Abstracts

After the Napoleonic Wars interstate war regularly occurred throughout the Western Hemisphere—until in matter of decades it disappeared. After the 1930s even low-level militarized interstate conflict became less frequent, shorter, and less severe over time. What explains the change in this specific region and historical jucture? We argue that leaders in the Americas identified territorial disputes and foreign intervention as interrelated problems that frequently caused the interstate war. In response, they developed a unique regional norm-complex that bundled together the norms of territorial integrity and non-intervention with the principle of peaceful conflict resolution. This norm complex emerged via Latin American entrepreneurship shortly after independence, cascaded with Pan-Americanism, and crystallized around the signature of the Saavedra Lamas Treaty in the early 1930s. We explain how, why, and when norm complexes develop. We then investigate the evolution and effects of the Latin American norm complex via statistics and within-case counterfactuals. We conclude that interstate war disappeared from the Americas with the acceptance and codification of this norm-complex.

Después de las guerras napoleónicas, el hemisferio occidental experimentó guerras frecuentemente hasta que las mismas desparecieron. Después de la década de 1930 incluso los conflictos militarizados interestatales de baja letalidad se fueron haciendo infrecuentes, más cortos y menos letales ¿Qué explica el abrupto cambio en esa coyuntura histórica específica? En este artículo argumentamos que la respuesta se encuentra en un complejo normativo que agrupa a las normas de integridad territorial y no intervención bajo el principio de solución pacífica de controversias. Este complejo normativo surgió de la mano de emprendedores latinoamericanos, cobró impulso con el panamericanismo y terminó de cristalizarse en torno al Tratado de Saavedra Lamas a principios de la década de 1930. En lo que resta del artículo explicamos cómo, por qué y cuándo se desarrollan estos complejos normativos. A continuación, investigamos los efectos del complejo normativo latinoamericano a través del uso de estadísticas y análisis contrafácticos. Concluímos que la guerra interestatal desapareció del continente americano cuando este complejo normativo ganó una amplia aceptación y recibió una codificación significativa.

Après les guerres napoléoniennes, des guerres ont régulièrement eu lieu entre les États des Amériques, pour pratiquement disparaître dans les années 1930. Les conflits militarisés mineurs ont perduré, tout en devenant moins fréquents, plus courts et moins importants au fil du temps. Nous affirmons que les dirigeants des Amériques ont identifié deux problèmes interreliés qui pouvaient engendrer des guerres entre les États. Ils ont ensuite mis au point un complexe normatif régional. Il regroupait deux normes principales (intégrité territoriale et non-intervention) et une norme corollaire (arbitrage). Ce complexe normatif est apparu en Amérique Latine peu après l'indépendance et s'est propagé avec le panaméricanisme, pour finalement se cristalliser autour du Traité Saavedra Lamas au début des années 1930. Nous expliquons d'abord comment, pourquoi et quand les complexes normatifs voient le jour, et analysons ensuite l’évolution du complexe normatif latino-américain au moyen de statistiques et d’éléments contrefactuels. Les guerres interétatiques ont disparu dans les Amériques avec l'acceptation large de ce complexe normatif et sa codification.

INTRODUCTION

During the century between the Napoleonic Wars and World War I, interstate war occurred twice as frequently in the Americas as in Europe. These wars also lasted four times longer than and produced five times more per capita battle deaths than their European counterparts (Schenoni 2024). At the end of the nineteenth century, the Americas then began transitioning towards peace, and after the Chaco War (1932–1935) ended, a “long peace” ensued. War disappeared as peace consolidated further, and the remaining, low-level militarized conflict became less frequent, shorter, and less severe—a trend that distinguishes the Americas from other regions and persists today.

What explains the abrupt change towards peace in this particular region and historical junture? We propose that the answer lies in the development of a regional norm complex—that is, a set of mutually reinforcing rules and principles that interrelate logically and historically. Norm complexes broadly develop according to the classic norm-cycle framework; they emerge, tip, cascade, and consolidate, gathering greater acceptance as the cycle advances and actors socialize one another to behavioral expectations (Finnemore and Sikkink 1998). Advancement, however, does not proceed linearly; it begins slow (incremental acceptance), accelerates (a rapid shift in acceptance), and then levels off (saturation causes a return to incremental acceptance). To explain this S-shaped dynamic, we reframe norms as solutions to policy problems and integrate the norm-cycle framework with a punctuated equilibrium model (Baumgartner and Jones 2009, 17).

Any significant policy change requires that actors (1) recognize and give attention to a problem, (2) have recourse to a solution, and (3) operate in an environment whose politics does not impede the solution's adoption (Kingdon 1984). Policy therefore tends toward stasis, until the rare occasions when the problem, solution, and politics streams align. That alignment often comes through shocks, after which policy changes quickly and abruptly—before settling into a (new) stasis.

Such a process broadly explains how the Western Hemisphere eradicated war. After independence, Latin American elites recognized two interrelated problems that initially threatened the survival (and later, sovereignty) of their states: violations of territorial integrity and major-power intervention, both of which frequently threatened or caused wars (i.e., another existential problem). No existing solution addressed these problems well (or holistically), so norm entrepreneurs advanced the foundations for one (norm emergence). Their solution took the form of a norm complex that combined two primary norms (territorial integrity and non-intervention) and one corollary norm (mechanisms for peacefully managing issues under the two primary norms, particularly arbitration). To secure wider support, the entrepreneurs had to convince others of the problems’ import, the solution's merits, and the wisdom of adopting that solution (i.e., the politics). Additional shocks (e.g., militarized episodes) helped these efforts; they highlighted the problem, demonstrated the urgent need for a solution, and opened windows of opportunity for reluctant actors to accept the norm complex. Once critical actors (e.g., the United States) accepted the solution's general outlines, acceptance tipped and accelerated (norm cascade), and actors began negotiating the norm complex's more formal and precise obligations.

These negotiations led to a stage—crystallization—that we add to the norm-cycle framework. During crystallization, actors fully develop, integrate, and codify the entirety of a norm complex's solution. Institutionalization begins early in the norm cycle. Entrepreneurs advance doctrines as possible solutions to the policy problem. Shocks, which highlight the problem, encourage conferences to discuss the problem and the available solutions (e.g., doctrines), incentivize more actors to join the process and produce outputs of their own (e.g., resolutions and treaties). A focal point solution eventually emerges, and the positive feedback loop this generates leads actors to return to and flesh out that solution in the wake of additional shocks (e.g., potential violations). At some point, however, this refinement accelerates and completes (i.e., abrupt policy change). This is the crystallization stage. After it ends, actors consolidate, restate, and reinforce the norm complex solution, but do not change its obligations significantly (norm consolidation).

Our Latin American norm complex crystallizes rapidly in the early 1930s—a period of critical import in the development of the region's international law (Becker Lorca 2014; Friedman and Long 2015; Scarfi 2017). During that period, states signed many treaties, but the 1933 Saavedra Lamas Treaty deserves pride of place. It was the first Inter-American treaty to integrate all three components of the norm complex: territorial integrity, strict non-intervention, and a robust mechanism for the collective, regional management of disputes.

Crystallization constitutes a key behavioral moment. Although actors increasingly conform to a norm complex's prescriptions as it strengthens, their behavior changes most drastically after crystallization. Actors then comply with the complex's obligations not only because of a “logic of appropriateness,” but also because of a “logic of consequences.” Violating the obligations carries material—as well as social—costs. In this way, constructivist and rational-institutionalist logics reinforce one another via the rational design of norms-based institutions (on the interrelationship of norms and institutions, see Koremenos et al. 2001).

Our study offers three main contributions. First, it extends the classic norm-cycle framework in two ways: supplying one causal mechanism that drives the cycle forward and theorizing a crystallization period. Second, it revindicates the Global South's importance in international norm development. Existing works privilege North American and European initiatives (e.g., the Kellogg–Briand Pact; see Hathaway and Shapiro 2017). These works, however, obscure how such initiatives mirrored earlier Pan-American ones, championed by Latin American states (see also Finnemore and Jurkovich 2014; Sikkink 2017). Finally, Latin American scholars identify the 1930s as a critical period in the region's international legal-normative development (Becker Lorca 2014; Friedman and Long 2015; Scarfi 2017). Our study situates their contention within a larger theoretical and historical context, while also demonstrating the behavioral effects of the norm complex's crystallization.

In the sections that follow, we first set the empirical puzzle by documenting a significant change in interstate militarized violence in Latin America after the early 1930s. We next outline a general theory that explains why, how, and when this occurs—and traces that theoretical process through the Latin American norm complex's development. Then, the remaining empirical sections quantitatively connect our norm complex's development to the dissipation of war; a decline in territorial disputes, territorial conquest, and foreign intervention; the wide use of arbitration; a shift in disputed issues; and a decline in the frequency and severity of the norm-related militarized conflict that remains. These trends consistently mirror not only the norm cycle's theoretically S-shaped curve but also feature a critical juncture around the time that the norm complex crystallizes. Mindful that quantitative data cannot illustrate the norm complex's behavioral constraints, these empirical sections also qualitatively explore two state-level counterfactuals—one for each primary norm—to do just that. Finally, we conclude with a brief discussion of potential alternative explanations and generalizability.

From War to Minor Conflict

We argue that a norm complex crystallized in the Americas during the early 1930s, and as a result, the region eradicated war and fundamentally altered interstate conflict behavior short of war. This argument not only sets the early 1930s as a critical period but also predicts that regional conflict behavior before and after the early 1930s differs significantly. Before proceeding to our detailed argument and the multi-method evaluation of it, we, therefore, ask a foundation question: Is the empirical record broadly consistent with this central prediction, particularly with respect to our main outcome of interest—war?

Table 1 lists every episode that at least one prominent study colloquially identifies as a post-independence, interstate war in the Americas during the period 1816-present, along with its dates, battle deaths, and duration. If we define war as a military confrontation that produces at least 1,000 battle deaths (hereinafter, “scholarly war”; asterisked in table 1; see Sarkees and Wayman 2010), an empirical puzzle indeed emerges. No new scholarly wars began after 1933; the Leticia (1932) and Chaco Wars (1932) were the region's last.1

Table 1.

Interstate war in the Americas, 1816–2010.

