Citation
The new politics of cooperative security in the Americas

Material Information

Title:
The new politics of cooperative security in the Americas
Creator:
Fullerton, William B
Publication Date:
Language:
English
Physical Description:
iv, 363 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Armed forces ( jstor )
Civilian personnel ( jstor )
Countries ( jstor )
Democracy ( jstor )
Hemispheres ( jstor )
Military operations ( jstor )
Military training ( jstor )
Peace ( jstor )
Treaties ( jstor )
War ( jstor )
Dissertations, Academic -- Political Science -- UF ( lcsh )
National security ( fast )
Political Science thesis, Ph. D ( lcsh )
Political science ( fast )
Politics and government -- Latin America -- 1980- ( lcsh )
Latin America ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1999.
Bibliography:
Includes bibliographical references (leaves 338-362).
Additional Physical Form:
Also available online.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by William B. Fullerton.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021557778 ( ALEPH )
43742035 ( OCLC )

Downloads

This item has the following downloads:


Full Text













THE NEW POLITICS OF COOPERATIVE SECURITY IN THE AMERICAS








BY

WILLIAM B. FULLERTON








A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENT OF THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA


1999















TABLE OF CONTENTS
page


A B STR A C T ..... ............................................................................................................ iv

CHAPTERS

1. COOPERATIVE SECURITY IN THE AMERICAS ...............................................1

Principal Them es ....................................................................................................... 1
Recent Trends in Hemispheric Security ............................................................... 11
The New Politics of Cooperative Security. ...........................................................20

2. THEORY ..........................................................................................................30

Introduction ....................................................... ................................................30
Cooperative Security Regimes ................................................................... ...32
Contending Paradigms ...................................................... ......................39
Contending Hypotheses.....................................................................................74
O perationalization ....................................................................... ......................87

3. ASYMMETRY AND SECURITY............................. ................ 94

Introduction ....................................................... ................................................94
The Role of Power in Hemispheric Relations .........................................96
Balancing at the Subregional Level............................................. 125
C onclusion ............................................................................................................. 129

4. HEMISPHERIC COOPERATIVE SECURITY INSTITUTIONS ................... 134

Introduction ........................................ 134
Institutional Evolution.................................................. .........................136
R egim e R ules ........................................................................................................ 158
Regime Effectiveness ........................................ 197

5. SUBREGIONAL COOPERATIVE SECURITY INSTITUTIONS..................218

Introduction ................................................. ............................................... 218
Southern Cone ............................................................................................... 220
Central America.....................................................................................................258


ii









A ndean Region.....................................................................................................290
Conclusions ...........................................................................................................307

6. CON CLU SION ..................................................................................................... 311

Review of the Evidence......................................................................................... 311
Contribution to Theory .......................................................................................... 320
Lim its of System -level Theories .................................. .............. 325
A M odest Proposal ................................................ ..........................................332

BIBLIO G RA PH Y ....................... ..................................................................... 338

BIO G RA PH ICA L SK ETCH ...............................................................................363






































Ill














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor Of Philosophy

THE NEW POLITICS OF COOPERATIVE SECURITY IN THE AMERICAS

By

William B. Fullerton

December 1999


Chairman: Terry M. McCoy
Major Department: Political Science

Hypotheses derived from institutionalism are tested against the pattern of

cooperative security observed in Western Hemisphere, especially since the end of the Cold War. The study addresses the hemispheric level and in three subregions: the Southern Cone, the Andean Region, and Central America. Findings confirm that security institutions have an independent effect in the areas of transparency in military matters, reductions in defense postures, and burden sharing. State power and interest play an important role, but power based theories alone are inadequate. At both the hemispheric and subregional levels, the processes of institutional politics produce rules that validated institutional hypotheses. The pattern of security behavior that emerges is generally rule based. The study makes three contributions to theory: (1) improved understanding of security politics in asymmetric systems; (2) the impact of relative gains on cooperation; and (3) the relationship of transparency, democracy and cooperative security.

iv














CHAPTER 1
COOPERATIVE SECURITY IN THE AMERICAS Principal Themes

This is a study of the patterns of security cooperation and competition in Latin America since the end of the Cold War. Its focus is on changes in security behavior in the 1990s, or the "process of moving beyond praetorianism and the doctrines of counterinsurgency" (Child, 1996, p. 11), which have characterized the hemisphere's Cold War alliance system for the past half century, to a new mode of conflict management based on security cooperation and military transparency. The theoretical tools used to study this process derive from realist and institutionalist traditions in international relations, the first based on power and the second on international organizations.

Many observers argue that a new security regime is emerging in the Western

Hemisphere, based on an understanding that the new threats to hemispheric security are largely transnational in character and require interstate cooperation. Consensus on how to jointly address these new security challenges remains elusive. But despite the absence of central direction, an astounding array of transparency measures and cooperative endeavors in the security field has arisen to capture the imagination of policy makers throughout the region. The "procedures and mechanisms to ensure a secure, prosperous and democratic region" are in evidence all around us only awaiting final agreement on organizational architecture to "transform disparate intentions into cohesive, functional, and comprehensive security arrangements" (Downes, 1996, p. 8).


1





2


Institutions and new rules regarding the use of military capabilities are emerging, but at the same time interstate rivalries and power politics remain important features of the foreign policies of all states in the region. Security cooperation itself is not new to the hemisphere, but rather a normal tool of foreign policy that occurs within the framework of the shifting informal ententes that characterize many security systems. A critical view of the "new cooperation" might point out that much of the activity is restricted to the conference rooms of international organizations that are far removed from military bases where weapons modernization is ongoing and from borders that remained armed and mined. The sources of conflict have not disappeared.1 They have endured despite previous attempts to engineer solutions that would replace power as the ultimate arbiter in the international relations of the region.

What theoretical optic is available to sort through these conflicting tendencies? The Latin American experience has been hard on theories. The field of political economy, for instance, has seen the rise and fall and occasionally the rise again of theoretical approaches as disparate as modernization theory, hegemonic stability theory, Marxism, dependency theory, and bureaucratic authoritarianism. Explanations for regime change have covered the theoretical waterfront as well. Consensus explanations of macro-social change have been elusive.




1 For instance, border disputes, which have been the trigger for most militarized disputes in the past two decades, are still an important source of conflict (Mares, 1997). Downes (1998) cites the Secretary-General of the Organization of American States as stating in 1995 that thirty-one border disputes were still active. For a listing of the more than one dozen serious border disputes, see Rojas (1997).





3


The field of security studies is divided as well. Realism remains the dominant paradigm for understanding patterns of security behavior in this hemisphere, relying on the concepts of power, national interest, and strategic balancing. The contender is institutionalism, which posits that institutions can facilitate communication, forge cooperative commitments, and encourage rule-based behaviors that go well beyond mere balancing.

The basic research design borrows from Grieco (1988a), who develops a set of contending hypotheses, derived from realism and neoliberal institutionalism, which look both at the emergence of security institutions and the security behavior predicted by each theory. Both theories expect states to act in their own best interest. But, can states cooperate to reap the absolute gains implied by lower levels of conflict, reduced defense expenditures, and burden sharing to address transnational threats? The realist approach ranks the likelihood of significant security cooperation as very low, since states are more concerned with relative than absolute gains and, in any case, lack credible assurance that neighbors will respect rules restricting the use of force. Institutionalism offers more hope

- the dilemmas of distributive gains and defection can be at least partially ameliorated through international institutions.

With hypotheses derived from these two approaches, I examine security

cooperation separately at the hemispheric and subregional levels, looking at the evolution of security institutions and observable patterns of security behavior relevant to those institutions. Comparison of the empirical record with explanations offered by systemlevel theories based on realism and institutionalism shows the latter to offer a more reliable framework for understanding security behavior in the region.






4


Contrary to Grieco, I argue that institutions do matter, although not always in

ways that promote cooperation. When interests are largely compatible and power evenly distributed, institutions can be effective; institutional mechanisms facilitate cooperation in a wide variety of ways. However, in the absence of clear consensus, especially under conditions of asymmetry, institutions may not foster cooperative behavior. That is not to say they are unimportant. Indeed, it is in these situations that institutional processes can have their most significant impact over power and interests.

As we shall see, the institutional politics of hemispheric security institutions can work against the interests of the most powerful states, particularly the United States.2 While military balancing against the United States may not be an option, smaller states can develop a diplomatic line of resistance that rallies opposition and prevents U.S. cooptation of institutional tools. This "institutional balancing" is most likely when the security agenda is complex and issue linkages are available conditions characterize the 1990s as a result of the processes of globalization. A divided security agenda emerges that stunts the growth of security institutions at the hemispheric level. On the other hand, in subregional areas, where mutual interests are more easily identified, a rough balance of





2 The power disparity between the United States and the rest of Latin America is so great that comparisons are not always meaningful. At the beginning of the decade, the United States had more men and women in uniform and spent about twenty times more on defense than all of Latin America. Defense spending accounted for a larger percentage of the U.S. economy (which was roughly three times that of Latin America) than any other country in the hemisphere (with the possible exception of Cuba). Similar comparisons of equipment, organization, technology, and training levels deepen the divide. See Poitras (1990).





5


power obtains between partners and the number of players is low, the same trends that make security cooperation problematic at the hemispheric level will instead promote it.

These phenomena are best explained through institutionalist analysis. While institutionalized cooperation based on mutual interest can sometimes be explained by realism "defensive positionalists" argue that states may rationally adopt institutional tools to facilitate cooperation under some circumstances (Jervis, 1986) realism cannot account for institutional outcomes that contradict the balance of power and interests, particularly in the area of security.

But as with all systemic models, there are gaps in both the realist and neoliberal explanations of the security behavior examined in this study. In the concluding chapter, I suggest that those gaps are best corrected by tools drawn from subsystemic institutionalism, which link external security behavior to the character of civil-military relations within individual countries. Interstate behavior may well be the outcome of a two-level game, in which both states and domestic actors compete and cooperate to realize their interests.

The scope of the study encompasses the Inter-American Security System (IASS), which consists of hemispheric security organizations (the Organization of American States, the Inter-American Defense Board and College, the presidential summits and defense ministerials), security-related legal instruments (treaties, conventions and declarations), and institutionalized rules pertaining to the defense forces of the region






6


(transparency requirements, cooperative arrangements and arms control).3 At the heart of the IASS lie two permanent organizations: the Organization of American States (founded in 1948) and the Inter-American Defense Board (founded in 1942). The OAS is the principal political organ of the system, with general oversight of conflict mediation and collective security activities. The IADB for its part is charged with the military-technical aspects of hemispheric security, to include defense planning, training, and standardization. The organizational structure of the IASS is presented in Figure 1-1, for reference at the end of this chapter.

The tension between these two organizations over the past half-century reflects uncertainty about the appropriate role of the military in hemispheric security institutions, the debate between broad and narrow definitions of security and the larger issue of civilmilitary relations in the hemisphere. These themes, which we will revisit later in the study, combine with the asymmetrical distribution of power in the IASS to make organization of hemispheric defense particularly problematic. Efforts to organize hemispheric security are almost inevitably associated with the United States, sparking concern among Latin American states that a centralized system will front for the U.S. security agenda.





3 Grossi (1997, pp. 242-3) sees the IASS as including the "interlinked treaties, agreements and institutions with specific responsibilities on security," including specific instruments like the Rio Treaty and technical bodies such as the Inter American Defense Board. He characterizes the IASS as a "collective security arrangement" with a degree of obligation below that of a formal alliance. Also, see Council on Foreign Relations (1962).





