that sporadic confli
significant effects on state
effect.
The control variables are standard fare for predictions
of the tax ratio (e.g., Cheibub 1998; Fauvelle-Aymar 1999;
Webber and Wildavsky 1986). They include a measure
of democracy, which is based on the polity2 score from
Polity IV, which varies from +10 (most democratic) to
-10 (most autocratic). The polity2 score is increased by
10 points and then divided in half so that the composite
score varies between 0 and 10 as in the original variables
on which it is based. Predatory theory based on the
European experience generally predicts a negative relationship
between democracy and the tax ratio, although
some studies have found a positive relationship in the developing
world, and perhaps not surprisingly, previous
empirical studies find mixed evidence for the effect of
democracy on extraction (e.g., Cheibub 1998).
I include a measure of external debt as a percentage of
GDP. This measure includes the total of public, publicly
guaranteed, and private nonguaranteed long-term external
debt, including the use of IMF credit, and short-term
debt, that is owed to nonresidents. Centeno is quite clear
that he believes Latin American states avoided increased
extraction in part because of the readily available foreign
sources of funding, hence his book title Blood and Debt.
The resort to external debt by developing states is often
viewed as one of the reasons that Tilly's model may not
generalize outside of early modern Europe. I include the
measure of external debt to control for this alternative to
increased domestic extraction.
The logarithm of GDP per capita in dollars is used
as a measure of national wealth and economic development.
States with higher levels of wealth and development
should have the ability to extract a larger portion of the
national income, as well as a larger initial pool of resources
to consider for extractive purposes. The logarithm of inflation
measured as the annual percentage change in the
Consumer Price Index (CPI) is typically expected to be
negatively associated with tax revenues. However, Latin
America has had an unfortunately long experience with
high and even hyperinflation. These levels of inflation
were often the result of the state's desire to placate labor,
allowing wages to rise to meet previous price increases,
thus creating a vicious circle whereby inflation begat inflation.
At the same time, states continued to provide subsidies
in transportation and food, in addition to more expensive
ventures launched under the import-substitution
industrialization programs popular between the 1920s
and the 1980s. Thus, the state continued to need enhanced
revenues during periods of rising inflation. Inflation and
extraction (along with other means of securing revenues,
such as debt) are expected to be positively related in this
situation. Trade openness (the total of exports plus imports
divided by GDP) is expected to be positively related
to tax revenues as international trade is often thought to
be relatively easy to tax as it enters or leaves through a
limited number of ports.
I also include a measure of the sectoral composition
of domestic product: agriculture as a percentage of GDP.
This sector of the economy is often included as an indicator
of the transaction costs associated with taxation,
along with per capita GDP, trade openness, and mining as
a percentage of GDP (e.g., Cheibub 1998). This article includes
all of these factors except mining, since the Oxford
data does not include a separate variable for this sector
of the economy. The transaction cost argument is that it
is relatively easier to tax the modern sectors of the economy,
as opposed to the traditional sectors like agriculture.
Therefore, agriculture is expected to be negatively related
to the tax ratio.
A time-trend variable is included to capture the variation
in the tax ratio across time throughout the region.
This variable is also a surrogate for age, since all of these
states emerged from the colonial period at approximately
the same time. Presumably, as states age, they should become
better at extracting from society, although at some
point the relationship reverses in mature democracies as
groups bargain through institutionalized political channels
to reduce their share of the tax burden (Jackman
1993). In the Latin American states, I expect a positive
relationship between time (age) and the tax ratio. I estimate
the following pooled cross-sectional time-series
with panel-corrected standard errors (PCSEs; Beck and
Katz 1995).
Analysis
I test Centeno's basic propositions about the relative impact
of interstate and civil war on state building in Latin
America in the model found in Table 4. I first employ Centeno's
measures of interstate and civil war in the model
on the left-hand side of the table. As Centeno suggests,
interstate war does not have a significant effect on extraction,
even allowing for a 10-year impact on the tax ratio.