WarYearBattle deathsDuration (months)
Argentina-Brazil1825–28*8,00032
Peru-Great Colombia1828–29*4,0007
Bolivia-Peru War183560010
War of the Confederation1837–39*2,90034
War of Texan Independence1836*1,7006
Pastry War18392883
Peru-Bolivia1841*1,0007
Mexican-American1846–47*19,28317
La Plata1843–52*6,30091
Haiti-Dominican1855–56*1,00014
Filibuster1856–57*7,70022
Franco-Mexican1862–67*35,85058
(First) Central American1863*1,00010
Ecuador-Colombia18631481
War of the Restoration1863–65*14,88823
Paraguayan War1864–70*130,00064
Chincha Islands1865–667948
Ten Years War1868–78*90,000117
(Second) Central American1876*6,0001
War of the Pacific1879–83*13,86846
(Third) Central American1885*1,0001
Cuban Independence1895–98*5,96635
Spanish-American1898*3,6853
(Fourth) Central American1906*1,0002
(Fifth) Central American1907*1,0002
Leticia War1932*1,0578
Chaco War1932–35*88,39733
Zarumilla War19418000.5
Soccer/Football War19692060.133
Paquisha War198191
Malvinas/Falklands War19829042
Cenepa War1995931
WarYearBattle deathsDuration (months)
Argentina-Brazil1825–28*8,00032
Peru-Great Colombia1828–29*4,0007
Bolivia-Peru War183560010
War of the Confederation1837–39*2,90034
War of Texan Independence1836*1,7006
Pastry War18392883
Peru-Bolivia1841*1,0007
Mexican-American1846–47*19,28317
La Plata1843–52*6,30091
Haiti-Dominican1855–56*1,00014
Filibuster1856–57*7,70022
Franco-Mexican1862–67*35,85058
(First) Central American1863*1,00010
Ecuador-Colombia18631481
War of the Restoration1863–65*14,88823
Paraguayan War1864–70*130,00064
Chincha Islands1865–667948
Ten Years War1868–78*90,000117
(Second) Central American1876*6,0001
War of the Pacific1879–83*13,86846
(Third) Central American1885*1,0001
Cuban Independence1895–98*5,96635
Spanish-American1898*3,6853
(Fourth) Central American1906*1,0002
(Fifth) Central American1907*1,0002
Leticia War1932*1,0578
Chaco War1932–35*88,39733
Zarumilla War19418000.5
Soccer/Football War19692060.133
Paquisha War198191
Malvinas/Falklands War19829042
Cenepa War1995931

Sources:  Mares (2001), Centeno (2002), Wimmer and Min (2009), and Sarkees and Wayman (2010). For battle-death data, we consult Clodfelter (2017), Gibler (2018), and official government sources.

Data availability determines the temporal range. To 2024 no additional wars have occurred. *Denotes war by the scholarly definition (Sarkees and Wayman 2010).

Table 1.

Interstate war in the Americas, 1816–2010.

WarYearBattle deathsDuration (months)
Argentina-Brazil1825–28*8,00032
Peru-Great Colombia1828–29*4,0007
Bolivia-Peru War183560010
War of the Confederation1837–39*2,90034
War of Texan Independence1836*1,7006
Pastry War18392883
Peru-Bolivia1841*1,0007
Mexican-American1846–47*19,28317
La Plata1843–52*6,30091
Haiti-Dominican1855–56*1,00014
Filibuster1856–57*7,70022
Franco-Mexican1862–67*35,85058
(First) Central American1863*1,00010
Ecuador-Colombia18631481
War of the Restoration1863–65*14,88823
Paraguayan War1864–70*130,00064
Chincha Islands1865–667948
Ten Years War1868–78*90,000117
(Second) Central American1876*6,0001
War of the Pacific1879–83*13,86846
(Third) Central American1885*1,0001
Cuban Independence1895–98*5,96635
Spanish-American1898*3,6853
(Fourth) Central American1906*1,0002
(Fifth) Central American1907*1,0002
Leticia War1932*1,0578
Chaco War1932–35*88,39733
Zarumilla War19418000.5
Soccer/Football War19692060.133
Paquisha War198191
Malvinas/Falklands War19829042
Cenepa War1995931
WarYearBattle deathsDuration (months)
Argentina-Brazil1825–28*8,00032
Peru-Great Colombia1828–29*4,0007
Bolivia-Peru War183560010
War of the Confederation1837–39*2,90034
War of Texan Independence1836*1,7006
Pastry War18392883
Peru-Bolivia1841*1,0007
Mexican-American1846–47*19,28317
La Plata1843–52*6,30091
Haiti-Dominican1855–56*1,00014
Filibuster1856–57*7,70022
Franco-Mexican1862–67*35,85058
(First) Central American1863*1,00010
Ecuador-Colombia18631481
War of the Restoration1863–65*14,88823
Paraguayan War1864–70*130,00064
Chincha Islands1865–667948
Ten Years War1868–78*90,000117
(Second) Central American1876*6,0001
War of the Pacific1879–83*13,86846
(Third) Central American1885*1,0001
Cuban Independence1895–98*5,96635
Spanish-American1898*3,6853
(Fourth) Central American1906*1,0002
(Fifth) Central American1907*1,0002
Leticia War1932*1,0578
Chaco War1932–35*88,39733
Zarumilla War19418000.5
Soccer/Football War19692060.133
Paquisha War198191
Malvinas/Falklands War19829042
Cenepa War1995931

Sources:  Mares (2001), Centeno (2002), Wimmer and Min (2009), and Sarkees and Wayman (2010). For battle-death data, we consult Clodfelter (2017), Gibler (2018), and official government sources.

Data availability determines the temporal range. To 2024 no additional wars have occurred. *Denotes war by the scholarly definition (Sarkees and Wayman 2010).

Even were we counterfactually to consider all episodes in table 1 as (scholarly) wars, 1933—or the end of our norm complex's crystallization period—represents a better breakpoint than any other year (e.g., 1881 or 1883, Kacowicz 1998; or 1945, Holsti 1996). Whereas, on average, 0.23 major conflicts begin annually before 1933, only 0.06 wars begin annually thereafter—a 74 percent decrease. The later “wars,” moreover, differ in kind from their predecessors. Prior to 1933, each “war” produced an average of 16,571 battle deaths; after 1933, battle deaths plummeted to roughly 402 per armed conflict (a 97 percent decrease), even as populations and army sizes grew. Finally, a pre-1933 “war” lasted approximately 2 years, whereas a post-1933 one lasted less than a month (a 96 percent decrease).

War is admittedly not the only form of militarized conflict. States also threaten, display, and use force short of war (i.e., Militarized Interstate Disputes, or MIDs; Palmer et al. 2021). Table 2 examines measures of MID frequency, severity, and duration for the sub-periods 1816–1933 and 1934–2010, distinguishing between conflicts within and outside the Americas. The analysis therein yields three main conclusions. First, controlling for the number of states, the MID frequency falls significantly inside (94 percent), but not outside (22 percent), the Americas across the two periods. Second, MID severity does likewise. It declines by a statistically significant 33 percent within the Americas, but a statistically insignificant 3 percent outside the Americas (for these data, see Goertz et al. 2023). Finally, MID duration in the Americas drops, on average, from 401 days before 1933 to 200 days after 1933—a statistically significant, 50 percent decrease. Outside the Americas, however, the average MID duration significantly increases by 37 percent.

Table 2.

Interstate conflict before and after 1935.

The AmericasOutside the Americas
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Dyadic MIDs per year0.01280.0008−3.1579***0.01000.0078−1.2846
Dyadic MID severity81.982755.2782−4.6512***78.625176.0006
−0.8347
Dyadic MID duration400.8557199.7320−4.0785***324.1755445.66075.3082***
The AmericasOutside the Americas
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Dyadic MIDs per year0.01280.0008−3.1579***0.01000.0078−1.2846
Dyadic MID severity81.982755.2782−4.6512***78.625176.0006
−0.8347
Dyadic MID duration400.8557199.7320−4.0785***324.1755445.66075.3082***

Notes: Excluding WWII years (1939–1945) yields similar results; duration is in days; *p < 0.10, **p < 0.05, and ***p < 0.01.

Table 2.

Interstate conflict before and after 1935.

The AmericasOutside the Americas
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Dyadic MIDs per year0.01280.0008−3.1579***0.01000.0078−1.2846
Dyadic MID severity81.982755.2782−4.6512***78.625176.0006
−0.8347
Dyadic MID duration400.8557199.7320−4.0785***324.1755445.66075.3082***
The AmericasOutside the Americas
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Mean
1816–1933
Mean
1934–2010
Difference in means
(t-value)
Dyadic MIDs per year0.01280.0008−3.1579***0.01000.0078−1.2846
Dyadic MID severity81.982755.2782−4.6512***78.625176.0006
−0.8347
Dyadic MID duration400.8557199.7320−4.0785***324.1755445.66075.3082***

Notes: Excluding WWII years (1939–1945) yields similar results; duration is in days; *p < 0.10, **p < 0.05, and ***p < 0.01.

In short, conflict in the Americas appeared similar to that found outside the region prior to the early 1930s, but not thereafter. A dramatic change occurred in the Western Hemisphere. Why did this change occur—and why in the early 1930s?

The Causal Mechanism Producing Peace

Any explanation for the evolution of peace in the Americas must rely on factors that kick in around the critical period we identify above—factors whose absence before the early 1930s permitted war to occur and whose presence thereafter constrained it. Because United States hegemony dwindled during the early 1930s (Grandin 2012; Butt 2013), the only compelling explanation involves norms and institutions (Holsti 1996; Kacowicz 2005), particularly the codification of the principles of territorial integrity and non-intervention.2

According to Finnemore and Sikkink's (1998) classic framework, norms pass through three stages as they develop: (1) emergence, during which individual entrepreneurs champion a norm; (2) cascade, during which actors increasingly accept the norm's behavioral constraints; and (3) internalization (or incremental-consolidation), during which actors reinforce the norm and eventually adhere to it routinely. This norm cycle explains how norms develop, but not precisely when or why (e.g., see Finnemore and Sikkink 1998, 901). Answers to the latter questions come through integrating the norm cycle with a punctuated equilibrium model of policy development (Baumgartner and Jones 2009), which supplies the mechanism that both produces the norm cycle's S-shaped curve and institutionalizes the norm.

Any major policy change requires attention to a policy problem, an existing solution to that problem, and a permissive environment in which policy change can occur (Kingdon 1984); yet these three streams—problem, solution, and politics—rarely converge. Policymakers face competing priorities, encounter a flood of information about diverse problems and their potential solutions, must find and debate solutions, and need to minimize any political impediments to policy change. Policy consequently tends toward stability. At times, however, problems emerge—often through shocks—that existing policy addresses poorly. Policymakers will first attempt to tinker with, or minimally adjust, the existing policy (Goertz 2003). If that fails and the demand for a policy solution becomes urgent, they then shift their attention, reprioritize, negotiate, and (on occasion) achieve compromises that abruptly change policy. A period of (new) policy stasis then follows until the process repeats. Importantly, the “new” stasis need not be the “old” stasis; policy can evolve. Such a punctuated-equilibrium model predicts long periods in which policy advances incrementally—as policy-makers tinker, and information and urgency build—peppered with episodic, abrupt, and more dramatic policy changes.

The dynamic of this model effectively powers the norm cycle engine. A policy-driven problem arises (e.g., a war that violates territorial sovereignty), which shocks the system. Norm entrepreneurs then advance a particular norm—or in the case of a multi-faceted problem, a set of interrelated norms (i.e., a norm complex)—as a potential solution to the problem (e.g., territorial integrity and non-intervention), and try to persuade other key actors to accept it (norm emergence). Because norms ultimately constrain behavior, that persuasion campaign takes time, and additional shocks play a critical role during it. Shocks, particularly manifestations or consequences of the policy problem (e.g., additional territorial or interventionary wars), create windows of opportunity for policy change (i.e., the politics). By underscoring the problem's persistence and the urgent need for a solution, these shocks eventually incentivize a small number of actors to agree on some foundational premises. Through these premises, a focal point solution emerges.3

The focal point creates a positive feedback loop. Subsequent shocks reiterate the urgent need for a solution, bring the problem to other actors’ attention, soften reluctant actors to the solution's foundational premises, and encourage entrepreneurs to take the next step in the norm complex's development—a type of path dependence. When these shocks occur, those working on the solution return to and develop the focal point's foundational premises further. After a sufficient number of actors accept the norm's foundational premises as a solution to the policy problem, norm acceptance crosses a tipping point, after which acceptance of it accelerates (norm cascade). Actors then negotiate and codify the norm's precise obligations in far greater detail.