7


Geographically, the hemispheric case study covers the member states of the

Organization of American States, which includes all independent states of the hemisphere minus Cuba. The subregional case studies focus on three subregions that share strong common traits that permit identification as regional sub-systems: the Southern Cone, Central America, and the Andean Ridge. Organizations, legal instruments and security rules particular to these subregions are analyzed in turn. While a part of the hemispheric case study, the English speaking Caribbean is not addressed as a separate subregion because of its markedly different history, language, culture and political tradition, as well as significant differences in its security environment. Only a handful of Caribbean states are signatories to the Rio Treaty and other traditional security arrangements of the region. Few are members of the Inter American Defense Board (IADB). Only recently have the Caribbean states become members of OAS. The subregion is important and merits study, especially in light of the significant security cooperation present there, but cannot be easily compared to the other subregions.4

The time frame is post-1990, a start point that coincides roughly with the end of the Cold War, the beginning of recent efforts to integrate trade and the end of authoritarianism in the region. A shift in U.S. policy also became evident early in the decade, from a security agenda focused on the Soviet and Cuban threats to a new set of priorities that included controlling proliferation of weapons of mass destruction, supporting new democracies and addressing transnational security threats along with a



4 Ivelaw Griffith's many contributions are important. In addition, see the recent volume on Caribbean security edited by Michael Desch (1998) features pieces by Rosenau, Dominguez, and Griffith, among others.








renewed willingness to work with multilateral institutions. As an historical point of comparison, previous security institutions of the hemisphere are covered in Chapter 3.

The lack of a fixed end date for the study complicates efforts to draw hard conclusions about the patterns of security cooperation and the effect of regional and subregional institutions and rules regarding the use of force. For instance, there is little that can be said about the robustness of the new institutions and rules in the absence of a significant extra-hemispheric challenge or change in power distribution globally or regionally. Such changes are unlikely in the short term. But we can still analyze how institutions emerge, how they reinforce norms and create rules and how they affect behavior patterns.

The topical focus is on the use of the armed forces in the foreign policies of states in the Western Hemisphere. Thus issues of transparency in military matters, reciprocal verification of defense postures, arms restraint and security cooperation (involving militaries), as well as conflict, are within the scope of the study. Security cooperation that might be present between police forces or other branches of the government is only tangentially discussed. Likewise, domestic aspects regarding the role of the armed forces within states are considered only briefly.

This chapter continues with a short discussion of recent developments in security relations in light of the end of the Cold War and the onset of globalization, followed by a more fully developed exposition of the main argument of this investigation. Chapter 2 discusses the theoretical literature in order to develop theories and hypotheses. These hypotheses are then situated in a research design that will test their relative explanatory power. Specific research questions will be developed in later chapters, after the requisite






9


empirical settings are examined. Chapter 2 also develops a working definition of cooperative security, with factors that can be operationalized for the detailed analysis to come. Institutional factors relate to organizational assets (administrative capacity, issue agendas and linkages) and behavioral rules (degree of bindingness and verifiability). Cooperative security behavior is measured in terms of military transparency, reductions in defense postures and combined security operations (burden sharing).

An historic case study the institutional politics of the IASS during WWII and its aftermath is the subject of Chapter 3. A convergence of interests, an uncomplicated security agenda, and the interests of United States and its ability to offer assurances and side benefits to key member states help explain the institutional design of the IASS and the cooperative security behavior characteristic of that epoch. When the context changed in ways that complicated interstate relations, divergent interests led to a process of institutional decay.5 Institutional balancing replaced "institutional bandwagoning."6 This historical review offers a case study of the enduring political processes present in hemispheric security relations, showing when cooperation is likely and when it is likely to fail.



5 This general finding is consistent with Desch (1998), as well as a number of other cooperation "pessimists." But I do not argue that hemispheric institutions are simply a tool of the United States, as Desch does. The bargaining dynamics are substantially more complex.

6 Corrales and Feinberg (1999, p. 8) employ this term, which they credit to
Hurrell. U.S. acts of reassurance helped maintain "institutional bandwagoning" towards a Free Trade Area of the Americas in the early 1990s. But with the loss of fast track authority and a cooling of enthusiasm after the Mexican peso crisis, dynamism has shifted to subregional agreements.





10


The second case study focuses on the main theme of the study hemispheric security cooperation in the 1990s. While the dynamics of institutional balancing and bandwagoning are similar, changes in the international context radically altered some of the underlying conditions for cooperation present in the earlier epoch. Efforts to consolidate a new hemispheric security regime have largely been frustrated, because its construction falls along the fault line that separates U.S.-Latin American priorities for regional security.

Despite this failure, however, cooperative security behavior is evident in the

region, going well beyond that prescribed by institutional rules at the hemispheric level. Furthermore, significant geographical variations emerge, such as the degree of military transparency and burden sharing in the Southern Cone; the depth of defense reductions in Central America; and levels of security cooperation in border areas within the Andean Pact. While some of the cooperative activity present conforms to the dynamics expected by realism, subregional institutions play an important role.

Since the deviations vary by geographical area, we examine subregions in Chapter

5. Subregional comparisons allow a more thorough test of our contending hypotheses and inform our conclusions with a greater degree of confidence. In the case of the Southern Cone and Central America, hemispheric security institutions are replicated and strengthened at the subregional level, while in the Andean Ridge countries, little security institutionalization exists. Furthermore, in all three subregions, the relative importance of the United States differs significantly. Thus the case studies allow a clearer look at the role of both institutions and power and interests. Here, too, institutionalism offers the better explanation, but here, as at the hemispheric level, it is still incomplete.





11


In the concluding chapter, we look at the limitations of the systemic approaches, both at the hemispheric and subregional levels, and consider extensions and alternative explanations. An institutional approach that connects domestic politics with hemispheric security institutions looks promising, although its explanatory power may be limited to special circumstances. We explore the approach and suggest a research design for further investigation. The chapter concludes with a modest policy recommendation.

Recent Trends in Hemispheric Security

Is the emergence of security transparency and cooperation a significant break from past interstate behavior in the hemisphere? The level and frequency of interstate conflict in the hemisphere has historically been low when compared to other regions. By most measures defense expenditures, size of armed forces, numbers of conflicts, battlefield deaths Latin American is the least militarized major region in the world.7 Security cooperation has not been uncommon and third party mediation has sometimes helped limit conflict that does arise. In this environment, slight increases in security transparency and cooperation may not be considered remarkable, especially in light of integrative processes currently ongoing in the political and economic arenas.

This disclaimer notwithstanding, the use and threat of force in Latin American foreign policies has been a frequent feature of the almost two centuries of independence



7 This characterization of Latin America has broad support. See Dominguez
(1998), Poitras (1990), Kelly (1997), Van Klaveren (1992), and Ramsey (1999). Desch (1998, 256) notes that Singer and Small found that the Western Hemisphere had the lowest number of wars per year (0.11) and the lowest battle deaths per year (3,100). But it does not follow that the region is free of security competition. In the early 1980s, the region accounted for almost 20% of the Third World's total arms expenditures (Varas, 1985, p. 38).





12


for the region. Mares counts in the twentieth century alone more than two hundred cases where states "either threatened or used military force or were the subject of such threats or force" (Mares, 1997, p. 195). While interventions involving the United States account for a significant portion of this number, most involve militarized disputes between neighboring states of Latin America that bear the same characteristics of conflict everywhere.

The twentieth century disputes that Mares (1997) cites are taken largely from the Militarized Interstate Disputes (MID) database. While many fall short of war, as defined by the standard 1,000 battlefield deaths, at least eight conflicts qualify by that criteria: the Second Central American War, 1906; Third Central American War, 1907; Dominican Republic-Haiti, 1937; Bolivia-Paraguay, 1931-35; Peru-Ecuador, 1939-1941; El Salvador-Honduras, 1969; Argentina-Great Britain, 1982; and Peru-Ecuador, 1995. Other disputes which were not counted came very close to the cutoff criteria. The PeruColombia conflict in Leticia in 1932, for instance, resulted in 852 deaths.

Against this historical backdrop of violence and mistrust, developments in the

field of security cooperation in the 1990s do indeed present an empirical contrast. In the Southern Cone, Argentina and Chile have participated in joint exercises involving their naval and air forces deployed in the Beagle channel, the object of a near war in 1978. The two countries also sent ground forces to work under Brazilian command on the border of Peru and Ecuador as part of a military observer mission. And in 1999, Presidents Carlos Menem and Eduardo Frei, in a reenactment of the 1899 "abrazo" which led to the Pactos de Mayo in 1902, signed a declaration in the austral port of Ushaiua






13


pledging even greater security cooperation and military transparency. 8 That year, the last of two-dozen border disputes between the two countries was resolved.

Argentina recently conducted joint military exercises with Brazil and Uruguay at a 19th century battlefield of what was then Brazil's eastern province of Cisplatine.9 Brazilian, Uruguayan and Chilean officers participate in instruction on peacekeeping activities at an Argentine training center at an old base near the same previously conflicted sector. All countries of the region to include Bolivia and Chile, still with no formal military ties have signed a declaration that establishes the Southern Cone as a United Nations Zone of Peace. The declaration commits its signatories to the removal of mines in the region significant given the recent publicity over the minefields that still exist along the disputed international border of Chile and Bolivia ("Foreign Ministers Spar," 1997).

In the Andean Ridge, Colombia and Venezuela have established a border

commission that coordinates joint military operations. Cooperation between those two countries extends to naval operations, despite an unresolved border dispute in the Gulf of Venezuela. And, most significantly, Peru and Ecuador have ratified a peace agreement




8 Argentina, Brazil, and Chile signed a disarmament agreement in 1902 known as the Pactos de Mayo. Under its terms, Argentina and Chile were compelled to resell ships to foreign arms suppliers (Varas 1985). Somewhat ironically, the original agreement was less an expression of trust and confidence than an act to solidify the balance of power system that emerged in 19th century South America (Burr, 1955).

9 War began in 1825 between Argentina and Brazil over the Cisplatine province (modern day Uruguay). It ended in 1828 under British mediation leading to the independence of Uruguay.






14


that promises a definitive end to a conflict which has engendered two wars and a series of militarized disputes.

In Central America, Honduras and Nicaragua are actively demining their common border, under the auspices of the Organization of American States and the Inter-American Defense Board. El Salvador is participating with the deployment of Salvadoran technical experts to Honduras. El Salvador, Honduras and Nicaragua conduct joint patrols in the Gulf of Fonseca. The armies of those countries, along with Guatemala, have formed a subregional security organization to promote cooperation. And all countries of Central America are signatories to the recently enacted Democratic Security Treaty, which contains many aspects related to cooperative security and military transparency.

Throughout the hemisphere, we find hundreds of examples of close coordination to address common threats, unprecedented transparency in military matters, arms control agreements, and significant changes in force structure, that demonstrate a new spirit of cooperation in the area of security. The incidence of militarized conflict has dropped to near zero.10

The hemisphere has seen periods of institutionalized security cooperation in the past, generally promoted by the United States in the face of a systemic threat, but cooperation in the 1990s is occurring in an environment that lacks a consensus threat. Furthermore, patterns of cooperative behavior reach across historic rivalries. The United



10 Wallensteen and Sollenberg (1998) compare incidence of Latin American
conflict from 1989-97 with other major regions. They report eight militarized disputes in 1989 (out of a global total of 47), and only 2 (of 33) in 1997 (internal conflicts in Colombia and Peru).





15


States is uninvolved or irrelevant to much of this activity. This empirical puzzle merits serious attention. What is the cause of this palpable change in the tenor of interstate relations in the region?

One possible explanation lies with factors commonly associated with realism.