The sign on this variable is negative, which does seem
to correspond to Centeno's general understanding that if
anything, the few wars that were fought at the wrong time
(in his view) might have actually undermined the state
building process. Civil wars, on the other hand, have a
significant, negative effect on extraction, which is exactly
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY AND STATE BUILDING 459
TABLE 4 The Effect of Interst
Independent Variable Coefficient PCSE Coefficient PCSE
Centeno Interstate War -.0777 .6624
Centeno Civil War -2.6418*** .5637
Centeno Interstate x Civil -2.4289* 1.2602
COW Interstate War -4.4734*** 1.3889
COW Civil War -2.6360*** .4999
COW Interstate x Civil 6.5303*** 1.5658
Democracy .2124* .0959 .2279** .0911
External Debt -11.2135*** 2.1763 -8.2430*** 2.1610
GDP per capita 1.8593*** .5432 1.6366*** .5284
Inflation .2611*** .0540 .2563*** .0503
Trade Openness 14.4286 8.3202 8.6642 7.7740
Agriculture -21.9131*** 3.4295 -23.7643*** 3.1945
Time-trend .1011*** .0174 .0814*** .0184
Constant -2.5987 3.5218 .4379 3.2446
N 653 653
Adj. R2 .5093 .5131
Wald X 2 556.06*** 614.70***
Note: All significance tests are two-tailed: *p < .05, **p < .01, ***
what Centeno argues. Civil wars waged by a variety of internal
rivals to the state have reduced the state's ability to
unify, centralize, and extract from its society. Consistent
with Andreski's argument, states involved in both internal
and external wars as measured by the interaction term
find their extraction rates hampered significantly as well.
Substituting the COW measures of interstate and civil
war provides an interesting twist in the model on the righthand
side of Table 4. The COW interstate war variable is
negative and significant, as is the civil war measure. The
COW measure suggests that not only was war an important
political shock, but its effects were in the opposite
direction expected by a literal application of predatory
theory derived from the European experience. However,
the negative impact of interstate and civil wars is entirely
consistent with Centeno's argument. The twist is the positive
and significant interaction term. According to the
COW measures, states experiencing both internal and external
conflict were better able to extract from society.
This finding actually fits with predatory or bellicist theory,
which would expect a state under siege from within
and without to find a way to enhance its extractive capacity
from groups willing to support its continued rule.
However, upon closer inspection of the data, one
finds only one case in either model of an interaction between
interstate and civil war, hence our confidence in
the generalizability of this finding is diminished. In the
model using Centeno's measures, Peru experiences two
interstate wars with Ecuador while experiencing a longrunning
civil war with El Sendero Luminoso, which resulted
in a decline in the tax ratio. In the model using the
COW measures, Colombia experiences the Korean War
at the same time as it experiences La Violencia, which resulted
in a positive effect on the tax ratio. While Centeno
and COW both agree on the coding of the civil wars in
these cases, Centeno does not recognize Colombia's participation
in the Korean War and the COW project does
not recognize Peru's 1981 and 1995 conflicts with Ecuador
as wars. Centeno's caution about the subjective nature of
coding wars is dramatically illustrated by these two models,
in which the interactive terms depended solely upon
one case in each instance. Dropping the interactive term
has no significant effect on either model. Overall, the results
conform closely to Centeno's argument that interstate
and civil war worked to depress the state's extractive
capacity in Latin America in contradiction to the literal
application of bellicist or predatory theory.
The control variables have their expected effects in
both models. Democracy has a positive and significant
impact on the tax ratio. States that have attained higher
levels of democracy in the developing world may possess
a greater degree of legitimacy, thus their tax ratios are
somewhat higher than their autocratic neighbors. Higher
external debt ratios reduce the reliance on extraction
as expected by Centeno's argument that Latin America
states were built on "blood and debt." External sources of
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
460 CAMERON G. THIES
funding are a definite alte
American context. Inflat
positive effect on the tax
America's history of infl
associated with the transac
collection also exert their
trade openness variable jus
at the .05 level. However, i
significance test.