The codification of normative solutions proceeds in a path-dependent fashion throughout the norm cycle. Doctrines initially set the agenda, place the problem on policymakers’ radars, and propose solutions. Conferences follow when enough actors recognize the problem and need to discuss a solution. Treaties—which often emerge as outputs of conferences later in the norm cycle—define the problem more concretely, craft specific solutions in response, and formally institutionalize obligations.

Each doctrine, conference, and treaty builds on what came before. It further clarifies the problem and solution; causes behavioral expectations to converge; and encourages the next doctrine, conference, or treaty. At some point, however, actors enter an intense codification period, during which they fully develop, integrate, and formally codify the entirety of their normative solution. This significant institutionalization characterizes the crystallization stage. Once crystallization ends, actors perhaps restate (or reinforce) their normative obligations in additional treaties, but the content of these obligations does not change significantly (incremental-consolidation). Their behavior, moreover, complies with the norm's prescriptions most strongly in this final stage, as a result of both constructivist (i.e., “appropriateness”) and institutionalist mechanisms (e.g., the “consequences” of non-compliance).4

The above process applies to the development of individual norms, as well as norm complexes—or sets of mutually reinforcing rules and principles that interrelate logically and historically. Figure 1 (left panel) visually summarizes the general theoretical argument through which a norm gains acceptance—from emergence to cascade to crystallization to incremental consolidation. A norm complex abstractly diffuses (light gray S-shaped line), constraining behavior somewhat by defining “appropriate” actions. The subsequent bundling and translation of the complex's norms into specific legal obligations (black S-shaped line) provides further constraints, for it aligns actors’ material incentives with the complex's prescriptions. This, however, takes time to develop. Legalization advances when the policy, solution, and politics streams converge—often as a result of shocks (e.g., wars) that highlight the policy problem and underscore the need for the (norm-based) solution. Each codification effort (e.g., doctrine, conference, and treaty), moreover, builds on prior efforts (i.e., a positive feedback loop derived from the norm complex as a focal point solution). Legalization therefore deepens over time. Actors iteratively create more precise obligations to curb unwanted behavior, and norm acceptance consequently expands.

Norm complex development: General and as applied to the Americas
Figure 1.

Norm complex development: General and as applied to the Americas

The remainder of this section presents a historical narrative that illustrates how a norm complex developed via the above, general process to virtually eliminate war in the Americas (see right panel of figure 1; see also Appendix A, online). Newly independent states recognized two major threats to state sovereignty: violations of territorial integrity and, relatedly, major-power intervention (e.g., attempts at re-colonization). Both threatened frequent, intense wars—another problem, not only because of the costs involved but also because such wars eased the path for further major-power intervention. Regional leaders consequently bundled these interrelated problems together and created a norm complex to address them collectively. This complex contains two primary norms, territorial integrity and non-intervention, as well as one corollary—the expectation that states manage disagreements related to the primary norms through peaceful conflict management tools, such as arbitration (see Powell and Wiegand 2021). This complex eradicated war in the Americas.

Emergence

The norm complex originated in the early nineteenth century. After independence, Latin American boundaries often lacked clarity, local rulers frequently used military force to address territorial disagreements, and European states regularly intervened in regional affairs (e.g., to attempt recolonization). Elites in Latin America consequently identified territorial conflict and major-power intervention as important, interrelated policy problems that threatened the survival and sovereignty of their new states. In response, a unique form of legalismo emerged (Jenne 2022), originally expressed in the form of two doctrines. First, the uti possidetis doctrine (1810–) based new states’ territorial limits on pre-independence administrative boundaries (Kacowicz 2005). Second, the 1823 Monroe Doctrine deemed European intervention in the Americas unlawful (Sexton 2011). Both doctrines sought to establish, clarify, and secure territorial sovereignty.

Because doctrines, by definition, are neither formally codified nor widely accepted (e.g., across a region), the problems persisted. Nevertheless, by 1837, Venezuelan jurist Andres Bello realized that the emerging doctrines constituted a distinct corpus of indigenous regional norms—one concerned with both territorial integrity and non-intervention. Table 3 lists the general doctrines within that corpus. As the table shows, doctrines punctuate both the emergence (–1888) and subsequent (1889–) stages of our norm complex—primarily because norm evolution involves a dialectic, in which actors define a problem-solution (set), transgressors find loopholes (i.e., create shocks), and renewed entrepreneurship closes the loopholes, thereby advancing norm strength further (i.e., a form of positive feedback/path dependence). As an illustration, European powers eventually acquiesced to the Monroe Doctrine but continued intervening in the region to protect foreign nationals and collect debt. During the Second French Intervention (1861–1867), President Benito Juarez consequently furthered Mexican political commitments to a broader non-intervention norm via the Juarez Doctrine. Carlos Calvo, an Argentine legal scholar, later argued that states should exhaust local dispute resolution mechanisms before resorting to force for either debt collection or citizen protection (Friedman and Long 2015).

Table 3.

Doctrines of international public law originating in the Americas, 1810–1933.

DoctrineYearPrinciple enshrined
Uti possidetis1810Territorial integrity
Monroe doctrine1823Non-intervention
Bello doctrine1837New Inter-American corpus
Juarez doctrine1867Non-intervention
Calvo doctrine1868Non-intervention
Varela doctrine1869Territorial integrity
Díaz doctrine1896Non-intervention
Drago doctrine1902Non-intervention
Tobar doctrine1907Limited recognition
Carranza doctrine1917Expanded recognition
Brum doctrine1927Non-intervention
Estrada doctrine1930Absolute non-intervention
Stimson doctrine1932Territorial integrity
DoctrineYearPrinciple enshrined
Uti possidetis1810Territorial integrity
Monroe doctrine1823Non-intervention
Bello doctrine1837New Inter-American corpus
Juarez doctrine1867Non-intervention
Calvo doctrine1868Non-intervention
Varela doctrine1869Territorial integrity
Díaz doctrine1896Non-intervention
Drago doctrine1902Non-intervention
Tobar doctrine1907Limited recognition
Carranza doctrine1917Expanded recognition
Brum doctrine1927Non-intervention
Estrada doctrine1930Absolute non-intervention
Stimson doctrine1932Territorial integrity
Table 3.

Doctrines of international public law originating in the Americas, 1810–1933.

DoctrineYearPrinciple enshrined
Uti possidetis1810Territorial integrity
Monroe doctrine1823Non-intervention
Bello doctrine1837New Inter-American corpus
Juarez doctrine1867Non-intervention
Calvo doctrine1868Non-intervention
Varela doctrine1869Territorial integrity
Díaz doctrine1896Non-intervention
Drago doctrine1902Non-intervention
Tobar doctrine1907Limited recognition
Carranza doctrine1917Expanded recognition
Brum doctrine1927Non-intervention
Estrada doctrine1930Absolute non-intervention
Stimson doctrine1932Territorial integrity
DoctrineYearPrinciple enshrined
Uti possidetis1810Territorial integrity
Monroe doctrine1823Non-intervention
Bello doctrine1837New Inter-American corpus
Juarez doctrine1867Non-intervention
Calvo doctrine1868Non-intervention
Varela doctrine1869Territorial integrity
Díaz doctrine1896Non-intervention
Drago doctrine1902Non-intervention
Tobar doctrine1907Limited recognition
Carranza doctrine1917Expanded recognition
Brum doctrine1927Non-intervention
Estrada doctrine1930Absolute non-intervention
Stimson doctrine1932Territorial integrity

Regional conferences began debating and codifying various doctrines as early as the 1826 Panama Conference, called by President Simón Bolívar of Gran Colombia. In the series of Latin American–led, “Bolivarian” conferences that followed (Kacowicz 2005, 49), regional leaders worked doggedly to bundle together and strengthen the principles of territorial integrity and non-intervention—prompted each time by shocks that reiterated the policy problems and the urgent need for agreed-upon solutions. During the Mexican–American War, for example, the 1847 Lima Conference codified uti possidetis to resolve territorial disagreements among neighboring states and encouraged regional unity against major-power intervention. Similarly, the Santiago Conference (1856) responded to William Walker's attempt to conquer Central America, while the Second Lima Conference (1864) convened as Spain invaded Peru and French troops crowned a Mexican Emperor. Conferences such as these promoted regional cooperation against major-power intervention, enshrined territorial integrity, and committed states to provide good offices and to accept arbitration to manage regional disagreements. These three goals linked for a common purpose: to secure state survival via the creation of a common, Latin American front.5

During this emergence stage, states did not widely adopt the embryonic norms as a policy guide. Instead, various shocks, doctrines, and conferences persuaded a growing cadre of states to devote attention to the problem (i.e., recurrent war and threats to state survival/sovereignty), to agree on some foundational premises for a solution, which became the focal point for continued discussions (i.e., that territorial integrity and non-intervention, however defined, offered a solution), and to begin designing alternative mechanisms to handle disputes under the emerging norm complex (i.e., arbitration; see the 1856 Santiago Conference). As more regional powers with the legal and diplomatic capacity to back the norms joined in—most notably, Argentina, Brazil, Chile, Peru, and Mexico—the norm complex approached the cascade stage.

Cascade

Around 1889, the norm complex hit a tipping point and transitioned to the cascade stage. Latin American states provided the entrepreneurship that drove the emergence stage; they not only developed most of the doctrines that appeared in this period (see table 3) but also organized the entire Bolivarian conference series. United States representatives either could not attend the Bolivarian conferences when invited (e.g., the 1847 Lima Conference, which convened during the Mexican–American War), or were excluded by organizers, particularly when Washington's behavior seemed to violate the norm complex's emerging foundational premises (e.g., the 1856 Santiago Conference, which convened during William Walker's Nicaraguan intervention).

Elites in Washington eventually sought to reassure their southern neighbors, stabilize the region, and open new markets. To achieve these goals, the United States recognized some of the foundational premises that the Bolivarian conferences developed—namely, territorial integrity and the peaceful resolution of disagreements—and inaugurated the Pan-American conference series, within which the United States held a prominent (albeit repeatedly contested) leadership role.

The First Pan-American Conference (1889) illustrates many characteristics common to the cascade stage. First, more actors joined in the norm complex's development. Unlike any prior international conference, the 1889 Conference adopted a rule of universal participation. That rule became a hallmark of Pan-American conferences (Finnemore and Jurkovich 2014). Second, the involvement of more actors signaled growing (minimal) agreement on the norm complex's foundational premises, even as actors continued bargaining about the exact obligations that sat atop those premises (Stimmer 2019). The United States, for example, opposed the intervention of European powers in the region, but itself benefited from the existence of a legal right to intervene in cases such as the protection of citizens or collection of debt. Absolute non-intervention, which Latin American states pushed, would constrain the United States too much. During conference negotiations, the North American diplomats consequently worked to preserve an option to intervene.