Consider for a moment what the end of the Cold War has meant from the perspective of U.S. strategic interests. As late as 1987, Soviet assistance to Nicaragua, heavy military involvement in Cuba and extensive contacts in Peru were viewed as serious threats to U.S. interests in the region. Argentina and Brazil, in the late 1980s, avidly pursued nuclear technologies and ballistic missile delivery systems. 11 As late as 1989, Cuba deployed the most powerful military forces in the region after the United States, Canada and Brazil and provided direct support to guerrilla movements in El Salvador and Guatemala. 12 The Salvadoran conflict was still contested at the beginning of the decade.

By the mid to late 1990s, all of these potential threats to U.S. interests had

disappeared. The disintegration of the Soviet Union, the collapse of Cuba's military forces, the end to the Central American civil wars and the abandonment of nuclear and missile technology programs in the Southern Cone left the hemisphere without a strategic







11 Into the 1990s, the Tlatelolco Treaty banning nuclear weapons from Latin America had not been accepted by the four countries most likely to proliferate: Argentina, Brazil, Chile and Cuba (Varas, 1994).

12 Dominguez (1998) notes Cuban support of movements in El Salvador, Guatemala and Chile.






16


threat to the United States, other than the transnational challenges of drug trafficking, organized crime and terrorism.13

These threats represent the "dark underside of globalization and interdependence" (Godson and Williams, 1998, p. 66). The same processes that accelerate the flow of goods, people, ideas and money through the global economy, facilitate criminal transnational networks, mass immigration flows, and volatile capital movements (Mufioz, 1996). To the extent that these threats affect the security of the states, they can be considered with realist tools.

Transnational threats may not be core survival issues for the United States, but

they do have an impact on security. The threats to security caused by cross-border flows of contraband (principally drugs), people (immigration and terrorism) and other phenomena (environmental degradation, natural disasters, etc.) affect the U.S. territory both directly (drugs, illegal immigration and terrorism) and indirectly (by undermining newly consolidated democracies and creating an environment of instability which exacerbates transnational threats). Since unilateral action against such threats is not costeffective, the United States might rationally seek ways to influence behavior through bilateral agreements or the creation of security institutions that foster cooperation and




13 See Perry (1997), Wilhelm (1998). General Wilhelm commanded the U.S. Southern Command in the late 1990s. There are numerous official publications that provide detailed lists of transnational threats of concern to the United States. Often cited are the transnational threats of drugs, terrorism, trafficking in arms, migrant flows, and internal instability. The Department of Defense, "United States Security Strategy for the Americas" (1995), offers such a similar catalog. See Einaudi (1996/7), Lowenthal (1992) and Van Klaveren (1992) for good treatments of the topic. Villagran de Leon (1993) offers a review of OAS activity in these areas.






17


burden sharing in these areas. A pattern of security cooperation consistent with this general optic would largely reflect U.S. interests.

Another possible explanation goes beyond a focus on state power and interest, drawing on the proliferation of new international organizations and the growing transnational linkages characteristic of a globalization of economic, cultural and political issues. The globalization literature emphasizes the important role of new norms14 and institutions in organizing the "turbulence" of world politics (Rosenau, 1990; Vaky, 1993; Mufioz, 1996). Despite the recent bearish turn in the United Nations (after euphoric predictions following the end of the Cold War), the general belief is that institutions can help, especially if carefully designed with incentives that appeal to national interests. These institutions are thought to be easier to create in regional organizations, such as the OAS (Lepgold and Weiss, 1998; Lamborn, 1998).15

Corrales and Feinberg characterize the region in the 1990s as passing through a high point of what they term "hemispherism": "the active attempt by the nations of the hemisphere to redirect their foreign policies in favor of closer and coordinated cooperation with one another" (Corrales and Feinberg, 1999, p. 2). Unprecedented cooperation in the 1990s is based on acceptance of new norms, creation of new


14 For example, some argue that a new norm on democracy and human rights appears to be challenging the old, statist norm of sovereignty and non-intervention (Cronin, 1998).

15 The importance of organizations as ways to institutionalize norms is
emphasized here. Ideas untethered from institutions are unlikely to have lasting impacts. Institutionalist treatments of Kant's "perpetual peace" are not vulnerable to assaults such as Carr's (1964) attack on WWI idealism.






18


institutions, and the inclusion of new items on the inter-American agenda. The U.S decision to work through hemispheric institutions and willingness on the part of Latin American states to engage in "institutionalized bandwagoning" has been key to the success of cooperative initiatives (Corrales and Feinberg, 1999, p. 8).

A renewal process at the OAS 16 was initiated in 1991 through the adoption of OAS Resolution 1080, "The Santiago Commitment to Democracy and the renewal of the Inter-American System."17 The resolution established a "firm resolve to stimulate the process of renewal of the Organization of American States, to make it more effective and useful in the application of its guiding principles and for the attainment of its objectives." While the renewal process is in large part an administrative reform to increase organizational efficiency, responsiveness and transparency of resource allocation, the movement includes rethinking and redefining several of the core functions of the OAS. For instance, economic development has been broadened and redesignated under the rubric of sustainable development and support for democracy has been strengthened by the creation of institutions to monitor, verify and improve governance in newly consolidating democracies.

Corrales and Feinberg claim that a new historical peak of hemispherism was

reached in 1994, at the Miami presidential summit. At that meeting, the presidents of the


16 See OAS (1995a) for a review of that process.

17 OAS Resolutions are listed by resolution number and title when that additional information adds to the argument. OAS conferences are referred to by title or subject matter and date. OAS publications are available through the Office of the Secretary General, Organization of American States, Washington, DC.






19


Americas committed their countries to: "cooperate toward an impressive array of goals in politics (advancement of democracy, good governance), in economics (free trade in the hemisphere, financial coordination), and in security (the fight against the drug trade, the collective defense of democracy)."

As Corrales and Feinberg indicate above, hemispheric security and conflict

resolution have been caught up in the reform movement. As a part of that process the ministers of foreign affairs and heads of delegations, through OAS Resolution 1080, "expressed their decision to start a process of consultation on hemispheric security in light of the new situation in the region and worldwide, from an up-to-date and comprehensive perspective on security and disarmament."

The renewal process initiated change in the IASS in several areas: (1)

reorganization of the institutions that address security; (2) redefinition of the principles that underlie hemispheric security; and (3) an ambitious plan of action that includes concrete measures for adoption by member states that constrain use of force and promote a vigorous defense of democracy.

In 1993, the President of the OAS Committee on Hemispheric Security, Dr. Hernin Patiiio Mayer, presented a new model of hemispheric security "cooperative security" to replace the flawed collective security arrangement established by the Rio Treaty. Under "cooperative security," countries would join together to address potential threats instead of waiting until open conflicts developed. Patio recognized that such a model was based on the premise of shared interests, shared threats and a willingness to cooperate to address these threats, using all elements of national power, to include the armed forces. In his view, the hemisphere was ready (OAS, 1995b).





20

There is substantial evidence that elements of this vision have taken root in the hemisphere. Institutionalized cooperation is emerging at all levels locally, subregionally and regionally. Its constituent parts overlap in places and leave voids in others, and mechanisms of coordination between levels are underdeveloped. Despite these limitations, this network of security institutions is becoming an important factor in security relations between states of the region and may be directly responsible for much of the security cooperation observed in the 1990s. This study is designed to evaluate that challenge to the realist paradigm.

The New Politics of Cooperative Security

Just as regional trade integration has slowed at the hemispheric level, giving way to subregional pacts, security cooperation demonstrates a similar trend. The evidence shows that cooperative security generally coalesces at the subregional level, especially where security can be nested in other cooperative regimes, such as Mercosur (Mercado Comdn Sur) in the Southern Cone and the peace process in Central America. But the security institutions that have emerged at the hemispheric level as a result of this process are far from fulfilling the vision of Patifio Mayer. Understanding of how the renewal process actually unfolded in the IASS requires a theoretical model that can encompass the dynamics of power, interest and institutional behavior.

The U.S.-led intervention in Haiti (1994) and the shock of the Ecuador-Peru

conflict (1995) confronted states in the region with two guideposts to the future: the first, an example of outcomes likely under an effective hemispheric security regime and the second, an example of outcomes likely under a failed regime. Latin American states, for the most part, were not satisfied with either scenario. On the one hand, Latin American





21


countries traditionally see hemispheric defense schemes and U.S. hegemony as "two sides of the same coin ... the power of the United States to dictate Latin America's foreign and domestic affairs" (Poitras, 1990, p. 123). The Haitian intervention, although it followed years of negotiations within the OAS and was based on an emerging norm of international relations (Tes6n, 1996), did nothing to ally Latin American suspicions that a new cooperative spirit in security meant cooperation on U.S. terms (Maingot, 1996). On the other hand, conflict between two democratic states of the hemisphere, Peru and Ecuador, both of which had participated in the development of confidence and security building measures (CSBMs) shown to be effective elsewhere in the region, undermined the new spirit of Latin American security cooperation.18 These seminal events corresponded with a turn away from new initiatives at the hemispheric level and a new focus on subregional security.

At first glance, such a trend seems inconsistent with the integrating processes of globalization. But given the asymmetrical power relations in the hemisphere, it is plausible to argue that globalizing influences in fact accelerated the trend towards subregionalism. Economic globalization, for instance has led not to consolidation of a free trade area of the Americas (still a goal for the year 2005), but rather to new or





18 Much of the work on security institutions in this hemisphere focuses on CSBMs, which are also are central to this study. Fundaci6n Latino-Americana de Ciencias Sociales (FLACSO) of Chile has published two important volumes on the subject, under the editorship of Franscisco Rojas Arevana (1996a, 1996b). Both volumes adopt an institutionalist focus, describing the hemispheric concepts and institutions that deal with security and illustrating their influence in a series of regional case studies.





22


revitalized subregional trading blocks: Mercosur, the Andean Pact, the Caribbean Common Market (CARICOM) and the Central American Common Market (CACM).

In the area of security, subregionalization was driven by a number of related

factors. In the first place, security issues were thought to be less salient with the end of the Cold War. The unifying theme of East-West confrontation had disappeared, given way to a new focus on economic relations. In particular, Latin American security issues were thought to be of marginal importance, given the lack of U.S. strategic concerns in the region. This new "marginalization" opened space for subregional initiatitives (Druetta, Tibiletti, and Donadio, 1991).

In the second place, the "dark side" of globalism, with the emergence of

transnational threats, complicated the regional security agenda significantly, leading to a natural breakout of specific subregions based on particular security challenges (Stanely, 1999). Hence, while the Southern Cone focused on overcoming a rather standard security dilemma in order to move economic integration forward, Central America and the Caribbean were confronted with the transnational challenges of contraband and criminality. In the Andean Ridge, countries faced a witch's brew of tense interstate relations and deeply rooted transnational security threats.

Transnational threats are often considered likely triggers for cooperative security, since solutions require multilateral approaches. But U.S. attempts to include transnational threats on the hemispheric security agenda have been met with resistance. Latin American countries are nervous about a broadly defined security agenda that includes many threats of a non-military nature and are unwilling to allow the United






23


States to dominate a process of"securitization" 19 that defines hemispheric threats (InterAmerican Dialogue, 1997). The asymmetry in power relations driven home by the 1989 invasion of Panama to address the results of a "narco-regime" make hemispheric consensus difficult to arrange around these sorts of issues.

The "internationalization of politics" is another feature of the impact of

globalization (Mufioz, 1996b). Its first order effect the awakening of ethnicities and nationalities in Eastern Europe did not have a major impact in Latin America, where the business of nation building is largely complete. But new norms in the area of human rights and democracy found ready acceptance in the region, which was just completing its transition from authoritarianism as the 1990s began. With democracy as a unifying theme and the absence of "alliance handicaps" stemming from ideological differences (Liska, 1962), the possibilities for hemispheric convergence seemed bright.