The time-trend variable also indicates that Latin
American states have on average slowly augmented their
ability to extract from society at a rate of approximately
one percent of GDP every 10 years. This finding should
make clear that state building is a long process, and one
that is never truly complete. One of the misunderstandings
of those who apply the bellicist approach to the developing
world is the idea that state building is completed
when the state is consolidated. However, the initial, tenuous
consolidation of centralized rule over a territory
and population marks the beginning, not the end of state
building.
The more generalized version of the bellicist approach
is found in the models presented in Table 5. Two separate
models are presented, one representing Diehl and Goertz's
(2000) approach to enduring rivalries on the left-hand
side of the table and the other representing Thompson's
(2001) approach to strategic rivalries on the right-hand
side. In both cases, the exclusive focus on interstate war
is replaced by the notion of an interstate rivalry that may
provide a stimulus to extraction over the long-term. These
types of external rivals are also assessed alongside the internal
rivals represented by Centeno's civil war measure.
Centeno's (2002, 61-66) classification of civil wars is comprised
of regional rebellions, ideological battles, caudillo
wars, race/ethnic wars, and revolutions. These are exactly
the types of political and ethnic rivals that early modern
European rulers had to pacify, neutralize, and bargain
with in order to continue their rule.5
In the first model, enduring and proto-rivalries are
significantly and positively related to the tax ratio, while
isolated rivalries have a significant, negative impact. In the
second model, strategic rivalries exert a significant, positive
influence on extraction as well. In both models, the
interaction terms representing situations in which states
are involved in both internal and external rivalry are all significant
and negative. The coefficients on the interaction
terms are also quite large compared to their component
variables. Whether the state is engaged in an enduring,
proto- or strategic rivalry, all of these long-term processes
have negative consequences for extraction when the state
is simultaneously involved in internal conflict, consistent
with Andreski's (1980) argument. The Centeno measure
of civil war is negative, but loses significance in this model,
as the interaction terms capture most of its variation.6 All
of the control variables are similar in size, sign, and significance
to the original war models found in Table 4. These
findings are interesting for a number of reasons.
Most importantly, the results demonstrate that the
more generalized bellicist model is clearly applicable to
Latin America. Long-term, threatening interstate relationships,
in the form of enduring, proto-, or strategic rivalry
prompt increased extraction on the part of the state.
This finding replicates Thies' (2004) previous work on rivalry's
impact on state building in the developing world
as applied to sub-Saharan Africa, Asia, and the Middle
East and North Africa. However, I expand the geographical
scope of the argument by including Latin America, as
well as the temporal scope by examining the entire twentieth
century. Future research should incorporate Latin
American states into the larger pool of developing states
in order to compare the relative impact of war and rivalry
on state building efforts across the globe.
As the interactive terms from the original war models
presented in Table 4 illustrate, substantive decisions
about coding can produce contradictory statistical effects.
Given the disagreement over the most appropriate way to
date rivalries in MID-based approaches, I have included
Bennett's Interstate I and II rivalries as a check on the
robustness of the findings stemming from the enduring
and strategic rivalry measures. The results presented in
Table 6 are very robust. Both additional measures of interstate
rivalry have positive and significant effects on the
tax ratio. The Centeno civil war measure is negative and
significantly related to the tax ratio. The interaction of
interstate rivalry and civil war is also negative and significant.
The control variables are all similar in size, sign,
and significance to the models presented in Table 5. As a
result, we can be confident that external rivalries exert a
significant effect on the state's extractive capacity, regardless
of the coding decisions made by researchers.