Third, actors began sewing the individual norms that comprised the complex more tightly together. For example, according to the recommended Plan of Arbitration treaty that the conference produced, arbitration would now be compulsory for most disputed issues, including “boundaries” and “territories.” Fourth, the actors began to codify the norm complex more elaborately. One unratified treaty that resulted from the 1864 (Second) Lima Conference, for example, sketched an arbitration process, but with significantly less detail than the later Plan of Arbitration. In fact, the 1889 Conference even produced a proposal for a Permanent Court of Arbitration (PCA); that proposal crossed the Atlantic, where states at the 1899 Hague Conference adopted it. Given such details, as well as broad regional agreement on them, the arbitration of regional territorial disputes expanded significantly in the wake of the First Pan-American Conference (see below; Owsiak et al. 2018).

Finally, bargains struck during the First Pan American Conference display “compensatory layering,” in which “transformative institutional change emerges from sequential bargaining over organizational design” (Long and Schulz 2022, 9). Latin American states resisted United States plans for a continental customs union at the conference but agreed—through a back-and-forth that accommodated the interests of both Latin America and Washington—to create the Commercial Bureau of the International Union of American Republics. Subsequent Pan-American conferences reveal the same dynamic with respect to the norm complex's content. The United States ultimately wanted greater regional stability and cooperation, while Latin American states primarily sought a broader non-intervention norm. This give-and-take further encouraged the region to codify the territorial integrity and non-intervention norms together—via instruments that provided peaceful dispute resolution mechanisms for violations of either norm.

Table 4 lists the main legal instruments that the Pan-American conferences produced (Fenochio 1953) to deal with “good offices, mediation, investigation, conciliation, arbitration, and judicial settlement” (Kacowicz 2005, 104–5). These instruments cover the cascade, crystallization, and (early) consolidation stages. The resolution to adhere to and ratify the 1933 Peace Treaties at the Montevideo Conference constitutes a critical juncture; it brings preexisting treaties together, binds all American states, and formally integrates and bundles the individual norms within the complex—across and within individual treaties. The 1936 agreements, in contrast, expressly intend to strengthen the previous consensus and to emphasize conflict prevention, rather than resolution.

Table 4.

Conflict resolution mechanisms endorsed at Pan-American conferences.

YearTitle of the instrument
1890Resolution on arbitration with European powers
1890Project for a permanent court of arbitration
1890Resolution on the right of conquest
1890Recommendation for an arbitration treaty
1902Protocol of Adherence to the Hague Conventions
1902Treaty of Compulsory Arbitration
1906Resolution on extended compulsory arbitration
1923Gondra Treaty
1928General Treaty of Inter-American Arbitration
1928Protocol for Progressive Arbitration
1928General Convention of Inter-American Conciliation
1933Protocol to the Convention of Conciliation
1933Declaration of Intention to Subscribe all Instruments for Peaceful Resolution
1933Resolution to ratify all “pacifist agreements”
1936Convention on the Maintenance, Strengthening and Reestablishment of Peace
1936Convention to Coordinate, Amplify, and Ensure the Commitment to Peace Treaties
1936Treaty for the Prevention of Controversies
YearTitle of the instrument
1890Resolution on arbitration with European powers
1890Project for a permanent court of arbitration
1890Resolution on the right of conquest
1890Recommendation for an arbitration treaty
1902Protocol of Adherence to the Hague Conventions
1902Treaty of Compulsory Arbitration
1906Resolution on extended compulsory arbitration
1923Gondra Treaty
1928General Treaty of Inter-American Arbitration
1928Protocol for Progressive Arbitration
1928General Convention of Inter-American Conciliation
1933Protocol to the Convention of Conciliation
1933Declaration of Intention to Subscribe all Instruments for Peaceful Resolution
1933Resolution to ratify all “pacifist agreements”
1936Convention on the Maintenance, Strengthening and Reestablishment of Peace
1936Convention to Coordinate, Amplify, and Ensure the Commitment to Peace Treaties
1936Treaty for the Prevention of Controversies
Table 4.

Conflict resolution mechanisms endorsed at Pan-American conferences.

YearTitle of the instrument
1890Resolution on arbitration with European powers
1890Project for a permanent court of arbitration
1890Resolution on the right of conquest
1890Recommendation for an arbitration treaty
1902Protocol of Adherence to the Hague Conventions
1902Treaty of Compulsory Arbitration
1906Resolution on extended compulsory arbitration
1923Gondra Treaty
1928General Treaty of Inter-American Arbitration
1928Protocol for Progressive Arbitration
1928General Convention of Inter-American Conciliation
1933Protocol to the Convention of Conciliation
1933Declaration of Intention to Subscribe all Instruments for Peaceful Resolution
1933Resolution to ratify all “pacifist agreements”
1936Convention on the Maintenance, Strengthening and Reestablishment of Peace
1936Convention to Coordinate, Amplify, and Ensure the Commitment to Peace Treaties
1936Treaty for the Prevention of Controversies
YearTitle of the instrument
1890Resolution on arbitration with European powers
1890Project for a permanent court of arbitration
1890Resolution on the right of conquest
1890Recommendation for an arbitration treaty
1902Protocol of Adherence to the Hague Conventions
1902Treaty of Compulsory Arbitration
1906Resolution on extended compulsory arbitration
1923Gondra Treaty
1928General Treaty of Inter-American Arbitration
1928Protocol for Progressive Arbitration
1928General Convention of Inter-American Conciliation
1933Protocol to the Convention of Conciliation
1933Declaration of Intention to Subscribe all Instruments for Peaceful Resolution
1933Resolution to ratify all “pacifist agreements”
1936Convention on the Maintenance, Strengthening and Reestablishment of Peace
1936Convention to Coordinate, Amplify, and Ensure the Commitment to Peace Treaties
1936Treaty for the Prevention of Controversies

All instruments in table 4—even the earliest—commit states to peacefully resolve disputes over territory and issues that historically motivated major-power intervention (e.g., debt collection or citizen protection). In 1902, however, Venezuela defaulted on its debt, and European powers blockaded it. Latin America reacted bitterly. The Argentinian Minister of Foreign Relations, Luis María Drago, argued that such behavior was impermissible; foreign powers could not use military force to collect debt. The 1906 Rio Conference supported this Drago Doctrine, thereby unifying the region's position on the matter and encouraging the region to spearhead two initiatives at the 1907 (Second) Hague Conference: compulsory arbitration through the PCA and absolute non-intervention. The United States, under regional pressure, reluctantly supported the latter, albeit “with fingers crossed” and hoping to find “a major loophole” (Friedman 2018, 212). Indeed, through the 1907 Drago-Porter Convention, the United States endorsed non-intervention for debt collection but preserved the right to intervene if the debtor state rejected arbitration of the debt disagreement.

Further norm institutionalization followed. At the 1906 Rio Conference, states created a commission of jurists to codify Inter-American law. The American Institute of International Law (AIIL), founded by James Brown Scott (United States) and Alejandro Alvarez (Chile), assisted in this process (Scarfi 2017). The involved scholars met repeatedly in legal Pan-American congresses (largely from 1912 to 1933), which shaped the agendas and outputs of subsequent political conferences (Petersen 2022). Typical of the cascade stage, the AIIL actively promoted acceptance and accelerated the legalization of the norm complex; it drastically reduced its activity when crystallization was completed in the early 1930s.

As expected, a series of important treaties appeared as the cascade stage reached its end. The Gondra Treaty, which emerged from the 1923 Santiago Conference, created mechanisms for compulsory conciliation and investigation. The 1928 Havana Conference then produced both a General Treaty of Inter-American Arbitration (1929) and a General Convention of Inter-American Conciliation (1929). The United States viewed these various instruments as pillars of a regional “peace machinery”; alongside the Kellogg–Briand Pact (1928), it expected the treaties to ensure regional stability. Latin American states refused to support these peace treaties, however, unless absolute non-intervention explicitly became part of the package (Scarfi 2017, 125). Indeed, they spearheaded yet another attempt to bundle territorial integrity and non-intervention together during the 1930 Hague Conference. Their efforts ended in a deadlock that highlighted the “continental divergence between American and European rules of international law” and “increased the pressure to get rules passed within the Pan-American context” (Becker Lorca 2014, 334).

Crystallization

Crystallization of the norm complex occurred in the late 1920s and early 1930s. Signatories to the 1929 Treaties intended them to “give conventional form to” both the commission (i.e., conciliation) system envisioned in the 1923 Gondra Treaty and “the organization of an arbitral system.” These significant codification efforts omitted any explicit reference to disagreements over territory or intervention. Regional actors therefore still had to integrate the territorial integrity—and more importantly—the absolute non-intervention norms with the emerging “peace machinery.”

To accomplish such a task, Argentinian Foreign Minister Carlos Saavedra Lamas drafted an Anti-War Treaty of Non-Aggression and Conciliation, which became known as the Saavedra Lamas Treaty (SLT). This treaty was the first to combine all three aspects of the norm complex: an admonition against territorial wars, a pledge of strict non-intervention, and robust (peaceful) mechanisms to prevent (or quickly end) militarized violence. Argentina, Brazil, Chile, Mexico, Paraguay, and Uruguay signed the treaty in October 1933—two months before the Montevideo Conference. In so doing, they created a focal point for negotiations as the Conference approached.

Latin American states wanted to secure Washington's support for the SLT, but the United States found two of its articles particularly unpalatable. Article II proposed that states shall not recognize the military occupation of territory—a nod to anti-intervention—while Article III explicitly noted that third-party states to a dispute shall not intervene, either diplomatically or militarily. Both articles merged absolute non-intervention and territorial integrity in ways Washington had historically rejected. Nevertheless, because the Great Depression had weakened the United States (Grandin 2012, 88), Latin American states hoped President Roosevelt would sign. He had, after all, announced a “Good Neighbor” policy in March 1933, though that policy had not yet taken shape.

The Chaco War further strengthened the position of the Latin American republics—and Saavedra Lamas in particular. It constituted a shock (i.e., the politics stream) that reillustrated the policy problem and reinvigorated the pursuit of a solution. To address the war, a United States–led Commission of Neutrals threatened to withdraw diplomatic representation in 1933, unless the belligerents accepted arbitration. Saavedra Lamas responded that: “The Argentine Chancellery will not go along with the Commission of Neutrals in any act which … might approximate an intervention … The adoption of coercive measures can be based only on a Treaty accepted beforehand by the countries to which it is to apply” (cited in Rout 1970, 77–78). The United States withdrew its diplomatic initiative in the face of this opposition, and a decentralized diplomatic effort that bundled territorial integrity and non-intervention began under the auspices of Argentina, Brazil, Chile, and Peru. On the fourth day of the Montevideo Conference, the latter efforts yielded a cease-fire that ended the deadliest phase of this last Latin American war—a diplomatic victory for Saavedra Lamas.