But as applied to security relations, the picture is more complex. In the early

1990s, ground work was laid for the "collective defense of democracy": OAS Resolution 1080 adopted in Santiago (1991) and the Washington Protocol amending the Charter strengthened automatic responses within the OAS to ensure a strong reaction in the event that democratic processes were interrupted in a member state. But as demonstrated during the Haitian crisis, the political will to violate sovereignty in the defense of democracy proved impossible to muster (Maingot, 1996). Mexican opposition to U.S. policy in this incident was strong. Mexican policy remains anchored in the principle of



19 This is the term employed by Buzan, Waever, and de Wilde (1998) to explain how an economic, social or cultural issue becomes identified as a threat within a given regional security complex.






24


non-intervention, which also forms the basis of its Cuba policy (GonzAlvez Gdlvez, 1998). In 1999, when the United States pushed for an even stronger mechanism to "defend" democracy, resistance from key Latin American states killed the initiative.20 Sovereignty and non-intervention, as well as democracy, are norms of OAS Charter.21

Thus at the regional level, democracy has not been as significant an integrating factor as some expected, in part because of the presence of marked power asymmetry. But at the subregional and unit levels, the impact of democratization on security policies has not been any easier to characterize. While most analysts agree that democracy has an impact on security, few agree on precisely what that impact is (Hirst, 1996). The consolidation of democracy in Latin America has taken a variety of paths and its foreign policy expression is equally varied.

Perhaps more significant than the simple adoption of democratic forms of

government is the general rise in transparency between countries, based on enhanced information flows, open markets and open political systems. This transparency has its major impact in security relations between neighboring states.

For all of these reasons, conditions for subregional security cooperation were

enhanced in the 1990s. In subregions with substantial economic interaction, shared civil20 At the 1999 OAS General Assembly in Guatemala City, the United States
authored language to supplement Resolution 1080. An automatic OAS response would be initiated upon detection of warning signs that democracy was failing, instead of waiting for an actual disruption.

21 The "guiding principles" of the OAS are numerous. As stated in the Charter, they include respect for international law, human rights, sovereignty, noninterference in domestic matters, rejection of wars of aggression, and a recognition of the importance of social development, justice and education, among others (OAS Charter, Articles 2 and 3).





25


military traditions, and common security threats, significant security cooperation has emerged. Often the trend towards more cooperative security was facilitated by specific institutions that allowed defense policies to gradually adjust from confrontational to cooperative strategies. In many cases, these subregional institutions have filled gaps in the overarching hemispheric model, which has remained incomplete and rift by divergent interests.

In Mercosur, for instance, links between economic and security institutions appear to be mutually reinforcing. This is not to say that economic integration always promotes cooperative security. The Andean Pact, created in 1969, initially triggered a dramatic improvement in relations between Ecuador and Peru, whose leaders signed a number of bilateral agreements for economic cooperation, but militarized clashes between the two countries increased dramatically from 1977 onward (Mares, 1996/7). Colombia and Venezuela engaged in an arms race during the height of the Andean Pact's success.22 El Salvador and Honduras fought a war in 1969 at the peak of Central American integration, over the status of Salvadoran nationals in Honduras who were in a sense the physical embodiment of that economic integration. 23




22 Mares states that "Venezuela had previously reinforced border defenses after Colombian guerrillas crossed. But the 1987 appearance of a Colombian navy vessel in Venezuelan-claimed waters provoked mutual military mobilizations. Similar incidents in the early 1970s also fueled arms acquisitions" (Mares, 1997, p. 213).

23 Perhaps these economic pacts leave behind institutional traces. Van Klaveren (1992) asserts that CACM, almost moribund in the 1980s, nevertheless provided a framework to begin the Esquipulas process that culminated in the Central American peace accords.





26


If the processes of globalization in the 1990s have favored subregional security

integration, what are the prospects for a hemispheric counterpart? Based on the pattern of institutional security politics in asymmetric systems, they are poor. Varas (1992) categorizes U.S. security behavior in the region as varying between "hegemonic control" and "coercive control." These labels exaggerate the influence of the United States, but nevertheless offer a key to understanding the new security politics. What Varas means by "hegemonic control" is a U.S. policy of using regional institutions to give a multilateral flavor to its security policies. When this course of action seems more cost-effective than unilateral action ("coercive control"), the United States attempts to create or remake security institutions that reflect its own interests in the region.

Under the best of circumstances, this is not easy. Because institutions tend to

restore symmetry through common decision-making procedures, opportunities for small countries to "voice" their interests, and facilitation of issue linkage new forms of political behavior are possible: "institutional bandwagoning" (support to U.S. initiatives) and institutional "balancing" (resistance to U.S. initiatives). The United States is forced into an assurance game to maintain support, requiring constant displays of multilateralism (acquiescing to majority decisions, accepting binding obligations, refraining from unilateralism), as well as significant side payments to potential balancers (e.g., Mastanduno security "outlets" and other benefits). This is because the U.S. commitment to multilateralism is rarely credible on its own merits. The "shadow of the future" for Latin American states is the certainty that the United States will defect or change the rules of the game when necessary to secure its own interests.






27


However, as long as there is a mutual interest in the gains to be secured through security cooperation (e.g., defense of the continent against the Axis countries in WWII), institutional support for U.S. initiatives is likely to be forthcoming. Over time, a natural decay may set in, as more issues are brought under the purview of the institutional arrangement, leading to a more complicated agenda that is harder to support and easier to subvert. At some point, a balancer may emerge at different times in U.S.-Latin American relations, that role has been played by Argentina, Brazil and Mexico. Since in an assurance game, cooperation quickly unravels once defection is probable, institutional balancing can quickly develop into significant resistance.

The conditions wrought by globalism in the 1990s are particularly auspicious for institutional balancing. The complicated security agenda makes it much more difficult to define a set of security objectives that are equally important to all member states. This same phenomenon makes it easier to link issues in ways that benefit smaller states. Finally, the expansion of the inter-American system, with the incorporation of Canada and the English speaking Caribbean, has added to the pool of potential balancers and exacerbated the well known cooperation dilemmas facing large groups.

If all of this simply meant that multilateralism is difficult and likely to cause the U.S. to resort to unilateral action or engage in states bilaterally ("divide and conquer" tactics), then the institutional politics described above would be of little import. But in fact, the outcomes that the process generates are important; they have an independent impact on cooperative security behavior and the behavior that results is not always consistent with U.S. priorities.






28


The logic of security politics at the hemispheric level is a special case that applies to institutions that encompass asymmetrical power relations. As such, it can be derived from the more general form of institutional theory. That theory, along with realism, is developed in the next chapter. Both provide a set of hypotheses with which to examine the successful security cooperation at the subregional level, as well as the limited cooperation at the hemispheric level.







29








Q U 0
2 em E < u 0

u I .


T o m
0 0 11


( E



cq
8C 3 < U Fl j r











III






4U.o
000 4)4 E0









E o
aP e











0 2 gZ o 0 z u
E0 5 00

ocl(0 '-"0E















<2
o) u U o
(CS E l I



''' Z', i U
zI rr ra0

2 ,og E~


00L M oU~ E P














CHAPTER 2
THEORY

Introduction

Following a brief introduction, this chapter develops a theory-based description of the components of a cooperative security regime, made up of institutions that facilitate cooperation by providing organizations for negotiation and rules that prescribe specific behaviors. Once that important mark is established, the chapter investigates the realist and institutionalist traditions in some depth, particularly with regard to the obstacles and opportunities these paradigms posit for security cooperation between states. From that examination, contending hypotheses emerge to explain the patterns of security cooperation present in the hemisphere. Finally, the chapter concludes with a discussion of the operationalization of key explanatory factors.

Realism and neoliberal institutionalism both address the problem of security

cooperation and conflict in the international system. The theoretical frameworks share many of the same assumptions of how the world works. Both assume that states are the principal actors in an anarchic global "system." In the absence of world government, states must rely primarily on their own resources to ensure survival. Since the main threat to survival is aggression from other states, international relations, especially in the



1 Mere membership in international organizations does not constitute
participation in a regime because independent decision-making is in no way constrained (Stein, 1990).


30






31

area of security, tend to be conflictual. Cooperation is difficult to achieve, even if all participating states would be better off in absolute terms.

The simple fact that a cooperative strategy exists that might reap absolute benefits for participating states does not mean that states, acting rationally, will adopt such a strategy. The dilemmas of cooperation the collective action (free rider) problem, cheating and relative gains must be overcome for states to reap the absolute benefits of cooperation. Institutionalism admits the possibility that multinational institutions comprised of states and non-governmental organizations can foster cooperation. Realism denies institutions any independent role.

Since the two theories differ strongly on the relative importance of institutions, these offer solid analytical traction for examining cooperative security behavior and the links between that behavior and norms regarding the use of force. However, neither the realist nor institutionalist approaches in their elementary form explain patterns of security competition and cooperation in the Western Hemisphere. Both theories need to be modified to reflect the marked power asymmetry present in the region. Power-based approaches must recognize that the United States, with its long time horizon and generous resource base, will at times bear the costs and promote the agenda of multilateral institutions, even in the face of short-term, tactical reversals. By the same token, institutional approaches must recognize that no multilateral regime can effectively constrain the United States from privileging its global or domestic interests in ways that may offend institutional norms.

From a theoretical standpoint, the net affect of this great power asymmetry is to mute the differences between the two theoretical approaches, making it more difficult to





32


design research questions that ably weigh the available evidence. In cases where institutions appear to succeed, success might in fact be due to institutional effectiveness or U.S. support. On the other hand, the failure of an institution to constrain the United States says little about the impact of institutions on state behavior in less extreme cases of power asymmetry.

One way to escape this complication is to drop down one level of analysis to the subregional level. Within the geographical confines of Central America, the Andean Ridge and the Southern Cone power is distributed more evenly. At this level, sharp distinctions appear in the behavior patterns predicted by realism and institutionalism. This becomes evident in the analysis of the subregional case studies in Chapter 5. At this point however, it is useful to consider what a cooperative security regime might consist of, in order to better focus discussion of the two theoretical paradigms.

Cooperative Security Regimes

The purpose of this section is to describe an ideal type that represents a potential end point for the developing set of security institutions in the region. We begin with a discussion of cooperative security, and then develop the constituent parts of what would make up a cooperative security regime. The end result of this discussion is a set of variables that can be operationalized and used to measure the degree to which such a regime might be evolving in the hemisphere.

To develop the concept of cooperation security, we will rely on the work of Janne Nolan (1994), Antonia and Abram Chayes (1994), Joseph Nye (1991), and Ivo Daadler (1992). According to Nolan et al. (1994), cooperative security is "designed to ensure that





33


organized aggression cannot start or be prosecuted on any large scale." It seeks to accomplish this purpose "through institutionalized consent, rather than through threats of material or physical coercion," an important element being the creation of "collaborative rather than confrontational relationships among national military establishments" (pp. 45).

A cooperative security regime supports these objectives through a set of

institutions, with norms that specify behaviors with regard to use of force.2 Sometimes the norms simply proscribe force (with the caveat that use of force in self-defense is always permitted); sometimes they encourage cooperative security arrangements to address certain shared threats; sometimes they create the expectation of collective responses to aggression.

An important point is that cooperative security requires a proactive attitude on the part of participating states to deal with emerging threats before they become open acts of aggression. Cooperative security is not collective security. Its relationship to collective security is like that of preventive medicine to surgery (Nolan et al., 1994).