As a final check on the robustness of these results,
I recode the COW-based measures of enduring rivalry
5Thies (2004) employed data on ethnic wars, revolutionary wars,
disruptive regime transitions, and genocides/politicides from Gurr
et al.'s (1997) State Failure Task Force as indicators of internal rivals
to the state. However, the state failure data only covers the post-1950
era. Centeno's (2002) classification of civilwars is a reasonable proxy
for the types of "rival claimants" to power required for this analysis.
6The correlations between the Centeno measure of civil war and
the interaction term for enduring rivalry is .49 and proto-rivalry is
.38. The correlation between the Centeno measure of civil war and
the interaction term for strategic rivalry is .73. A visual inspection
of Tables 1 and 2 confirms that most of the Latin American civil
wars occur in the context of external rivalries.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY, AND STATE BUILDING 461
TABLE 5 The Effect of Enduring and Strategic Rivalries on th
in Latin America
Independent Variable Coefficient PCSE Coefficient PCSE
Enduring Rivalry 2.1979*** .3730
Enduring x Civil -6.3762*** 1.1833
Proto Rivalry 4.5999*** .9234
Proto x Civil -6.3323*** 1.3690
Isolated Rivalry -1.6540** .6793
Strategic Rivalry 1.9223*** .4023
Strategic x Civil -4.1324*** 1.3660
Centeno Civil War -.3336 .4894 -.0014 1.0108
Democracy .1647 .0884 .1789 .0954
External Debt -15.5281*** 2.6244 -13.0496*** 2.3899
GDP per capita 1.5406*** .4651 2.5797*** .6049
Inflation .3900*** .0577 .2649*** .0510
Trade Openness 17.3384* 7.9138 10.8389 8.2823
Agriculture -25.2786*** 3.5496 -19.7460*** 3.4911
Time-trend .0931"** .0157 .1092*** .0172
Constant -.1369 3.3834 -9.1412* 4.1641
N 653 653
Adj. R2 .5729 .5187
Wald X2 515.00*** 531.72***
Note: All significance tests are two-tailed: *p < .05, **p < .01, ***p
TABLE 6 The Effect of Interstate I and II Rivalrie
in Latin America
Independent Variable Coefficient PCSE Coefficient PCSE
Interstate I Rivalry 1.0055** .3628
Interstate I x Civil -4.0041*** 1.1345
Interstate II Rivalry 1.0225** .3464
Interstate II x Civil -4.0657*** 1.1239
Centeno Civil War - 1.6109*** .4798 -1.5413*** .4742
Democracy .1979* .0938 .1872* .0938
External Debt -11.3175*** 2.2054 -11.4733*** 2.2194
GDP per capita 1.7504*** .5399 2.0239*** .5362
Inflation .2744*** .0559 .2703*** .0554
Trade Openness 14.6059 8.3713 16.0999 8.5776
Agriculture -23.8680*** 3.6340 -22.8569*** 3.6416
Time-trend .1040*** .0167 .1094*** .0169
Constant -2.0732 3.5142 -4.4144 3.5578
N 653 653
Adj. R2 .5160 .5159
Wald X 2 567.03*** 587.46***
Note: All significance tests are two-tailed: *p < .05, **p < .01, ***p < .00
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
462 CAMERON G. THIES
TABLE 7 The Effect
Ratio in Latin America
Independent Variable Coefficient PCSE Coefficient PCSE
Enduring Rivalry Severity .1027 .4656
Enduring Rivalry Hostility .2953* .1343
Civil War Severity -.4440*** .0750 -.4358*** .0757
Democracy .2075* .0941 .2186* .0950
External Debt -9.7626*** 1.9723 -9.6792*** 1.9584
GDP per capita 1.8878*** .5542 1.8292*** .5507
Inflation .2056*** .0941 .1995*** .0495
Trade Openness 15.5057 8.0726 16.3445* 8.1799
Agriculture -22.5666*** 3.5836 -23.0434*** 3.6370
Time-trend .1051*** .0167 .1046*** .0163
Constant -3.0911 3.5860 -2.7772 3.5950
N 653 653
Adj. R2 .4946 .4972
Wald x 2 587.54*** 596.23***
Note: All significance tests are two-tailed: *p < .05, **p < .01
and civil war for severity and hostility. All of the aforementioned
measures of external and internal rivalry have
assumed a constant threat level, while we know that the
threat level present in these long-term processes varies.