As the Montevideo Conference loomed, Roosevelt acquiesced. He warned his delegates that the SLT was perhaps too broad with respect to dispute settlement, might conflict with Washington's intervention rights embedded in other treaties, and could curtail the United States government's ability to protect nationals abroad. At the same time, he noted that “an expressed willingness … to sign Señor Saavedra Lamas’ Anti-War Treaty might conceivably be of considerable assistance to our delegation” and “strengthen materially the American peace machinery.”6 In the end, Roosevelt authorized the United States to accept the non-intervention clauses in the SLT in exchange for additional Latin American signatures on the 1923, 1928, and 1929 treaties, which the Montevideo “peace commission” subsequently promoted.7

The crystallization stage ended when the SLT emerged in the early 1930s—that is, once actors fully developed, integrated, and formally codified the entirety of the norm complex. As in earlier stages, the norm complex strengthened throughout the entire stage, but not linearly. It instead proceeded in fits and starts, at moments when the problem, solution, and politics streams converged. Latin American states, for example, needed to convince others (e.g., the United States) that lingering major-state intervention constituted part of the policy problem—one that the norm complex addressed. The United States struggled to accept the focal point solution (i.e., the entirety of the norm complex) because of policy inertia; a strict non-intervention pledge seemed inconsistent with its existing treaty commitments and values.

Various shocks (i.e., politics) helped actors surmount such obstacles. The Chaco War, which coincided with the Great Depression and the election of Roosevelt, opened a negotiating “window of opportunity.” Latin American states seized this opportunity to accelerate codification, under the leadership of Saavedra Lamas.8 Their efforts eventually resulted in the SLT, after which the norm cycle passed to the incremental-consolidation stage.

Incremental-Consolidation

The incremental-consolidation stage begins after states agree to the Montevideo Conference's treaties in the early 1930s. Thereafter, the region expected its members to conform to the entirety of the norm complex's behavioral prescriptions (see case studies below), reinforced the norm complex's obligations repeatedly (see Appendix G, online), and did not alter those obligations significantly. It lies beyond the scope of this study to trace the norm complex's incremental-consolidation stage in great detail (1935-present). Nevertheless, Appendix G (online) tracks consolidation from the 1933 Saavedra Lamas Treaty to the 1948 Pact of Bogotá—a period that includes the Charter of the Organization of American States (OAS). It maps the direct lineage that connects the former to the latter. When, therefore, actors cite obligations under the OAS Charter or the Pact of Bogotá as motivations for their behavior—specifically with respect to territorial integrity and non-intervention issues—they effectively reference the crystallized norm complex that these later documents restated.

Evidence of conformity with and reinforcement of the norm complex abound. When, for example, the United States signed and ratified the SLT in 1934, it concurrently repealed the Platt Amendment (that stipulated conditions for United States intervention in Cuba) and withdrew its remaining forces from Caribbean states. Similarly, at the next Inter-American Conference (1936), Latin American states reinforced their commitment to resolve territorial disputes peacefully (e.g., through mediation and conflict prevention), while Washington withdrew its remaining legal reservations to the SLT. For championing this consensus, Saavedra Lamas received the Nobel Peace Prize that same year. The new normative environment, moreover, led to “active satisfaction with the territorial status quo” (Kacowicz 1998, 49), within which Latin American states committed to enforce peace collectively in exchange for a United States promise of non-intervention.

After World War II, the Pact of Bogotá (1948) superseded the treaties of 1923–1936 (see Article LVIII) and harmonized the region's “peace machinery” within the United Nations framework. This admittedly consolidated and incrementally advanced the norm complex; yet, the substance of the “norms of peace” in the Americas (i.e., territorial integrity, non-intervention, and the specific peaceful mechanisms for dispute resolution) changed only marginally after 1933, even as new “norms of security” developed (e.g., on disarmament or common security; Kacowicz 2005). In this sense, “Montevideo [1933] marks a culmination” in the “semi-peripheral transformation of basic rules of international law” (Becker Lorca 2014, 7).

Summary

A norm complex developed in the Americas to address territorial integrity and major-power intervention—two prominent threats to state sovereignty and recurrent causes of regional war—through peaceful conflict management, particularly arbitration (for a richer historical narrative, see also Appendix A, online). That development followed the classic norm cycle but also highlighted the need for (and value of) two theoretical additions to the cycle: a causal mechanism that drives development forward (i.e., shocks that converged the problem, solution, and politics streams, as well as positive feedback loops) and a theoretical transition period from the cascade to consolidation stages (i.e., crystallization in the early 1930s). Using quantitative and qualitative evidence, the remainder of this study examines the norm complex's effects on state behavior. Consistent with our argument, the norm complex increasingly restricted state behavior prior to crystallization; stark changes in behavior then followed crystallization, as incremental-consolidation unfolded.

The Effects of Territorial Integrity in the Americas

As Kacowicz (2005, 72) notes, “two major relevant norms of peace for Latin America [are the] peaceful settlement of disputes and uti possidetis.” Indeed, after the 1847 Lima Conference codified uti possidetis, the Latin American territorial landscape consolidated roughly as we know it today. States continued to fight wars over territory, but mainly because they disagreed about how to distribute sovereign jurisdiction along their mutual borders. These wars faded after the 1889 Pan-American Convention, even as territorial disagreements continued (Frederick et al. 2017). In this sense, the Letica (1932) and Chaco (1932–1935) Wars constitute outliers—ones that encouraged the norm complex's crystallization.

Evidence of a strengthening territorial integrity norm emerges in the late nineteenth century through a more robust commitment to (1) peaceful dispute resolution, particularly arbitration, (2) boundary settlement, and (3) non-forceful territorial transfers. Figure 2 shows arbitration in territorial boundary disputes during the period 1830–2001—for states within (left panel) and outside (right panel) the Americas (Owsiak et al. 2018). Comparing the figure's panels highlights American states’ entrepreneurship. The use of arbitration expands before its formal codification at the Pan-American conferences and blossoms alongside these early codification efforts. Similar patterns do not appear outside the Americas (Simmons 1999; Kacowicz 2005); states within the region use arbitration far more often—and relatedly, at this time—than their counterparts elsewhere.

Number of ongoing territorial boundary arbitrations, 1830–2001. Source:  Owsiak et al. (2018).
Figure 2.

Number of ongoing territorial boundary arbitrations, 1830–2001. Source:  Owsiak et al. (2018).

Arbitration trends do not supply direct evidence of the norm complex's constraints. States can challenge arbitral awards or attempt to reclaim lost territory via force. Figure 3 therefore shifts the focus to territorial claims (left panel; Frederick et al. 2017) and legal border settlement under international law (right panel; Owsiak et al. 2018). It demonstrates that, in the Americas, the territorial integrity norm crystallized in the 1930s and consolidated thereafter. We reach this conclusion through three features that correspond with our broad theoretical argument. First, both panels contain the S-shaped curve that accompanies norm complex development: slow change (i.e., emergence), rapid change (i.e., cascade), and a return to slow change (i.e., consolidation). Second, the rapid change occurs precisely when the norm complex strengthens most. Between 1900 and 1939, states increasingly—and most often voluntarily—removed territorial issues from their foreign policy agendas, and did so via peaceful conflict management, rather than territorial conquest (Owsiak et al. 2018).9 After the 1930s, only a few territorial disagreements resisted settlement (Schenoni et al. 2020), and states stopped raising new ones, causing both curves to flatten again.

Territorial claims and border settlement in the Americas, 1830–2000. Sources:  Frederick et al. (2017); Owsiak et al. (2018).
Figure 3.

Territorial claims and border settlement in the Americas, 1830–2000. Sources:  Frederick et al. (2017); Owsiak et al. (2018).

Finally, the 1930s presented a critical juncture—consistent with the crystallization period. The 1930s contains the greatest reduction in dyads (i.e., state-state pairs) with pending territorial claims—cases in which official representatives of one state diplomatically challenge another state's status quo sovereignty over a specific territory (Frederick et al. 2017). It also contains the largest jump in the percentage of neighboring states that delimit the entirety of their mutual borders via interstate treaty (Owsiak et al. 2018).

This evidence suggests a strengthening territorial integrity norm, but our central claim is that MID frequency and severity will decline precipitously after the norm complex crystallizes. We expect these declines to begin during the cascade stage, as the norm complex gains wider acceptance. A stark behavioral change, however, should appear after crystallization finishes and consolidation begins. This is a critical point, for the norm complex requires states not to end all ongoing territorial disagreements, but rather to address those disagreements non-violently.

Although table 2 earlier demonstrated such a general MID pattern, figure 4, shows that the same pattern relates specifically to territorial disagreements. It charts the proportion of all MIDs in the Americas that involve territorial issues over time (solid line). For comparison, it also tracks the similar proportion of MIDs that are unrelated to any of the norm complex's issues. Prior to 1933 (the vertical reference line), MIDs generally occurred as often over territory as over issues that do not fall under the norm complex's content, with some variation. Starting around the 1930s, the general relationship changes; MIDs over territory become less common than their counterparts, coinciding with the norm complex's crystallization and consolidation.

Proportion of territorial and non-territorial MIDs in the Americas, 1830–2010. Note: Intervention-related MIDs are not charted. See figure 7. For research design decisions, see Appendix F (online). Source:  Maoz et al. (2019).
Figure 4.

Proportion of territorial and non-territorial MIDs in the Americas, 1830–2010. Note: Intervention-related MIDs are not charted. See figure 7. For research design decisions, see Appendix F (online). Source:  Maoz et al. (2019).

The norm complex prevents, but also manages, militarization. We should therefore see a decline in territorial MID severity too. Figure 5 plots the timing and severity of individual, dyadic territorial MIDs (Goertz et al. 2023). A (dotted) ceiling line connects the most severe MIDs over time, supplying a sense of how maximum MID severity evolves. A (solid) local regression line then captures the overall temporal trend (Dul 2016). As the figure demonstrates, territorial MID severity declines in the Americas over time, particularly once the cascade stage begins. Crystallization of the norm complex then pushes the severity ceiling downward starting in the 1930s (vertical reference line). The norm complex, in other words, constrains MID severity, even though states still occasionally threaten, display, and use force to address territorial issues.

The severity of territorial MIDs in the Americas, 1830–2010. Note: Figure created using the R NCA package (Dul 2016). Sources:  Maoz et al. (2019); Goertz et al. (2023).
Figure 5.

The severity of territorial MIDs in the Americas, 1830–2010. Note: Figure created using the R NCA package (Dul 2016). Sources:  Maoz et al. (2019); Goertz et al. (2023).

Episodes after 1933 occasionally generated significant battle deaths, albeit short of scholarly war (i.e., 1,000 + battle deaths). The modal disagreement here involved smaller, more remote geographic areas. Moreover, on the rare occasions when states attempted to acquire territory through force (e.g., Ecuador-Peru), the norm complex restrained conflict severity. Thus, even though several dyads had active territorial claims after 1930—and the norm did not require these to terminate—territorial militarization changed radically. Further support for this claim derives from examining the Ecuador-Peru dispute.

Evidence of the Norm Complex's Constraints

To investigate if and how the territorial integrity norm constrained military behavior after the 1930s, we counterfactually analyze the territorial dispute between Ecuador and Peru (for extended analysis, see Appendix C, online). Actors put a norm under stress when they threaten to violate its prescriptions. If the norm truly constrains behavior, then the potential violator—because of the norm—either constrains its own behavior or the community works to constrain it (or both). These are the “dogs that didn't bark.” When no barking occurs whatsoever (e.g., no regional war after 1933), one ideally identifies the dog most likely to bark and analyzes why it did not. This is the most serious challenge to the norm's strength.