As Nolan et al. (1994) point out, a "rich fabric of constraints" has already

appeared in some parts of the world that offers instructive examples of elements that might help construct a cooperative security arrangement:

limits on military operations, such as the various CSBMs in place in Europe and
the Middle East; agreements for averting wars, such as measures covering
accidents, hot lines, and crisis centers between the superpowers; limits on force


2 For Daadler (1992), cooperative security is based on a "set of principles, norms and rules that govern the military dimension of interstate relations" created through an "intensive dialogue and habitual cooperation" (p. 52).





34


size, weapon types, and operational practices, such as those contained in the
Strategic Arms Reduction Talks (START), Conventional Armed Forces in Europe
(CFE), and Intermediate-Range Nuclear Forces (INF) agreements; and the
nuclear, biological, and chemical weapons and ballistic missile nonproliferation regimes ... cooperative verification and transparency measures, such as the data exchanges and on-site inspections required by various arms control agreements,
the UN arms sale registry, and the 1990 Open Skies agreement. (p. 7)

All of these measures have been described as CSBMs.3

In this hemisphere, in the 1990s, there are rules already in place associated with

the composition of forces, such as prohibitions on weapons of mass destruction, arms

control/restraint, ceilings on troop strength and constraints on military exercises and

deployments, and restrictions on certain other military activities. There are also rules

relating to transparency, such as exchange of information about military activities (e.g.,

exercise notification), capabilities (e.g., participation in arms transfer registries), and

doctrine (e.g., publication of so-called white papers). There are mechanisms to promote

coordination in security matters, including meetings between high-level representatives of

the defense ministries, exchanges of personnel and military units, and participation in

conferences, symposiums, etc. Some measures are designed to develop cooperation in

certain joint endeavors. They include measures that promote interoperability, such as



3 It is easy to see that such an expansive catalogue of CSBMs may cause some to decry the vagueness of the concept of CSBMs one author calls them "the fastest growing business of the post-Cold War era" and points out the lack of accepted theory to explain their growth or impact. Without a doubt, the term CSBM has become a watchword for all sorts of new security relations throughout the world and includes procedures that go well beyond the "modest measures" agreed to at the 1975 Helsinki Conference, where the concept was first coined. These observations come from Desjardins (1996), a critic of generalized acceptance of the value of CSBMs. Desjardins' work is an invaluable corrective for tendencies prevalent in the literature to expand the definition of CSBMs and exaggerate their value.






35


development of common military doctrine, establishment of a common military education system, and participation in joint training operations. There are discussions in some subregions regarding procedures to create multinational units, share logistics and purchase common hardware, and adopt unified command structures.

How can we make sense of these disparate elements?

The recent work in the area of cooperative security shows that there are several important features to consider.4 First, a cooperative security regime should have the administrative capacity to facilitate negotiation of rules, track compliance (through data collection), evaluate results (through data analysis) and report to members. Second a cooperative security regime should have binding rules regarding use of force. Use of force may only be employed as a last resort (after exhausting all other means), must be applied multilaterally (except in the case of self-defense), and can only be used in support of multilateral, consensus objectives (Nolan et al., 1994). Clearly, there must be prior agreement on what sorts of crises should invoke a military sanction. Third, rules must be produced through an inclusive decision making procedure. The Chayes (1994) assert that "a regime is inclusive to the extent that the states affected by it have a reasonable opportunity to participate in its processes and it does not unfairly discriminate against any of its members" (p. 72). These two general requirements for rules that they be binding and produced via an inclusive decision making process are important characteristics that should be present regardless of the particular aspect of security cooperation considered.



4 This derives from Nolan (1994), the Chayes (1994) and Daadler (1992).






36


Other characteristics help define specific security behavior. First, capabilities of member states must be transparent. Information must be reliable (verifiable), relevant and timely, providing early warning of defection. Second, military postures should be kept at low levels, with limited armaments capacity and constraints on exercises, ceilings on forces, restrictions on deployments (Daadler, 1992). Third, a cooperative security regime must have operational capacity to respond to crises at varying levels, through good offices, formal mediation and ultimately through collective response and burden sharing agreements. This operational capacity need not include a command element (as is the case with NATO), but some mechanism should be available for organizing a collective response to violations of the regime norms.

These three specific injunctions yield the behavior clusters that form the key components of rule-based security behavior: transparency and verification; reduced defense postures; and burden sharing (collective capacity to sanction offenders).

There are clear absolute gains from cooperation in these three behavior clusters.5 Military openness allows states to make a more informed choice about their security needs. Their calculations can thus avoid a "worst case" analysis requiring them to expend more resources than necessary. Reduced defense postures offers a similar absolute gain. With a lower requirement for defense, states can spend resources on other priorities, such as economic development and social programs. Regional burden sharing takes this one



5 The presence of absolute gains is an important prerequisite for cooperation under the assumptions of institutionalism. See Keohane (1984), Keohane and Martin (1995) and Hasenclever, et al. (1997). Military relations among states must be seen as a problem of "common security" (Nye, 1991, p. 162).





37


step further. If states pool their capabilities or agree to specialize in roles where their individual capabilities are already substantial and devote those capabilities to the region as a whole, all states enjoy the benefits of more efficient allocation of security resources. In addition, there are some regional security problems which may require cooperative approaches for their solution. Problems such as trafficking in contraband, illegal immigration, environmental degradation are either impervious to unilateral action or extremely costly.

This last point brings us back to the beginning. What sorts of threats or acts might trigger multilateral military action under a cooperative security regime? Presumably, an aggressor state that openly attacked a member or members of the regime would invite a military response. But what about the Haitian crisis? Does the disruption of democracy, the violation of human rights or the threat of instability also merit military intervention? What about uncontrolled trade in contraband-trafficking in drugs or small arms, for instance? If a state becomes a regional base for "narco-terrorism," is military intervention then justified? What about use of the armed forces to cooperate in nontraditional roles, such as natural disaster relief, environmental protection, peacekeeping and other humanitarian operations? The short answer to this is that an urgent security threat is whatever the regime members say it is.

In this study, we will focus on two conceptual clusters that can be derived from this discussion: institutions and behaviors. Institutions are "related complexes of rules and norms," which vary in terms of their level of "institutionalization," ranging from





38


informal conventions to formal intergovernmental organizations.6 Their rules "prescribe behavioral roles for actors" in accordance with general norms (Keohane, 1989, p. 163). Together, organizations, norms and rules regarding the use of military force make up what we identify as the cooperative security regime in the hemisphere. Tracing the evolution of these elements and comparing them to underlying power and interests of states in the region especially along the fault line of U.S.-Latin American relations offer a constructive comparison of realism and institutionalism.

I treat institutions as conceptually distinct from observable behaviors in areas of transparency and verification, defense postures and burden sharing.7 Those behaviors demonstrate a pattern of security cooperation that can be compared to that promoted by security institutions. Such an exercise offers a second test of the explanatory power of the contending theoretical paradigms. This research requires, then, data that can support discussion of the evolution of institutions vis-A-vis power and interests, as well as data that substantiates the depth and scope of cooperative security behaviors. These are the principal methodological challenges of the study.

The remainder of this chapter examines both the realist and institutionalist

paradigms in light of their explanatory power regarding cooperative security behavior, first at the hemispheric level and then at the subregional level. The dilemmas of



6 See Keohane (1989) for a longer discussion of typologies and factors related to this discussion. Nye (1991) offers a similar definition: "persistent sets of norms that shape expectations and constrain behavior" (p. 155).

7 Not all forms of institutionalism adopt this perspective, as we shall see when reviewing "reflective" approaches to the study of institutionalism.





39


cooperation, especially in the context of the current theoretical debate over absolute and relative gains, figure largely in this examination.

Contending Paradigms

Realism

Realism, tracing its origins back to Thucydides, has enjoyed a long run as the

principal paradigm for understanding interstate behavior, especially as relates to security. classical realism, which took its modem form under Hans Morganthau (1973) after World War II, adopts the state as its basic unit of analysis and the twin factors of power and national interest as the best explanations for state behavior in the international system.8

The tradition, while rich and varied, coalesces around a set of mainstream

assumptions. Keohane (1989) lists three which together make up what he calls the "hard core" of Classical realism: (1) states are the principal actors in world politics; (2) they act as unitary, rational actors; and (3) seek power and define their interests in terms of power.9 Because power is a relational concept, international politics becomes a "struggle for power" (Morganthau, 1973). States tend not to trust one another, tend to rely on their own capabilities to ensure their survival and strive to maintain their military capabilities




8 Keohane (1986) offers a useful review of the development of this school of thought.

9 Mearsheimer (1994/5) lists five: (1) states operate in an anarchic system; (2) the principal threat to a state is another state; (3) intentions can never be known with certainty; (4) survival is the highest goal of states; and (5) states are rational actors. By rational, we mean that states select strategies based on consistently ordered sets of priorities (Keohane, 1986).






40


relative to other states in the system. The result is constant security competition, whereby one state's efforts to enhance its security degrades the security of others.10

Waltz (1979) refined classical realism with a systemic approach, carefully laying the baseline assumptions of neorealism: States are unitary, rational actors attempting to maximize power in relation to other states in an anarchic, self-help system. His assumption that states are unitary actors meant that detailed analysis of the source of preferences was unnecessary. State interest could be reduced to the need to ensure survival. Thus, as rational actors states will seek strategies to ensure survival in the interstate system. That system is anarchic in the sense that no world government exists to manage security relations between states. States operate in a self-help system meaning they can only rely on themselves to ensure survival. Consequently, states vie inexorably to maintain a position of relative strength vis-A-vis their rivals.

Perhaps because of its privileged position, realism has been subjected to periodic challenges that depart from its basic assumptions and attempt to construct an alternate view of international relations. The principal challengers since the advent of the Cold War have been variations of liberalism, ranging from the functionalism of the 1950s to the regime theory of the 1970s, to what is now known as neoliberal institutionalism. 11

Realism, as we shall see, is not sanguine about the possibility that institutions can change the basic conditions in the international system in ways that promote cooperation.


10 Jervis (1978) found that some cooperation was possible in a defensedominated world. See Glaser (1997) for a recent answer to Jervis' challengers.

11 Keohane (1989) adopts this term, from the "liberal institutionalism" coined by Grieco (1988a).






41


Institutions simply reflect the underlying power and interest of participating states. Institutions are arenas for acting out power relationships (Mearsheimer, 1994/5) or representations of power aggregation in the form of alliances.12

This does not mean that security cooperation never occurs, just that it is closely circumscribed and not likely to result from institutions. To better understand why cooperation under the assumptions of realism is so difficult to achieve, it is necessary to examine more closely the theoretical literature on the standard dilemmas of interstate cooperation.

Olson (1964), and others, laid out a "logic of collective action" to show why

"collective goods" are normally supplied at suboptimal levels. Because collective goods are "public" or non-discriminatory in the sense that no state can be excluded from reaping the benefits, states have powerful incentives to shirk and free-ride.13 States will tend to allow others to bear the costs for the provision of collective goods because the benefits accrue more or less equally to all. Clearly, if each state follows this logic, collective







12 Classical realists such as Aron (1973) and Morganthau (1973) recommended formation of alliances as stabilizing, balance-of-power mechanisms. Alliances are institutions, based on Snyder's (1990) definition: "formal associations of states for the use (or non-use) of military force, intended for security or aggrandizement of members" (p. 104). But latter day realists have disagreed on the impact of alliances, as we will discuss shortly.

13 Olson also listed as a key characteristic 'jointness of supply,' meaning "an individual's consumption does not preclude consumption by others" (Gowa, 1989, p. 1245). Public goods theorists also employ these concepts (Kindleberger, 1973).