The severity of enduring rivalries and civil wars is based
on the COW fatality measure, which scales the number
of deaths between 0 and 6.7 Enduring rivalries are also
coded for their hostility level, which ranges from 1 to 5.8
Rather than using the highest level of hostilities undertaken
by either state, which is often used in the analysis of
dyads, each rival state is coded for the highest action they
take individually. This avoids the problem of overstating
the overall hostility level of a dyadic dispute (e.g., Diehl
and Hensel 1994), while also accounting for the hostility
exhibited by each individual state and its potential impact
on extraction.
Taking the severity of a rivalry into consideration,
one finds a positive, though insignificant relationship
to the tax ratio in the model on the left-hand side of
Table 7. Considering a rivalry's hostility level, on the other
hand, one finds a positive and significant relationship to
the tax ratio in the model on the right-hand side. This
set of findings is particularly important to the more general
version of bellicist theory presented here. What they
suggest is that the Latin American state is indeed running
a protection racket. Many displays and uses of force in
Latin America involve no fatalities, yet the state is able
extract more from its society by maintaining a long-term
hostile relationship than it does when that relationship
incurs actual fatalities. The actual severity of the rivalry
is insignificant, while the hostility displayed by the state
is important to extraction. This finding is consistent with
Thies' (2001b) analysis of the Argentine-Chilean rivalry,
in which the militaries exaggerated the threat posed by
their rivals to justify their role, status, and benefits through
enhanced extraction and expenditures. Finally, in both
models, the increased severity of civil wars has the expected
impact of a decreased effect on extraction. Battles
between the central government and internal rivals over
control of the state are not a stimulus to extractive efforts,
particularly when large numbers of citizens perish in the
process.
Conclusion
The evidence provided here both challenges and supports
the bellicist approach to state building in Latin America.
On the one hand, a literal application of Tilly's dictum that
"states make wars and wars make states" is not supported
by the data. Centeno and L6pez-Alves provide qualitative
7The coding is 0 = no fatalities, 1 = 1-25 deaths, 2 = 26-100
deaths, 3 = 101-250 deaths, 4 = 251-500 deaths, 5 = 501-999
deaths, 6 = more than 999 deaths. The precise numbers of fatalities
for all of the MIDs encapsulated within the enduring rivalries are
not available, though the 0-6 coding is assigned. The COW measure
of civil war fatalities is coded using the same scale, although actual
fatality estimates are available.
8Hostility levels are coded as 1 = no militarized action, 2 = threat
to use force, 3 = display of force, 4 = use of force, and 5 = war.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY, AND STATE BUILDING 463
analyses that demonstrate interstate wars did not enhance
the extractive capacity of the state, which is supported by
the quantitative analysis presented in this article. However,
I argue that Tilly's model should be generalized for
an appropriate application to the developing world. Tilly's
model places the state at the center of a balancing act between
encroaching external and internal rivals. When the
narrow focus on war is expanded to consider interstate
rivalry, we begin to see the true impact of external threats
on state-building efforts. External rivals, regardless of the
operational measure employed, have significant, positive
effects on the state's extractive capacity. The evidence provided
here suggests that these long-term, slow-moving
processes represent a protection racket organized by the
rulers of the state to maintain themselves in power. Given
the history of military dictatorship, it seems appropriate
to conclude that the Latin American state was indeed a
predator, albeit one that fed more slowly on society, over
longer periods of time, than its early modern European
counterparts.