We select the Ecuador-Peru case based on this rationale (see Appendix B, online, for the systemic process through which we selected this case). After the early 1930s, the territorial disagreements between Ecuador and Peru fueled more frequent and severe MIDs than any other in the region (Palmer et al. 2021). These episodes yielded colloquial “wars” (i.e., short of 1,000 battle deaths), as well as skirmishes with few battle deaths (e.g., in 1984). We claim that our norm complex explains why a scholarly war does not occur in this dyad after the early 1930s.

After 1935, the Oriente-Mainas dispute remained the largest territorial disagreement in the Americas—at 125,000 km2 (Huth and Allee 2002). Ecuador and Peru agreed on the validity of uti possidetis in the 1847 Lima Treaty; yet ambiguous Spanish cartography and various failed arbitrations prevented the dispute's resolution (St John 1977). Both Ecuador and Peru signed the 1933 Peace Treaties, and they opened border negotiations in its wake. By 1936, they agreed on a status quo line but suspended further negotiations in 1938.

The open territorial wound fueled the colloquial Zarumilla “war” (1941). This conflict never escalated beyond an estimated 600–800 battle deaths, however, because of the unprecedented quick reaction of Argentina, Brazil, and the United States. The 1933 Peace Treaties, as well as the documents that further consolidated the norm complex (e.g., the 1938 Declaration of Lima), obligated third parties to adopt a “common and solidary [regional] attitude.” It also required that parties use all means at their disposal consistent with international law (i.e., not military intervention) to address territorial disagreements, including a compulsory conciliatory procedure (see the SLT). The region therefore monitored the disagreement closely. When the disputants started staging their armed forces, Argentina, Brazil, and the United States quickly offered their good offices, and when Peru launched an offensive on 23 July, they imposed themselves as mediators. These efforts produced an effective ceasefire on 26 July. Had these states delayed their involvement—even a few days—the incident most likely would have produced many more casualties, thereby escalating to a scholarly war.

Mediation continued during the Third Meeting of Consultation of Ministers of Foreign Affairs in 1942. That meeting further consolidated the norm complex. More importantly here, it also constituted the consultative mechanism within the consolidated norm complex and tied that mechanism specifically to the Ecuador-Peru dispute (1942 Resolution XXI). This had two effects. It obligated the region to address the dispute, and it obligated Ecuador and Peru to attend the discussion. As a result, Ecuador and Peru signed the 1942 Rio Protocol. That Protocol delimited the boundary, encouraged demarcation, and committed Argentina, Brazil, Chile, and the United States (hereinafter, the “guarantors”) to robustly underwrite the Protocol's terms—through the deployment of military observers, the promise of further diplomatic involvement, and the provision of technical demarcation assistance. The guarantors, in other words, accepted a “legal obligation” to secure a stable territorial boundary “entailing both quasi-military and diplomatic functions.” These functions are unique to the Protocol; they exist in no other region (Simmons 1999, 22).

Demarcation began shortly thereafter, but ceased in 1948, when Ecuador discovered a misinterpretation of the Cenepa River's hydrography. On these grounds, Ecuador declared the Rio Protocol void in 1960. It argued that the fundamental circumstances on which it based its acceptance of the Protocol had changed (i.e., the legal principle of rebus sic stantibus) and that it had signed the Protocol under duress. The norm complex, however, undercut both arguments. On the one hand, the Third Meeting of Consultation of Ministers of Foreign Affairs—in which Ecuador and Peru participated—passed a general commitment to the legal principle of pacta sunt servanda (Resolution XXI); signatories could not unilaterally renege on treaties. The territorial integrity norm reinforced this specifically with respect to boundary treaties. On the other hand, the guarantors always based their involvement on their international legal obligations, which included never intervening militarily; without such an intervention, there existed no duress. The guarantors consequently continued to enforce the Protocol's terms and regarded Ecuador's position as “a subversive challenge of the accepted rules of international law, in the name of justice and morality” (Kacowicz 2005, 111). The guarantors’ stubborn, robust involvement, backed by the strength of the norm complex (e.g., using the OAS to engage Ecuador), continued to prevent new crises and to defuse those that flared up in 1981, 1984, and 1991 (see St John 1977).

The strength of the norm complex proved consequential once again during the Cenepa “war” of 1995. In December 1994, the Peruvian military failed to comply with OAS-brokered confidence-building measures, which served as an early warning to the region. Immediately after hostilities broke out on 23 January, Ecuador (re)accepted the validity of the Protocol, and the guarantors both provided their good offices and intensely pressured the disputants to cease. By 17 February, the guarantors had brokered the Itamaraty Peace Declaration, which instituted a ceasefire. By 21 February, they had deployed 112 soldiers throughout the disputed zone to enforce that ceasefire (the Military Observer Mission Ecuador-Peru, or MOMEP). Had the guarantors not intervened as quickly or robustly, the conflict would likely have lasted longer and produced far more battle deaths than the roughly 93–500 that occurred (Gibler 2018). A scholarly war, in other words, would likely have happened. Ecuador and Peru, moreover, then permitted the guarantors to arbitrate the final, binding settlement terms that they accepted—another feature of the norm complex.

The crystallized norm complex repeatedly and drastically constrained dispute severity in this case. Without the treaties from the early 1930s, the guarantors would lack the unique, robust enforcement powers they employed. Without a resilient non-intervention norm, Ecuador would have gained a stronger argument for nullification (i.e., signing under duress). Without a strong territorial integrity norm, Peru and Ecuador would have shown less restraint, and the guarantors’ ability to enforce the 1942 Protocol would have weakened. Without a commitment to resolve territorial disagreements peacefully—and without precisely specified mechanisms—the Inter-American community would have possessed imprecise guidance on whether and how to manage the dispute collectively; Ecuador and Peru may not have accepted the validity of mediation or arbitration; and the guarantors would likely have committed fewer resources to good offices, brokering agreements, assisting with demarcation, observing cease-fires, and deploying a peace operation. In short, the norm complex changed the disputants’ incentives in myriad ways, thereby impeding war.

The Effects of Non-Intervention in the Americas

Non-intervention constitutes an essential part of the Americas’ unique normative culture (Kacowicz 2005, 59); yet previous work underappreciates the positive effect of this norm on regional peace. Figure 6 tracks all instances of overt, major-state intervention in the Americas by decade (i.e., one state deploys troops to another state's sovereign territory; see left panel; Gibler 2018; Congressional Research Service 2020). Because European interventions faded in the late nineteenth century, the precipitous decline in a major-state intervention during the norm complex's cascade and crystallization stages corresponds specifically with a sharp decline in interventions by the United States (see right panel). The starkest drop occure as norm crystallization was completed in the early 1930s.

Overt interventions in the Western hemisphere, 1830–2000. Source: Congressional Research Service (2020); Gibler (2018).
Figure 6.

Overt interventions in the Western hemisphere, 1830–2000. Source: Congressional Research Service (2020); Gibler (2018).

Figure 7 lowers the bar for intervention. It charts the proportion of MIDs in the Americas—threats, displays, and uses of force—that involve major-power intervention over time, alongside the proportion of MIDs that are unrelated to the norm complex's issues. Initially, MIDs that challenge the non-intervention norm generally occur at least as often as MIDs over issues outside the norm complex. In the 1910s, the trend changed. United States interventions in the Caribbean then resulted in lengthier occupations, which deterred European interventions and reduced the opportunity any new great power interventions. After 1933 (the vertical reference line), however, the United States withdrew its last militarily troops from the Caribbean, releasing these constraints. Opportunity for major-power intervention therefore increased—but intervention remained relatively infrequent. The norm complex offers the only compelling explanation.

Proportion of MIDs involving great power intervention in the Americas, 1830–2010. Note: Territory-related MIDs are not charted. See figure 4. Source:  Maoz et al. (2019).
Figure 7.

Proportion of MIDs involving great power intervention in the Americas, 1830–2010. Note: Territory-related MIDs are not charted. See figure 4. Source:  Maoz et al. (2019).

As the crystallization stage neared completion, President Roosevelt announced the Good Neighbor policy in March 1933, signed the SLT, and withdrew United States troops from Nicaragua and Haiti. Thereafter, only three overt interventions by the United States did not result from either a (sub)regional collective security mechanism or an invitation from the sovereign state in question: the Dominican Republic (1965), Grenada (1983), and Panama (1989).10 Washington's behavior in these cases admittedly violated the norm complex's prescriptions; yet, the norm complex responded as it should. In each case, the region collectively condemned the action and pressured the United States to withdraw its forces quickly—before the conflict escalated to scholarly war (see Appendix C, online, for extended analysis).11

Why would a hegemon restrict its right to intervene in its sphere of influence? Our norm complex supplies an answer. After the norm complex crystallized (i.e., institutionalized), United States interests changed. It had strong incentives to adhere to the norm complex's obligations (see also Finnemore 2009) because any violation threatened not only to unleash concerted, regional pressure but also to undermine regional cooperation more broadly. Thus, after 1933, overt military intervention in the region dissipated, and the United States relied primarily on covert operations. The latter—along with the frequent decision to seek regional permission for many covert operations in advance—reduced public scrutiny (Poznansky 2020).12

MIDs that challenge the non-intervention norm also decline in severity after the norm complex crystallizes. Figure 8—whose interpretation follows that of figure 5—considers the severity of overt British, French, Spanish, and United States interventions in the Americas during the period 1816–2010. As the figure demonstrates, fatal MIDs (i.e., severity > 99; see Goertz et al. 2023) occurred frequently before norm crystallization, but rarely thereafter. In the 1930s–1950s, the rare intervention-related MIDs fell within the severity range of 0–50, which corresponds to unreciprocated military actions. The Cold War subsequently provoked more American interventions; nevertheless, with the exception of two outliers (i.e., Grenada 1983 and Panama 1989, both still short of the scholarly war threshold), the maximum severity of such MIDs remained in the 50–100 range, which corresponds to nonfatal, reciprocated displays or uses of force.

Severity of MIDs involving great powers, 1830–2010. Sources:  Maoz et al. (2019); Goertz et al. (2023).
Figure 8.

Severity of MIDs involving great powers, 1830–2010. Sources:  Maoz et al. (2019); Goertz et al. (2023).

The quantitative data presented above support the contention that the norm complex constrained conflict behavior in the Americas. As the norm complex cascaded, major-power interventions declined in frequency and severity, and European interventions disappeared. After the norm complex crystallized and consolidation began in the early 1930s, the decline became precipitous. United States policy shifted from overt to covert intervention, so that it would avoid obvious norm violations, and regional actors prevented smaller powers from intervening in one another's affairs, thereby impeding scholarly war onset.13

Evidence of the Norm Complex's Constraints

To understand if and how the non-intervention norm constrained military behavior after the early 1930s, we once again turn to counterfactual analysis—a necessary method when the outcome of primary interest (i.e., war) does not occur. How does the norm complex respond when potential (or actual) interventions place the non-intervention norm under stress? We answer this question by analyzing the United States-Cuba dyad. The United States did not overtly intervene in Cuba during the Cold War.14 Nevertheless, it was highly likely to do so and, through such an intervention, to produce a scholarly war (for the criteria we used to select this case, see Appendix B, online). Cuba carried the threat of extra-regional intervention close to the homeland, hosted a strategic military outpost, involved the expropriation of American assets, featured a government hostile to the United States, and generated frequent and severe militarization throughout the post-crystallization period. These characteristics historically provoked intervention, but the norm complex constrained that option here.15 It caused United States officials to consider the greater Inter-American consequences of intervention and facilitated the rapid organization of a unified Latin American front, which opposed the United States and opened other diplomatic channels. It also strengthened the position of those in Washington who advocated a less aggressive policy.