42


action will normally not be possible in the international system and as a consequence collective goods will not be supplied.14

Of the three clusters of behavior we are most concerned with military

transparency, defense postures and burden sharing transparency and the enhanced regional security produced by burden sharing most closely approximate collective goods. Information about military activities, once made public, is available to all and its utility does not diminish as the number of states with access to the information increases. Enhanced regional security is more problematic more difficult to organize and much easier to privatize. But in its pure form, regional security also approximates a public good.

Thus both of these "goods" are subject to free rider problems, which means, that absent other factors, we should expect to see transparency and regional security provided at suboptimal levels.

Another dilemma of cooperation is nicely illustrated in the context of the security dilemma described earlier. There are two distinct problems captured by the security dilemma: cheating and relative gains.

Cheating problems are often illustrated by game theoreticians through the

Prisoners' Dilemma. 15 In brief, two prisoners, accused of collaborating in a criminal activity carrying a stiff sentence, are held in separate cells. Their jailers have only


14 See Stein (1990) for a summary of this reasoning and a list of its major contributors.

15 The numerous models used in game theory to describe the strategic behavior of states are presented in Oye (1986).





43


enough information to convict on a less serious charge. Both are offered freedom in exchange for evidence against their partner. If only one provides information, the other will receive a harsh sentence. Absent reliable information and assurances that the partner will remain silent, both prisoners are likely to offer evidence against each other leading to imprisonment. In short, they rationally select strategies that lead to suboptimal outcomes. 16 This problem potentially affects all three of our behavior clusters.

In the case of transparency, states that report information about their military activities may expose critical weaknesses. If potential rivals do not also report their activities, then the reporting state may increase its vulnerability by compliance. Thus states may not provide information or may provide partial or misleading information.

Reduction of defense postures is also affected by the possibility of cheating.

States that do reduce their postures will increase their vulnerability if potential rivals do not reciprocate. Thus states may not reduce their postures, or may offset those reductions by seeking alliances engaging in competitive balancing.

Burden sharing faces a similar dilemma, to the extent that it is based on the

pooling of capabilities and specialization of roles. States that design their forces based on the assumption that others will provide support in certain areas (sea patrols, air defense, etc.) become vulnerable if others do not uphold their commitments. Thus states will tend not to cooperate or may offset vulnerabilities by building military capabilities, thus affecting the second behavior cluster. 17


16 See Oye (1996) for a description of the Prisoners' Dilemma and related games.

17 See Snyder (1990) on burden sharing concerns.






44


The theoretical literature attempts to address the problem of defecting as

illustrated by the Prisoners' Dilemma through a variety of means. To the extent that security cooperation is not a "one-move game," states can signal their cooperative intent by small incremental steps designed to elicit a positive response. If cooperation is not forthcoming, a state can retaliate. Because of the "shadow of the future" gains from future cooperation states are likely to cooperate if they do not discount future gains too heavily and the last move of the game is unknown.

Axelrod (1984) allows for cooperative activity between two partners following such a "tit-for-tat" strategy. If communication is enhanced through iteration, some cooperative behavior could theoretically emerge. But in the security field this would rarely amount to more than small incremental steps related to specific security problems, such as protocols to reduce tension caused by unforecasted troop movements near disputed border areas. These steps would have to be reciprocal and easily reversible (Oye, 1986).

In any case, even if the problem of lack of communication and uncertainty of intent were somehow solved, the security dilemma contains the seeds of a second dilemma to cooperation, the problem of relative gains. While inherent in the security dilemma, the relative gains problem is not obvious in the Prisoners' Dilemma. Even if information and commitments are somehow made reliable, states will gain in different measures from cooperation.

Grieco (1988a) clearly lays out the problem of relative gains from the realist

perspective, his argument rooted in the core assumptions of realism. He recognizes that under some circumstances relative gains matter less and states may be able to cooperate






45


to realize absolute gains. But in the area of security, because of core concerns regarding survival and easy translation of relative gains into military capabilities, this possibility is limited.18

Matthews (1996), drawing heavily on the offense-defense balance developed by Jervis, shows the impact that future, cumulative gains might have on a state's strategy to cooperate in the area of security: "If a relative gain in a current round of interaction creates advantages that allow additional gains in future rounds, relative gains will be more important" (Jervis, 1978, p.114). In an offense-dominated world, first round wins can confer enormous positional advantages. If defense is dominant, recovery is possible and hence cooperation more likely. However, as Mearsheimer (1994/5, p. 23) reminds us, in practice it is very difficult to distinguish offensive weapons from defensive ones. 19

Other criticisms are more general. Snidal (1991) shows that in large groups, relative gains do not offer important obstacles to cooperation, since "each cooperative dyad can achieve relative gains against the other actors" (p. 394). But Snidal's overly stylized model is sharply criticized by Grieco (Snidal, Powell and Grieco, 1993) for its assumption of constant returns from cooperation. There are apparently no empirical examples of such a system (Mearsheimer, 1994/5). In any case, not all "cooperative dyads" are equal, at least with respect to security issues. It matters if a state's neighbor



18 See Grieco, (1988a, 1988b) and Powell (Snidal, Powell and Grieco, 1993) for a discussion of various factors which affect the acuteness of the relative gains problem. Jervis (1999) and Keohane and Martin (1995) note a convergence on the issue of when relative gains will matter.

19 See also Jervis (1999).






46


reneges on an arms control agreement. That a distant partner did not may offer little solace.

In the case of transparency, some states may rely on secrecy more than others. A weak state may wish to present misleading displays of strength, while a strong, aggressive state may wish to hide its strength. A strong, status-quo state, on the other hand, would have no reason, based on relative gains, to fear transparency. In fact, transparency could enhance deterrence in this case. Thus, even in the absence of cooperation, transparent behavior on the part of these states might be observed.

A different argument, based on the nature in which military information is handled in various states, exposes a second relative gains dilemma related to transparency. In some states with open systems and mature civil-military relations, much information about military activities is already public. That is not the case in other states

- in fact military information may not even be available to representatives of the government. Thus, some states do not incur additional risk by endorsing military transparency, while others do. This dilemma is well documented in studies of transparency measures attempted between members of the North Atlantic Treaty Organization (NATO) and the Warsaw Pact.20 In short, states that have much to lose from transparency may refuse to provide information or may provide misleading information intending to dissimulate weaknesses or intent.

With regard to reductions in defense postures, the most obvious relative gains

problem is the impact on arms producers, principally the United States but also including



20 See Desjardins (1996).






47


Brazil, Argentina and Chile (three countries who together account for 90% of intra-Latin America arms transfers). Other relative gains problems associated with defense reductions are present as well. Some types of reduction (mechanized arms, aircraft, etc.) will fall more heavily on states with modernized forces, while others (troop ceilings, limits on maneuvers) may primarily affect states with larger, but less technologically advanced forces. Some states may rely more heavily on defensive systems (such as landmines or anti-aircraft missile systems) than others. Thus reductions in some areas may not be possible, or may be offset by increases in other areas. The net effect would be marginal reductions, at best.

The issue of regional burden sharing raises a different question. To what extent will a state compromise its security autonomy for advantages gained by cooperation, accepting increased vulnerability in exchange for gains associated with efficient allocation of regional resources to address common problems?

Regional burden sharing allows all states to use their defense assets more

efficiently. But some states will benefit more than others by the reduction of regional threats. Clearly a state without a coastline receives no immediate benefit from agreements to share sea-patrolling duties. In cases of more complex, transnational threats, such as illicit trafficking in narcotics, calculating relative gains is much more difficult. How are the benefits of reduced drug trafficking distributed among producer, transit and consumer states?

Another relative gains problem relates to the training advantages of taking part in combined operations. These exercises and operations require intense interactions that provide access to state-of-the art military techniques and technology, perhaps most






48


valuable to less advanced states. These considerations make burden-sharing arrangements difficult to achieve under the assumptions of realism.

Taken together, the three dilemmas to security cooperation discussed here freeriding, cheating and equitable distribution of benefits combine to form a powerful deterrent to cooperative security behavior. Table 2.1 summarizes the problems associated with cooperation.






49


Table 2-1: Dilemmas of Security Cooperation

Security Absolute Collective Good Cheating Relative Gains Behavior Gain Problem and Problem and Problem and Non-Cooperative Non-Cooperative NonBehavior Behavior Cooperative Behavior
Military Better Transparency Increased Some states Openness security approximates a vulnerability if may rely on
calculation collective good. weaknesses are secrecy more avoids worst exposed. than others, to
case analysis Strong incentive hide
to free ride. weaknesses or disguise
States may not aggressive provide intentions. information or
may provide Since partial or transparency misleading may not be information. reciprocated, it is unlikely to
develop.
Reduced More No collective Increased Some types of Defense resources goods problem; vulnerability if reduction may Postures available for gains are private. other states do not benefit some
development reduce defense more than
and other postures. others.
goals
States may not Reductions in reduce postures or some areas will may offset not be possible vulnerabilities by or will be offset seeking alliances in other areas. (balancing).






50


Table 2-1-- continued

Security Absolute Collective Good Cheating Relative Gains Behavior Gain Problem and Problem and Problem and Non-Cooperative Non-Cooperative NonBehavior Behavior Cooperative Behavior
Regional More Regional security Increased Some states
Burden efficient can approximate a vulnerability if may gain more
Sharing allocation of collective good. states to do not from interaction
for security uphold and some
Regional resources commitments. members may
Security gain more from increased
Strong incentive regional to free ride. security or solutions to
States may not transnational cooperate or may problems. offset
vulnerabilities Cooperation in through increase in many areas may defense posture or not be possible. seeking alliances
(balancing.)



Thus, under the assumption of realism, the logic of collective action and the

security dilemma demonstrate that sustained security cooperation is difficult to achieve.

And yet we know that realism admits some forms of security cooperation. To better

understand what cooperative security behavior might emerge under realism, it is

necessary to look at the questions of sensitivity to relative gains and the dynamics of

alliance formation.

Grieco (1988b) develops the concept of sensitivity to relative gains (his "k"

factor), arguing that, among other criteria, states faced with increasing levels of threat, an






51


historical adversary and a relative decline in power will have a high sensitivity to relative gains. Threat, as laid out by Walt (1987), includes the systemic elements of capability, proximity, and state of technology (offense-defense), as well as the classic realist factor of aggressive intent. This last element enriches Waltz's systemic approach with a unitlevel variable, at the cost of some loss of parsimony. Nevertheless, to the extent that the presence of active border disputes can be taken as an indicator of aggressive intent, it is relatively easy to measure and incorporate into our analysis.

Thus, in bounded subregions composed of proximate states with a history of

conflict, enduring border disputes and recent changes in power distribution, relative gains dilemmas should be difficult to surmount. This will be especially so during times of technological and organizational change which favor offensive operations, such as the current "revolution in military affairs," based on unprecedented mobility, precision weaponry and information technologies (Mares, 1996/7; National Defense University, 1995 and 1996). Conversely, in subregions with a pacific history, low threat levels and stable power relations, relative gains sensitivity will be lower and some cooperation may be possible, using standard "cooperation under anarchy" techniques (Oye, 1986).

Grieco (1988a) also argues that high levels of economic interaction in these situations will tend to make relative gains problems more severe, assuming economic advantages can be transformed into military capabilities (Grieco, 1988b). Taken together, these factors help explain patterns of security cooperation and conflict between states.





52


From Walt (1987), based on his theory of alliance formation,21 we would expect to see balancing behavior predominate states will seek partners to rally against states they perceive as most threatening. Balancing patterns should be readily observable in the behavior clusters developed to explain the make-up of cooperative security regimes: transparency/verification, defense postures and burden sharing.