Internal rivals, as measured by the variety of rival
claimants seeking power through civil wars, operate exactly
as the qualitative analyses suggest. Internal rivals reduce
the state's extractive capacity, whether through the
damage they cause to the economy's productive capacity,
the bargaining they conduct with the state to reduce
their share of the tax burden, or their temporary occupation
of territory that places productive wealth beyond the
state's grasp. Consistent with Andreski's (1980) argument,
states that simultaneously experienced both internal and
external rivalries suffered a reduction in their extractive
capacity, perhaps as a result of the inefficient or incompatible
use of force both at home and abroad. However, I
demonstrate that internal rivals did not completely overwhelm
the state as Centeno has suggested. External and
internal rivals placed competing pressures on the state's
ability to mobilize society for extractive purposes.
The argument that Latin American states often
turned to external debt in lieu of increased extraction at
home is also supported by the quantitative analysis. Generally,
critics of the application of Tillyesque models to the
developing world point to two differences between early
modern Europe and the present: the lack of interstate war
and the ability to borrow on the world market. I have already
dealt with the former, but it is true that developing
states often take on external debt instead of taxing their
domestic populace. However, they do not borrow to the
complete exclusion of taxing. The results of this article
would amend Centeno's thesis to focus on blood, debt
and taxes.
Centeno also attempts to deal with the problem of
regional order. His conclusion resonates across the literature
as other scholars, such as Mares (2001) and Kacowicz
(1998) recognize the existence of a violent or negative
peace in the region. Their diagnosis is much the samehigh
levels of internal violence combined with low levels
of external violence produce a "violent form of peace."
Kacowicz and Mares both recognize that there is a considerable
amount of interstate conflict in the region, yet
it is generally not allowed to spiral out of control into
full-scale warfare. The conceptual focus on interstate and
intrastate rivalries allows us to consider how these pressures
affect both the strength of the state at home and the
resulting effects for the region. While external rivalries
may prompt increased extraction, they may not do so at
a level commensurate with war. The dramatic impact of a
political shock like war did not translate into increased extraction
in Latin America, yet the longer-term incremental
addition to extractive capacity may have been a rational
adaptation by rulers to their circumstances. The notion
of a protection racket seems to fit these states well. The
states of the region perhaps failed to achieve the capacity
and durability of their European counterparts because
while the internal pressures they face are just as strong
as their predecessors experienced, the external pressures
are not.
Latin American states, much like their newer counterparts
in Africa and Asia, are left negotiating an international
system where the same basic forces are at work
today that were operating in early modern Europe, however,
the results of these competing pressures are not exactly
the same. State building will be marked by continued
attempts to control internal violence, defend against
threats of external violence, and provide public goods
for the citizenry through whatever means are available.
Blood, debt, and taxes are dependably recurrent aspects
of the long and painful process of building the modern
state.
References
Ames, Barry. 1987. Political Survival: Politicians and Public
Policy in Latin America. Berkeley: University of California
Press.
Andreski, Stanislav. 1980. "On the Peaceful Disposition of Military
Dictatorships." Journal of Strategic Studies 3(3):3-10.
Bates, Robert H. 2001. Prosperity and Violence: The Political
Economy of Development. New York: W.W. Norton.
Baumgartner, Frank R., and Bryan D. Jones. 1993. Agendas
and Institutions in American Politics. Chicago: University of
Chicago Press.
Beck, Nathaniel, and Jonathan N. Katz. 1995. "What to Do (And
Not to Do) With Time-Series Cross-Section Data." American
Political Science Review 89(3):634-47.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
464 CAMERON G. THIES
Bennett, D. Scott. 1996. "Secur
terstate Rivalry." Internatio
83.
Bennett, D. Scott. 1997a. "Democracy, Regime Change, and Rivalry
Termination." International Interactions 22(4):369-97.
Bennett, D. Scott. 1997b. "Measuring Rivalry Termination."
Journal of Conflict Resolution 41(2):227-54.
Bennett, D. Scott. 1998. "Integrating and Testing Models of Rivalry."
American Journal of Political Science 42(4):1200-32.