Among the United States-Cuba MIDs that might provoke an overt intervention, the most intriguing is the Bay of Pigs.16 After the Cuban Revolution, Washington thought Fidel Castro could become an ally. Castro's harsh anti-American rhetoric, however, alongside his expropriations, quickly alienated United States officials. By late 1959, the intelligence and the military pressured President Eisenhower to act. He instituted an arms embargo and covert support for the Cuban exiles trying to create a coherent opposition to Castro. In response, Castro courted Moscow.

The prospect of an extra-regional major-state intervention in the Western Hemisphere historically merited a strong military response by the United States, but the norm complex dictated otherwise. Indeed, Eisenhower understood that military action “required hemispheric support and, in turn, a sustained effort” to avoid antagonizing Latin American states (Sewell 2016, 119). He therefore focused not on overt intervention, but rather on replicating the (covert) Guatemalan coup of 1954, which the region diplomatically justified. The United States called two Inter-American meetings—in Santiago (1959) and San José (1960)—to promote this policy. The Latin American position prevailed, however, with eight states supporting “both Cuba's right to self-determination and the revolution's economic reforms” (Schoultz 2009, 131).

The San José Resolution produced numerous reinforcing effects. It dissuaded Eisenhower from any Cuban intervention, even the covert support for Cuban exiles that the CIA began planning in March 1960. Perhaps more importantly, by emboldening Castro (e.g., in Haiti, Nicaragua, the Dominican Republic, and Panama), it made US policy more hawkish. Hardliners, such as Vice President Nixon, publicly argued that the United States could resolve its issues with Cuba in less than forty-eight hours—a hint at invasion (Schoultz 2009). These hardliners eventually won control of US policy toward Cuba. When the administration subsequently reassigned Ambassador Bonsal to the OAS, the United States Embassy in Cuba became more hardline as well. Amidst the expanding litany of concerns and growing hardline chorus, and after failing at yet another coordinated Inter-American effort, Eisenhower broke diplomatic relations with Cuba.

After President Kennedy's inauguration, Secretary of Defense Robert McNamara outlined a menu of Cuban policy options, from economic pressure to a military invasion. The State Department perceived that an invasion would severely damage Inter-American relations (Schoultz 2009); Washington would face regional uprisings and jeopardize its regional-hegemonic leadership if its forces entered Cuba. As Ambassador John Kenneth Galbraith reminded Kennedy, the United States “got Guatemala at the price of losing all South America” (Schoultz 2009, 153). From the State Department's perspective, the decision about whether (and how) to intervene in Cuba therefore hinged on the malleability of the Inter-American, normative consensus on non-intervention.

Kennedy sent his private assistant, Arthur Schlesinger, to meet key Latin American leaders in February 1961, primarily to determine if the United States could bend the norm. Schlesinger concluded that the region overwhelmingly did not support an overt intervention in Cuba; privately, though, most Latin American states pledged not to obstruct a covert intervention. The norm, in other words, could bend, provided Washington's behavior did not flagrantly and publicly violate it. Kennedy himself further demonstrated the strong, constraining effect of the norm; he moved forward with the Bay of Pigs plan on the exact terms that Schlesinger determined to be regionally acceptable. In fact, “from the early discussions to the bitter end, President Kennedy was adamant that conventional US military forces would not be permitted to engage Castro's army directly” (emphasis added; Blasier 1985, 202).

After the Bay of Pigs fiasco, the security and intelligence communities pressed Kennedy to intervene overtly. Instead, he again turned to Inter-American diplomacy. Four days after the Bay of Pigs, the State Department initiated consultations for an OAS meeting, framing Cuba as a case of extra-regional (i.e., Soviet) intervention. Regional actors initially rejected the call, but as passions cooled and Cuba built stronger military bonds with the Soviet Union, the OAS convened such a meeting in January 1962, at which it suspended Cuba's membership. Although the position of the United States prevailed, Argentina, Brazil, Chile, and Mexico—the region's diplomatic heavyweights—abstained, signaling that the Inter-American willingness to enforce non-intervention remained strong.

The Bay of Pigs case shows that the crystallized norm complex constrained state behavior. The United States faced myriad conditions that historically motivated and justified overt intervention. It therefore entertained such an intervention here but bridled under the norm's constraints. Until the Soviet Union's involvement, Latin American states repeatedly conveyed intolerance for an overt intervention. The United States consequently adapted. Under two administrations drawn from distinct points along the political-ideological spectrum, it deprioritized overt intervention, despite intense domestic pressure; sought ways to bend, not break, the norm complex; and activated regional diplomatic bodies in lieu of unilateral action. These moves did not occur because the United States lacked the capability to intervene; rather, the United States understood overt intervention to be regionally unpalatable. Had the norm complex not existed, the odds of an overt American intervention in Cuba would likely have increased. That, in turn, would have raised the probability of war significantly.

Conclusion

Interstate (scholarly) war occurs regularly throughout the interstate system. The Americas, however, have not seen such wars in almost a century. How and why did this understudied regional peace emerge? In search of a generalizable explanation, this study augments the classic norm-cycle framework, theoretically adding the concept of a norm complex, a crystallization stage, and a mechanism to explain the cycle's S-shaped curve (i.e., a punctuated equilibrium model of policy change, with path dependence and positive feedback). It then chronicles how the Western Hemisphere developed a norm complex to solve two interrelated, war-prone policy problems. The theory predicts that behavior increasingly conforms to a norm complex's prescriptions as it strengthens, but does so most drastically after crystallization, when constructivist and rational-institutionalist logics both reinforce compliance. A battery of quantitative and qualitative evidence supports our position. War virtually disappeared from the region after the norm complex crystallized in the early 1930s, even though the region continued to experience situations that historically led to war. This occurred both because actors were self-constrained and because the region acted collectively to enforce behavioral expectations.

Any alternative argument must—as our argument can—account for all the empirical evidence we present, including the evolution of peace's S-curve and the importance of the early 1930s. No alternative explanation successfully accomplishes this task (see Appendix D, online, for an expanded treatment of what follows, including alternatives not discussed here). The democratic peace, for example, cannot explain regional conflict trends before the 1980s; few states were democratic prior to that. Moreover, in post-1980s cases where it should operate best (e.g., Ecuador-Peru), democracy promoted escalation—not de-escalation. The “dictatorial peace” suffers a similar fate.

Global norms (e.g., the 1928 Kellogg–Briand Pact) likewise fail. Such an explanation misses not only the Western Hemisphere's significant progress before global norms emerged but also that global norms often copied those in the Western Hemisphere. Finally, some alternatives present variables that weaken as conflict trends change most. Regional trade, for example, contracted as the norm complex crystallized. Similarly, although United States regional hegemony reached its height in the early twentieth century, United States power waned as the norm complex crystallized (Grandin 2012; Butt 2013). The United States was also reluctant, if not opposed, to many regional normative initiatives (Mares 2001). It bargained with Latin America to achieve crystallization—and the Saavedra Lamas Peace.

Although this study presents an analysis of a specific norm complex in Latin America, three features likely to generalize to other regions and norms: the concept of a norm complex, the general theoretical framework (i.e., the integrated norm cycle/punctuated-equilibrium model, as well as the crystallization period), and the theory's main prediction (e.g., that behavior changes most drastically after crystallization). Future research might explore these possibilities in other contexts—with as much detail as we bring to the Americas and with an eye toward the agency, entrepreneurship, and advocacy of actors in the Global South.

Author Biography

Luis L. Schenoni is a Lecturer (Assistant Professor) in International Relations and Director of the Security Studies Programme at University College London (United Kingdom). He is the author of Bringing War Back In (New York: Cambridige University Press, 2024), which explores how war made the state in nineteenth-century Latin America.

Gary Goertz is an independent scholar. He is the author of several books and numerous articles on peace, war, conflict management, methodology, concepts, and research design.

Andrew P. Owsiak is a Professor of International Affairs and Josiah Meigs Distinguished Teaching Professor at the University of Georgia (United States). His research examines the causes and consequences of militarized conflict, as well as its management.

Paul F. Diehl is an independent scholar. He is a former President of the International Studies Association and Peace Science Society (International).

Notes

Author's note: We presented earlier versions of this project at the 2020–2021 Kroc-Kellogg Peace, Conflict, Crime, and Violence Workshop series at the University of Notre Dame, the 2021 Annual Meeting of the American Political Science Association, and the 2022 Annual Meeting of the International Studies Association. We thank participants at these venues, as well as the journal editors and anonymous reviewers, for their helpful feedback. The data underlying this article are available on the ISQ Dataverse at https://dataverse-harvard-edu.libproxy.ucl.ac.uk/dataverse/isq.

Footnotes

1

Gibler (2018) claims that the Soccer War produced 1,000 + battle deaths. Scheina (2003, 306) disagrees (see also Clodfelter 2017). This disagreement likely results from counting non-battle-related, civilian casualties.

2

We discuss alternative explanations in the conclusion and Appendix D (online).

3

If competing solutions exist, one eventually emerges as a focal point.

4

As Koremenos et al. (2001, 762, 767–8) note, the rational-institutional/normative distinction is a false dichotomy. When norms codify, they institutionalize, and that institutionalization follows rational design principles.

5

After the threat of European colonialism receded, Latin American states continued to develop the norm complex to constrain US intervention and to reduce the military and economic costs of war (Simmons 2005; Sicotte and Vizcarra 2009).

6

Instructions to US Delegates to Seventh International Conference of American States, Montevideo, Uruguay, Nov. 10,1933. MS. Department of State, file 710.G/371; For. Rel. vol IV, p. 53.

7

Memorandum by Mr. J. Butler Wright, Montevideo, 1933. MS. Department of State, file 710.G/Economic and Financial Problems/48; For. Rel. vol IV, document 91, p. 178.

8

Events during the cascade and crystallization stages can be consistent with the argument that Latin American states were soft-balancing against the United States (Friedman and Long 2015). Nevertheless, arguments that center on the role of Washington as an adversary fail to account for the norm complex's emergence (during which no hegemon existed and European intervention was the main concern) and self-interested efforts by the United States to advance parts of the norm complex.

9

An analysis of forceful territorial transfers (i.e., conquests and annexations) yields a similar conclusion. See Appendix E (online).

10

The Organization of Eastern Caribbean States requested US intervention in Grenada, making it a more ambiguous case of unilateral intervention.

11

Although the norm complex contributes significantly to the US withdrawals, we do not claim it is the only contributing factor.

12

Regional outrage at discovered (unapproved) covert operations further illustrates the norm complex's power.