In his empirical work, Walt deals with the broader concept of "alignment," of which alliances are a subset. Alignments need not be formalized (Snyder, 1990). The "tightness" of such arrangements is the subject of much debate, but the presence of a common enemy is probably the most important factor. 22 Of course, high levels of threat may not always induce "external" balancing; arming, or "internal" balancing is also a rational response to threat (Morrow, 1993). In any case, alliances are still subject to the same problems inherent in less formal forms of security cooperation.23





21 Walt, studying alliance formation in the Middle East, found that regional balancing and bandwagoning were not closely related to superpower penetration even during the Cold War. He determined that security behavior can best be explained by the dynamic of competitive balancing. The study of military conflict and cooperation in the Middle East remains attractive to security analysts. For a recent look at regional security cooperation during the early 1990s, see Jentleson and Kage (1998), who find that realist factors best explain the trend.

22 See Christensen and Snyder (1990) and Weitsman (1997) on the "tightness" of alliances.

23 Snyder (1990) describes the alliance "games" as burden sharing (concern that costs are borne equally), abandonment (fears that commitments will not be upheld when a state's security is threatened) and entrapment (fears that binding commitments will draw a state into a conflict it would have rather avoided).





53


Grieco does not believe that relative gains are a significant problem in strongly asymmetric power situations (Hasenclever, et al., 1997). To gain a better idea of what patterns of security cooperation and conflict might be under that condition, we return to Waltz. The main variable left after Waltz's depiction of the international system is the distribution of capabilities within that system. Waltz focused on the different behaviors expected under conditions of bipolarity and multipolarity.24 Bipolarity was deemed more stable, because of the clarity that such a system offers and the subsequent predictability of great power behavior. Interventions by the United States in Central American and the Caribbean during the Cold War are consistent with this theoretical model.

Great powers in a bipolar world also engage in cooperative behavior with smaller states, as great power competition for influence and positional advantage penetrates local conflicts throughout the globe. Cooperative arrangements entail some level of risk, since a prospective partner could defect once conditions change, but great powers can accept risk in their relations with smaller states since local, tactical reversals do not affect the overall balance of power.25 Smaller states do not have this luxury (Waltz, 1979, p. 194), and hence would be unlikely to cooperate with each other in the absence of great power leadership. Hence, the Cold War pattern of security behavior was marked by security


24 Unipolar systems were not specifically addressed by Waltz. But we can infer from his discussions on the disadvantages of world government that he would have predicted them to be unstable because a powerful center invites revolution.

25 The longer time horizon of large states with little direct threat to survival from smaller states is an argument that appears in numerous works. See Stein (1990, Chapter 4, esp. 109-110) for a representative argument.






54


competition between the United States and the Soviet Union and security cooperation within each allied camp.

A strong current in the theoretical literature develops this point further. Olson

(1964) had already shown that under special conditions the provision of collective goods may be possible notwithstanding strong incentives to free ride. If one state, or a small group of states, determined that the provision of such a good provided private benefits that outweighed the costs of provision, then a collective good could be produced. The great powers would bear the costs.

Working from this framework to address problems of international political

economy, Krasner (1976) and Gilpen (1987) accomplished much theory-building in the area of hegemonic stability and produced strong hypotheses and a compelling research agenda that proved very popular for a time. Under hegemonic stability theory, a large power establishes rules for cooperative behavior for its own benefit and bears the costs of maintaining those rules. This approach is consistent with Walt's view that great powers can accept the costs of short-term tactical reversals.

In recent years, this approach has suffered a decline, in part as a result of its own overreaching. As Lake (1993) points out, a great power or small group of great powers may not necessarily choose to exert its influence. In any case, great power behavior can be either benevolent (tolerant of free riding behavior) or coercive (payment extracting). Coercive behavior will depend partly on the resources available for coercion and the likelihood that coercion will be effective.

Even if coercion can be cost effective, however, it does not follow that a hegemon will adopt coercive practices. Mastanduno (1997) demonstrates that a hegemon might






55


choose to provide positive incentives for cooperation even if coercion can be cost effective in the short run. It is possible that a great power such as the United States might choose to underwrite significant security cooperation at this particular juncture in history. Mastanduno argues that the United States, which enjoys privileges of the "unipolar moment," faces strong incentives to reassure the other states in the system of its peaceful intent.

In making this argument, Mastanduno relies on the hypotheses developed by

Stephen Walt to explain alliance behavior. In point of fact, Waltz had little to say about security behavior in a unipolar world, which characterizes to a great extent conditions in the Western Hemisphere during the 1990s. Waltz believes that any "unipolar moment" will be short lived because of the propensity of other states to revolt against overweening power at the center. In any case, the kind of security cooperation that Waltz envisioned was simply a byproduct of superpower competition. Absent that competitive dynamic, cooperation with prospective allies should diminish.

Mastanduno borrows from Walt to argue that the United States would rationally take steps to appear less threatening to other states, in order to preclude their forming a balancing alliance that would act as a counterweight to U.S. power. Thus the United States could be expected to work through multilateral institutions, promote transparency to signal peaceful intent and offer security outlets to potential rivals (burden sharing opportunities such as leading roles in peacekeeping operations). While this hypothesis may not square with current U.S. policy in Central Europe and Russia, it offers a theoretically sound prediction of U.S. behavior to enhance security cooperation that is consistent with realism.






56


A set of security institutions designed by the United States under these conditions might look very much like a cooperative security regime. For instance, the United States can facilitate transparency by providing examples (its own public reporting, for instance) and training ("how-to" instruction on preparation of defense white papers, for instance). The United States can facilitate verification by sponsoring specific mechanisms, such as "open skies" fly-overs, and by participating in third party verification, such as the Military Observer Mission in Ecuador and Peru (MOMEP). In addition, the U.S. traditional guarantee of protection against extra-hemispheric threats may permit lower levels of defense postures.

Finally the predominance of the U.S. military model in the hemisphere means that simply by following U.S. standards (protocols, communications, arms, training, doctrine, equipment) which are readily available and essentially cost-free coordination for regional burden sharing is substantially facilitated. Largely, though not exclusively, it is the exercise of U.S. influence over the militaries of the region and the strong presence of the United States in multilateral institutions that is responsible for the emergence of a common "military" language which promotes interoperability a prerequisite for effective burden sharing and collective security action. In the functional language that Keohane (1984) applied to regimes, the United States would be reducing the "transaction costs" which serve as one barrier to cooperation.26 But such a regime would be based on


26 The existence of a common military framework helps solve some
"coordination" problems (e.g., coordination of ship passage), but is not enough by itself to forge security cooperation. Coordination solves dilemmas of common aversions (Stein, 1990). Cooperation, practiced to realize common interests, requires policy adjustments to accommodate partners (Keohane, 1984).





57


and responsive to U.S. security priorities. Krasner (1991), in his study of power and institutions, provides some examples of how those priorities might be expressed in institutional arrangements: (1) determination of who plays; (2) establishment of the rules of the game; and (3) manipulation of payoffs through tactical linkages.

In sum, realism maintains that cooperation is especially difficult in security

matters, due to the security dilemma, cheating, shirking and free riding, and relative gains considerations. Security cooperation, unless supported by a large power, is only likely as a short-term instrument to balance threats. When the threat subsides, so will the cooperation. While some cooperation is possible through unilateral bargaining strategies of "defensively positioned states" (Grieco, 1988a, p. 507), the degree of cooperation actually taking place will be limited to small, incremental, reversible steps, in issue areas where survival is not at stake and early gains do not translate into large advantages.27

Institutions matter little in this conceptualization. The institutions "governing"

cooperative activities in the hemisphere reflect underlying power realities. They are more likely to be arenas of conflict than instruments to enhance cooperation. States may be involved in "talks," but are more likely to view them as an instrument of power politics



27 See Stein (1990) and Jervis (1999) for a more expansive view of the
possibilities of cooperation under anarchy. In his attempt to bridge the gap between realists and neoliberal institutionalists, Jervis includes many cooperative strategies within what he terms "defensive" realism. Institutions, even effective ones, are also in his view consistent with some forms of realism because they are created when states see them as instruments to reap benefits of cooperation. Stein adopts a similar approach, although he is not sanguine about the possibilities of cooperation in security matters. In any case, I do not employ such a broad version of realism, partly because it makes development of contending hypotheses impractical.





58

than good faith bargaining. In Krasner's view, institutions are "crucial means by which winners win and losers lose" (Moe, as cited in Krasner, 1997, p. 361).

Liberal institutionalism, as we will see, is much more sanguine about the

possibilities of cooperation. Institutions can be designed to overcome incentives to cheat and address distributive gains problems.

Institutionalist Theory

The theoretical framework of institutionalism focuses on how multinational

institutions help states cooperate. The approach has a long pedigree, which derives from the liberal belief that progress is possible in human affairs, partly through the design of social institutions.28 Classical idealists believe that international organizations, norms, and values are important. If institutions are properly constructed, peace is the "natural" state of affairs.

Institutionalism benefited from the work of the functionalists of the 1960s

developed by authors as disparate as Mitrany (1966), Haas (1964), Deutsch (1957) and Pentland (1973) who sought to generalize the experience of European integration, hypothesizing that increased cross-border flows would build a web of cooperation due to spill-over and spin-offs from the core economic cooperation, creating a matrix of peace that war could not sever.




28 1 use "liberalism" here in the sense that Keohane (1989) applies the term: a school of thought that "stresses the role of human-created institutions in affecting how aggregations of individuals make collective decisions (Keohane, 1989, p. 10)."






59


In the 1980s, a particular form of institution international regimes became the subject of most the conceptualizing and theory building associated with the institutionalist framework. Hasenclever, et al. (1997) offer a detailed analysis of the principal concepts employed in institutionalist theory, beginning their conceptual review with the consensus definition of"international regime" proposed in 1982 in a special issue of International Organization devoted to theoretical discussion of the concept. In the introduction to that volume, Stephen Krasner defines regimes as set of "implicit or explicit principles, norms, rules and decision-making procedures around which actors' expectations converge in a given area of international relations." Principles are "beliefs of fact, causation, and rectitude." Essentially, they offer a coherent set of theoretical statements about how the world works in a given area of international relations. Norms are general standards of behavior. Rules are "specific prescriptions or proscriptions for action." Decision-making procedures are "prevailing practices for making and implementing collective choice."

Various regimes have been catalogued in terms of their constituent parts. The interstate system itself can be considered as a regime that operates on the principle of sovereignty, a norm of non-intervention and rules/procedures established in the UN charter (Krasner, 1985). Cohen (1983) outlines an international monetary system that operates on the principle of a stable international currency, the norm that states should receive adequate but not unlimited funds for balance of payment problems and rules/decision making procedures as outlined in the charter of the International Monetary Fund. Keohane, in After Hegemony (1984), defines the international liberal trade regime as one





60


based on a set of neoclassical economic principles that demonstrate that global utility is maximized by the free flow of goods ... a basic norm ...
that tariff and nontariff barriers should be reduced and ultimately
eliminated ... specific rules and decision-making procedures ... spelled
out in the General Agreement on Tariffs and Trade. 29 (p.4)

While most of regime-based analysis belongs to the field of international political economy, the approach has a strong tradition in environmental studies and international law. Theoretical work in the area of security regimes is much less common, partly in recognition of the power of realism to predict interstate security behavior, although Keohane and Martin (1995), among others, express the hope that regime analysis will "gradually invade security studies" (p. 44).30 Security regimes have been "discovered" in the area of strategic nuclear weapons reduction, the "European military order," prevention of nuclear war, and nuclear non-proliferation.31

Efforts to categorize the theoretical insights developed within the institutionalist paradigm are legion. Haggard and Simmons, working specifically with the concept of international regimes, organize variants by their theoretical component. From their



29 The application of institutionalist theory to explain international trade behavior is perhaps more complete than in any other area of interstate relations. The scholarly debate on the topic emerged in the context of the declining U.S. economic hegemony in the 1970s. Institutionalists argued that what they saw to be an enduring practice of free trade despite the U.S. decline was linked to the presence of institutions and norms. Critics disagreed both with the institutionalist characterization of trading practices as largely free and with the claim that such behavior was in any case due to the impact of institutions. See Keohane (1984).