Campbell, John L. 1993. "The State and Fiscal Sociology." Annual
Review of Sociology 19:163-85.
Centeno, Miguel A. 1997. "Blood and Debt: War and Taxation
in Nineteenth-Century Latin America." American Journal of
Sociology 102(6):1565-605.
Centeno, Miguel A. 2002. Blood and Debt: War and the NationState
in Latin America. University Park: The Pennsylvania
State University Press.
Centeno, Miguel A., and Fernando L6pez-Alves, eds. 2001.
The Other Mirror: Grand Theory through the Lens of Latin
America. Princeton: Princeton University Press.
Cheibub, Jose A. 1998. "Political Regimes and the Extractive
Capacity of Governments: Taxation in Democracies and
Dictatorships." World Politics 50(3):349-76.
Cohen, Youssef, Brian R. Brown, and A.F.K. Organski. 1981.
"The Paradoxical Nature of State Making: The Violent Creation
of Order." American Political Science Review 75(4):901-
10.
Collier, Ruth B., and David Collier. 1991. Shaping the Political
Arena: Critical Junctures, the Labor Movement, and Regime
Dynamics in Latin America. Princeton: Princeton University
Press.
Desch, Michael C. 1996. "War and Strong States, Peace and Weak
States?" International Organization 50(2):237-68.
Diehl, Paul E, and Gary Goertz. 2000. War and Peace in International
Rivalry. Ann Arbor: University of Michigan Press.
Ertman, Thomas. 1997. Birth of the Leviathan: Building States
andRegimes in Medieval and EarlyModern Europe. NewYork:
Cambridge University Press.
Fauvelle-Aymar, Christine. 1999. "The Political and Tax Capacity
of Government in Developing Countries." Kyklos
52(3):391-413.
Ghosn, Faten, and Glenn Palmer. 2003. "Codebook for the Militarized
Interstate Dispute Data, Version 3.0." Available at:
http://cow2.1a.psu.edu.
Goertz, Gary, and Paul F Diehl. 1995. "The Initiation and
Termination of Enduring Rivalries: The Impact of Political
Shocks." American Journal of Political Science 39(1):30-52.
Goldstone, Jack A. 1991. Revolution and Rebellion in the Early
Modern World. Berkeley: University of California Press.
Gurr, Ted R., Barbara Harff, and Monty G. Marshall. 1997.
Codebook: Internal Wars and Failures of Governance, 1954-
1996. McLean, VA: Science Applications International
Corporation.
Hensel, Paul R., and Paul F. Diehl. 1994. "It Takes Two to Tango:
Non-Militarized Response in Interstate Disputes." Journal of
Conflict Resolution 38(3):479-506.
Herbst, Jeffrey. 1990. "War and the State in Africa." International
Security 14(4):117-39.
Herbst, Jeffrey. 1996/1997. "Responding to State Failure in
Africa." International Security 21(3):120-44.
Herbst, Jeffrey. 2000. States and Power in Africa: Comparative
Lessons in Authority and Control. Princeton: Princeton University
Press.
Jackman, Robert W. 1993. Power without Force: The Political
Capacity ofNation-States. Ann Arbor: Universityof Michigan
Press.
Jaggers, Keith. 1992. "War and the Three Faces of Power: War
Making and State Making in Europe and the Americas."
Comparative Political Studies 25(1):26-62.
Kacowicz, Arie M. 1998. Zones of Peace in the Third World: South
America and West Africa in Comparative Perspective. Albany:
SUNY Press.
Kirby, Andrew, and Michael D. Ward. 1991. "Modernity and the
Process of State Formation: An Examination of 20th Century
Africa." International Interactions 17(1):113-26.
Levi, Margaret. 1981. "The Predatory Theory of Rule." Politics
and Society 10(4):431-65.
L6pez-Alves, Fernando. 2000. State Formation and Democracy in
Latin America, 1810-1900. Durham: Duke University Press.