13

For an alternative argument, see Butt (2013). The timing of the Leticia and Chaco Wars aligns more closely with our argument.

14

The US intervened in the Dominican Republic (1965), Grenada (1983), and Panama (1989) during the Cold War. The norm complex constrained military behavior during these interventions, thereby impeding war in these cases too. See Appendix C (online).

15

Although the norm complex constrains US behavior significantly in this case, we do not claim that it is the only constraint.

16

The Cuban Missile Crisis perhaps constitutes another “likely” case. Any normative constraint, however, becomes empirically confounded by deterrence after the Soviet Union's involvement (see Appendix C, online).

References

Baumgartner
 
Frank R.
,
Jones
 
Bryan D.
.
2009
.
Agendas and Instability in American Politics
, 2nd ed.  
Chicago
:
University of Chicago Press
.

Becker Lorca
 
Arnulf.
 
2014
.
Mestizo International Law: A Global Intellectual History, 1842–1933
.
Cambridge
:
Cambridge University Press
.

Blasier
 
Cole.
 
1985
.
The Hovering Giant
.
Pittsburgh, PA
:
University of Pittsburgh Press
.

Butt
 
Ahsan I.
 
2013
. “
Anarchy and Hierarchy in International Relations: Examining South America's War-Prone Decade, 1932–41
.”
International Organization
.
67
(
3
):
575
607
.

Centeno
 
Miguel A.
 
2002
.
Blood and Debt
.
University Park, TX
:
Pennsylvania State University Press
.

Clodfelter
 
Michael.
 
2017
.
Warfare and Armed Conflicts
.
Jefferson, NC: McFarland
.

Congressional Research Service
.
2020
.
“Instances of Use of United States Armed Forces Abroad, 1798–2020.” CRS Report No. R42738. Accessed September 1, 2023. https://crsreports.congress.gov
.

Dul
 
Jan
.
2016
. “
Necessary Condition Analysis (NCA): Logic and Methodology of ‘Necessary but Not Sufficient’ Causality
.”
Organizational Research Methods
.
19
(
1
):
10
52
.

Fenochio
 
Andres.
 
1953
.
Indice por Materias de los Diversos Instrumentos Interamericanos
.
Washington, DC
:
Pan-American Union
.

Finnemore
 
Martha
,
Jurkovich
 
Michelle
.
2014
. “
Getting a Seat at the Table: The Origins of Universal Participation and Modern Multilateral Conference
.”
Global Governance: A Review of Multilateralism and International Organizations
.
20
(
3
):
361
73
.

Finnemore
 
Martha
,
Sikkink
 
Kathryn
.
1998
. “
International Norm Dynamics and Political Change
.”
International Organization
.
52
(
4
):
887
917
.

Finnemore
 
Martha.
 
2009
. “
Legitimacy, Hypocrisy, and the Social Structure of Unipolarity: Why Being a Unipole Isn't All It's Cracked Up to Be
.”
World Politics
.
60
(
1
):
58
85
.

Frederick
 
Bryan
,
Hensel
 
Paul
,
Macaulay
 
Christopher
.
2017
. “
The Issue Correlates of War Territorial Claims Data, 1816–2001
.”
Journal of Peace Research
.
54
(
1
):
99
108
.

Friedman
 
Max Paul
,
Long
 
Tom
.
2015
. “
Soft Balancing in the Americas: Latin American Opposition to U.S. Intervention, 1898–1936
.”
International Security
.
40
(
1
):
120
56
.

Friedman
 
Max Paul
.
2018
. “
Latin American Strategies against US Intervention
.”
In
 
Donald J. Trump y las Relaciones Cuba-Estados Unidos en la Encrucijada
,
edited by
 
Castro Marino
 
Soraya M.
,
Crahan
 
Margaret E.
,
211
26
..
Havana
:
Grupo Editor Orfila Valentini
.

Gibler
 
Douglas
,
Miller
 
Steven V.
,
Little
 
Erin K.
.
2016
. “
An Analysis of the Militarized Interstate Dispute (MID) Dataset, 1816–2001
.”
International Studies Quarterly
.
60
(
4
):
719
30
.

Gibler
 
Douglas.
 
2018
.
International Conflicts 1816-2010
.
New York
:
Rowman & Littlefield
.

Goertz
 
Gary
,
Diehl
 
Paul F.
,
Owsiak
 
Andrew P.
,
Schenoni
 
Luis
.
2023
.
“Tracking the Evolution of Conflict: Barometers for Interstate and Civil Conflict.” Discussion Paper Series 23-004. Washington, DC: United State Institute of Peace
.

Goertz
 
Gary.
 
2003
.
International Norms and Decision Making: A Punctuated Equilibrium Model
.
New York
:
Rowman & Littlefield
.

Grandin
 
Greg.
 
2012
. “
The Liberal Traditions in the Americas: Rights, Sovereignty, and the Origins of Liberal Multilateralism
.”
The American Historical Review
.
117
(
1
):
68
91
.

Hathaway
 
Oona A.
,
Shapiro
 
Scott
.
2017
.
The Internationalists
.
New York
:
Simon & Schuster
.

Holsti
 
Kalevi J.
 
1996
.
The State, War, and the State of War
.
Cambridge
:
Cambridge University Press
.

Huth
 
Paul
,
Allee
 
Todd L.
.
2002
.
The Democratic Peace and Territorial Conflict in the Twentieth Century
.
New York
:
Cambridge University Press
.

Jenne
 
Nicole.
 
2022
. “
The Origin of Regional Ideas: International Law, External Legitimization, and Latin America's legalismo
.”
Journal of the History of International Law/Revue d'histoire du droit international
.
24
(
1
):
500
32
.

Kacowicz
 
Arie.
 
1998
.
Zones of Peace: South America and West Africa in Comparative Perspective
.
New York
:
SUNY Press
.

Kacowicz
 
Arie.
.
2005
.
The Impact of Norms in International Society
.
Notre Dame
:
University of Notre Dame Press
.

Kingdon
 
John W.
 
1984
;
Agendas, Alternatives, and Public Policies
.
Boston, MA
:
Little, Brown & Co
.

Koremenos
 
Barbara
,
Lipson
 
Charles
,
Snidal
 
Duncan
.
2001
. “
The Rational Design of International Institutions
.”
International Organization
.
55
(
4
):
761
99
.

Long
 
Tom
,
Schulz
 
Carsten
.
2022
. “
Compensatory Layering and the Birth of MMIGO in the Americas
.”
International Organization
.
77
(
1
):
1
32
.

Maoz
 
Zeev
,
Johnson
 
Paul L.
,
Kaplan
 
Jasper
,
Ogunkoya
 
Fiona
,
Shreve
 
Aaron P.
.
2019
. “
The Dyadic Militarized Interstate Disputes (MIDs) Dataset Version 3.0: Logics, Characteristics, and Comparisons to Alternative Datasets
.”
Journal of Conflict Resolution
.
63
(
3
):
811
35
.

Mares
 
David.
 
2001
.
Violent Peace
.
New York
:
Columbia University Press
.

Owsiak
 
Andrew P.
,
Cuttner
 
Allison
,
Buck
 
Brent
.
2018
. “
The International Border Agreements Dataset
.”
Conflict Management and Peace Science
.
35
(
5
):
559
76
.

Palmer
 
Glenn
,
McManus
 
Roseanne W.
,
D'Orazio
 
Vito
,
Kenwick
 
Michael R.
,
Karstens
 
Mikaela
,
Bloch
 
Chase
,
Dietrich
 
Nick
,
Kahn
 
Kayla
,
Ritter
 
Kellan
,
Soules
 
Michael J.
.
2021
. “
The MID5 Dataset, 2011–2014: Procedures, Coding Rules, and Description
.”
Conflict Management and Peace Science
.
39
(
4
):
470
82
.

Petersen
 
Mark.
 
2022
.
The Southern Cone and the Origins of Pan-America, 1888–1933
.
Notre Dame
:
University of Notre Dame Press
.

Powell
 
Emilia
,
Wiegand
 
Krista
.
2021
. “
Conflict Management of Territorial and Maritime Disputes
.”
In
 
What Do We Know About War?
. 3rd ed.,
edited by
 
Mitchell
 
Sara
,
Vasquez
 
John
,
191
205
..
Lanham, MD
:
Rowman and Littlefield
.

Poznansky
 
Michael.
 
2020
.
In the Shadow of International Law
.
Oxford
:
Oxford University Press
.

Rout
 
Leslie.
 
1970
.
Politics of the Chaco Peace Conference, 1935–1939
.
Austin
:
University of Texas Press
.

Sarkees
 
Meredith
,
Wayman
 
Frank
.
2010
.
Resort to War, 1816–2007
.
Washington, DC
:
CQ Press
.

Scarfi
 
Juan Pablo
.
2017
.
The Hidden History of International Law in the Americas Empire and Legal Networks
.
New York
:
Oxford University Press
.

Scheina
 
Robert.
 
2003
.
Latin America's Wars
.
Washington, DC
:
Brasseys
.

Schenoni
 
Luis
,
Goertz
 
Gary
,
Diehl
 
Paul F
,
Owsiak
 
Andrew P
.
2020
.
The Evolution and Causes of the Territorial Peace in the Americas
.
International Studies Quarterly
.
64
(
1
):
57
70
.

Schenoni
,
Luis
.
2024
.
Bringing War Back In: Victory, Defeat, and the State in Nineteenth Century Latin America
. New York:
Cambridge University Press
.

Schoultz
 
Lars.
 
2009
.
That Infernal Little Cuban Republic
.
Chapel Hill, NC
:
University of North Carolina Press
.

Sewell
 
Bevan.
 
2016
.
The US and Latin America
,
London
:
Tauris
.

Sexton
 
Jay.
 
2011
.
The Monroe Doctrine
.
New York
:
Hill and Wang
.

Sicotte
 
Richard
,
Vizcarra
 
Catalina
.
2009
. “
War and Foreign Debt Settlement in Early Republican Spanish America
.”
Revista de Historia Económica / Journal of Iberian and Latin American Economic History
.
27
(
2
):
247
89
.

Sikkink
 
Kathryn.
 
2017
.
Evidence for Hope
.
Princeton, NJ
:
Princeton University Press
.

Simmons
 
Beth A.
 
2005
. “
Rules over Real Estate: Trade, Territorial Conflict, and International Borders as Institutions
.”
Journal of Conflict Resolution
.
49
(
6
):
823
48
.

Simmons
 
Beth.
 
1999
. “
Territorial Disputes and Their Resolution: The Case of Ecuador and Peru
.”
USIP Peaceworks
.
27
:
1
20
.

St. John
 
Ronald
.
1977
. “
The Boundary Dispute between Peru and Ecuador
.”
American Journal of International Law
.
71
(
2
):
322
30
.

Stimmer
 
Annette.
 
2019
. “
Beyond Internalization: Alternate Endings of the Norm Life Cycle
.”
International Studies Quarterly
.
63
(
2
):
270
80
.

Wimmer
 
Andreas
,
Min
 
Brian
.
2009
. “
The Location and Purpose of Wars around the World: A New Global Dataset, 1816–2001
.”
International Interactions
.
35
(
4
):
390
417
.

This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic-oup-com.libproxy.ucl.ac.uk/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Supplementary data