L6pez-Alves, Fernando. 2001. "The Transatlantic Bridge: Mirrors,
Charles Tilly, and State Formation in the River Plate."
In The Other Mirror: Grand Theory through the Lens of
Latin America, ed. Miguel A. Centeno and Fernando L6pezAlves.
Princeton: Princeton University Press, pp. 153-
76.
Lustick, Ian. 1997. "The Absence of Middle Eastern Great
Powers: Political 'Backwardness' in Historical Perspective."
International Organization 51(4):653-83.
Mahoney, James. 2001. The Legacies of Liberalism: Path Dependence
and Political Regimes in Central America. Baltimore:
The Johns Hopkins University Press.
Mares, David R. 2001. Violent Peace: Militarized Interstate Bargaining
in Latin America. New York: Columbia University
Press.
North, Douglass C. 1981. Structure and Change in Economic
History. New York: W.W. Norton.
Organski, A.F.K., and Jacek Kugler. 1980. The War Ledger.
Chicago: University of Chicago Press.
Peacock, Alan T., and Jack Wiseman. 1961. The Growth of Public
Expenditures in the United Kingdom. Princeton: Princeton
University Press.
Pierson, Paul. 1996. "The Path to European Integration: A Historical
Institutionalist Analysis." Comparative Political Studies
29(2):123-63.
Pierson, Paul. 2003. "Big, Slow, and...Invisible: MacroSocial
Processes in the Study of Comparative Politics." In
Comparative-Historical Analysis: Achievements and Agendas,
ed. James Mahoney and Dietrich Rueschemeyer. Cambridge:
Cambridge University Press, pp. 177-207
Pierson, Paul, and Theda Skocpol. 2002. "Historical Institutionalism
in Contemporary Political Science." In Political Science:
The State of the Discipline, ed. I. Katznelson and H.V. Milner.
New York: W.W. Norton & Company, pp. 693-721.
Rasler, Karen A., and William R. Thompson. 1985. "War Making
and State Making: Governmental Expenditures, Tax Revenues,
and Global Wars." American Political Science Review
79(2):491-507.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms
WAR, RIVALRY AND STATE BUILDING 465
Rouqui6, Alain. 1987. The Military and the State in L
ica. Berkeley: University of California Press.
Russett, Bruce M. 1970. What Price Vigilance? The B
National Defense. New Haven: Yale University Pres
Stubbs, Richard. 1999. "War and Economic Devel
Export-Oriented Industrialization in East and S
Asia." Comparative Politics 31(1):337-55.
Thies, Cameron G. 2001a. "An Historical Institutio
proach to the Uruguay Round Agricultural Nego
Comparative Political Studies 34(4):400-28.
Thies, Cameron G. 2001 lb. "Territorial Nationalism in
valries: An Institutionalist Account of the Argentin
Rivalry." International Interactions 27(4):399-431.
Thies, Cameron G. 2004. "State Building, Intersta
trastate Rivalry: A Study of Post-Colonial Developi
try Extractive Efforts, 1975-2000." Internationa
Quarterly 48(1):53-72.
Thompson, William R. 1995. "Principal Rivalries."
Conflict Resolution 39(2):195-223.
Thompson, William R. 2001.
World Politics." Internatio
86.
Thorp, Rosemary. 1998. Progr
nomic History of Latin Am
Washington: Inter-Americ
Tilly, Charles. 1975. The Form
Europe. Princeton Universi
Tilly, Charles. 1985. "War M
nized Crime." In Bringing
Dietrich Rueschemeyer,
Cambridge University Pres
Tilly, Charles. 1992. Coercio
990-1992. Cambridge: Blac
Waldner, David. State Build
Cornell University Press.
Webber, Carolyn, and Aaron
tion and Expenditure in the
and Schuster.
This content downloaded from 147.251.55.39 on Mon, 11 Sep 2017 12:50:02 UTC
All use subject to http://about.jstor.org/terms