30 Axelrod and Keohane (1986) advocate a "single framework" for militarysecurity and political-economic issues (Axelrod and Keohane, 1986, p. 227). See Jervis (1983) for the classic explanation of "why security is different."

31 See Mueller (1993), Tate (1990) and Brzoska (1992).






61


analysis, four basic forms emerge: structural, which links regime creation and maintenance to hegemonic power (Gilpin, 1987); functional, which claims that regimes play a facilitating role that reduces the transaction costs associated with formulating agreements (Keohane, 1984); game theoretic, where regimes help structure the context in which states make strategic choices in pursuit of their preferences (Stein, 1990); and cognitive, which is strongly related to the conditioning role that norms/ideas have on behavior (Ruggie, 1983 and Haas, 1964).

Institutionalism, as currently employed (Keohane, 1989), contains strong traces of this prior work. Its principal characteristic is its view that institutions have independent causal significance rather than simply reflecting power relationships.32 Thus, hegemonic stability theorists, who see institutions simply as creations of powerful states, are not liberal institutionalists. However, the functional notion that institutions can help states reduce costs associated with forging cooperative agreements, as well as the game theoretic concept that institutions affect "pay-offs" for cooperation and defection, are central to what Keohane (1989, 166) calls the "rationalistic" study of institutionalism, commonly referred to a "neoliberal institutionalism." Cognitive regime theorists have contributed to what Keohane refers to as "reflective" institutionalism, which is distinct from neoliberalism in that it operates below the system level, where unit-level processes (such as the formation of preferences) are important variables.




32 As Jervis (1999) notes, it is not enough to show that cooperation and the
presence of institutions are correlated, as they most certainly are (see work of Russett, Oneal, and Davis, 1998). Institutionalism must demonstrate causality.





62


Neoliberal institutionalism, operating at the system level, accepts the neorealist assertions that states are the most important actors of the international system and that they act rationally in their interactions with each other. However, the assumption that the distribution of capabilities is the most important system-level variable is sharply questioned. Neoliberal institutionalism holds that "deeply embedded institutions are as fundamental to world politics as are the power resources of units" (Keohane, 1989, pp. 89). Institutions change the context in which states operate (a game theoretic notion) by facilitating communications so that agreements can be made (a functional notion), by providing information to reassure states that agreements are being respected, by institutionalizing norms so that expectations converge around rule-based behaviors, and by linking compliance to other areas of concern to states so that the costs of defection are raised.

These institutionalized settings facilitate cooperation in two primary ways. First, they reduce uncertainty by providing information that can be considered in a state's decision making process. Second, they reduce the transaction costs of entering into negotiations by providing an arena for communication.

Institutionalists hold that even in the absence of changes in the international distribution of power or the preferences of actors, information can "profoundly affect international political behavior" (Keohane, 1986, p. 19). In the area of security cooperation, information is especially valuable because it avoids worst-case analysis (Keohane and Martin, 1995; Jervis, 1986). In fact, Keohane (1984) sees the provision of high quality information as a main function of these institutions.






63


If agreements are to be made on the basis of this information, however, its relevance and reliability are critical. Information about the resources and formal negotiating positions of potential partners is not sufficient. What is needed is information about their "internal evaluations of the situation, their intentions, the intensity of their preferences" (Keohane, 1989, p. 118). That information must be verifiable, through formal inspections, visits, exchanges, data cross-checking, or other mechanisms (Nye, 1991; Oye, 1986; Anthony, 1993). Institutions can provide this service by levying reporting requirements on members and establishing rules for verification. Institutions also help verification through "explicit codification of norms," making it easy to evaluate the degree of compliance (Oye, 1986, pp. 16-17). Since these mechanisms taken together greatly facilitate the operation of reciprocity, they are critical to agreements regarding reductions in defense postures and burden sharing.

In addition to providing information, institutions facilitate the making of

agreements in a variety of ways. In the first place, they provide a forum, or "constructed focal point" for negotiation (Keohane and Martin, 1995, p. 45), which increases opportunities to negotiate (Nye,1991). This allows bargaining to proceed in an iterative fashion, promoting the tit-for-tat pattern of cooperation. Furthermore, institutions can enhance communication by standardizing signals through the adoption of standard formats and reporting calendars.

These steps make it easier for states to reach agreements regarding cooperative security. They also create the expectation that states will comply with security rules. This is because institutions formalize standards for behavior and create the means to monitor compliance and evaluate state reputations (Hasenclever, et al., 1997). As long as






64


states "discount the value of agreements on the basis of past performance" (Keohane, 1984, p. 105), a reputation for compliance is valuable due to the prospect future gains.33 The costs of defection are raised because of the existence of institutions, just as the transaction costs of cooperation are lowered.

Institutions that can be effective in this sense must have the administrative capacity to support negotiations and collect, analyze and report data. These are essentially performance criteria, which can be observed and measured in security institutions.

Institutions rarely operate in a vacuum. They may overlap issue areas with other institutions (Keohane, 1986), creating multiple links between states that connect, for instance, rules on cooperative security with regional mechanisms to promote respect for human rights, encourage economic development and strengthen democratic governance. In the broadest sense, the effect of these issue linkages is to create a dense web of constraints that fortifies rules and makes rule-based behavior more likely.

More specifically, linkage is important to institutionalist theory in several

concrete ways. First, linkage allows for more iteration, increasing the opportunities for "tit-for-tat" reciprocation (Oye, 1986). Second, linkage may erode hierarchy, by allowing small states to link security issues to other areas (Keohane and Nye, 1989). Third, linkages increase the benefits of cooperation and the costs of defection. They "allow for




33 Institutions can also reward cooperative behavior through provision of
leadership positions (president of the OAS Committee on Hemispheric Security, for example) and sponsorship of security conferences.






65


more effective retaliation against cheaters and ... create scope for mutually-beneficial exchanges" (Keohane and Martin, 1995, pp. 49-50).

This last point touches on the globalization literature,34 which resurrects some of the early thinking regarding the beneficial impact of trade (Viner, 1944). Oneal et al. (1996) note that Kant, in his treatise Perpetual Peace, asserted that "economic interdependence reinforces constitutional constraints and liberal norms by creating transnational ties that encourage accommodation rather than conflict" (Oneal et al, 1996, p. 12). This proposition is contrary to the realist view that interdependence creates vulnerability and heightens concerns over distribution of gains, thereby increasing the propensity for conflict (Waltz, 1979). Realists see this as especially valid in cases of asymmetrical interdependence, such as Latin America (Hirschman, 1945; Sanderson, 1992). 35

Empirical tests of the "perpetual peace" proposition have failed to sway proponents of either side of the debate, in part due to disagreements over testing







34 Barbieri and Schneider (1999) note that "the term 'globalization' has largely superceded the term 'economic interdependence' to describe the rapidly growing links between nations, economics, and societies" (Barbieri and Schneider, 1999, p. 387). For two recent critiques of the liberal argument that free trade leads to peace, see Rowe (1999) and Ripsman and Blanchard (1996/7).

35 Mares (1997) argues that economic globalization will not change the nature of security relations in the hemisphere: "economic growth ... in the absence of a stable and credible balance of power" is likely to lead to "increased conflict, rather than the first steps toward integration or a pluralistic security community" (Mares, 1997, p. 215).






66


methodology.36 Even if agreement were possible on the methodology, however, results showing simple correlation are not convincing, since causal connections are not clear. Gowa (1989) finds for instance that because of the security externalities associated with open borders, free trade is more likely between allies than adversaries. It her analysis, the security environment determines trade relations, not vice versa.

Even if institutions are able to craft linkages between security and trade that

expand potential gains, the struggle over distributional gains still needs to be addressed. Despite claims that institutionalism has ignored relative gains (Grieco, 1988a, Mearsheimer, 1994/5), Keohane and Martin (1995) assert that situations where relative gains considerations are important offer an especially productive area for institutions:

Liberal theory argues that institutions provide valuable information, and
information about the distribution of gains from cooperation may be
especially valuable if the relative-gains logic is correct. Institutions can facilitate cooperation by helping to settle distributional conflicts and by
assuring states that gains are evenly divided over time, for example by
disclosing information about the military expenditures and capacities of
alliance members. (Keohane and Martin, 1995, pp. 45-46)


Thus, the mechanism of reciprocity, facilitation for negotiation of side agreements, and the presence of linkages all help institutions deal with distributional problems.

Institutions do not always have a strong impact on behavior. According to

Hasenclever, et al. (1997), institutions differ in terms of their robustness (the extent to which institutions resist change when the interests of the most powerful member diverge from the norms) and effectiveness (the extent to which actual behavior can be


36 See the recent debate in Journal of Peace Research (vol. 36, no 4), especially Barbieri and Schneider (1999) and Oneal and Russett (1999). Also, see Morrow (1993).





67


characterized as rule-based). But the important point is that while institutions may reflect the distribution of power and the interests of their members, they have an independent impact on behavior and mute the impact of power over outcomes (Keohane and Nye, 1989). Cooperation between states, acting as unitary, rational actors attempting to maximize interest in an anarchic, self-help system, is possible: (1) because states are interested in attaining benefits of reduced tension and enhanced security cooperation and

(2) because institutions help them overcome the dilemmas of defection free riding and cheating and help them negotiate solutions to the distributive problems that relative gains present.

How might this work in situations of asymmetry? The kind of cooperation

described under realism in which a large power establishes institutions as tools to achieve its own ends is a possibility admitted by neoliberalism. In fact, where mutuality of interests is high, it may be difficult to find significant differences in the manner in which these kinds of institutions evolve. But the politics of security cooperation under asymmetry are likely to be more complex. Keohane and Nye (1989) provide a set of potential outcomes that differ sharply from realism. While these were constructed for a specific context what the authors called "complex interdependence" I argue that the effects they predict are evident in cases where a large state practices a policy of reassurance (as described by Mastanduno) that opens negotiating room for smaller states.

The United States may opt for a multilateral approach as a cost-effective solution that simultaneously reassures its security partners and engenders the cooperation it seeks. However, such a strategy is likely to produce perverse outcomes over time. This is because the institutional forum it promotes restores symmetry by offering smaller states





68


an opportunity to voice preferences, build coalitions and reinforce negotiating positions through linkages.37 Small states can leverage these advantages to engage in balancing behavior that would not be possible in the absence of the institution. Furthermore, in order to build support for its agenda, the United States may need to provide side benefits and security outlets which ultimately strengthen the bargaining position of the smaller states. While an equilibrium may be theoretically possible, a more likely outcome is an eventual retreat from multilateralism, leaving the institution void of resources and effectiveness. Mutuality of interests declines and the shadow of the future shortens as smaller states come to expect a return to unilateralism on the part of the United States.

The determinants of U.S. security behavior are largely beyond the reach of

hemispheric institutions. The primacy of global interests and the impact of the domestic agenda are clearly more important to the formulation of U.S. security policy than regional concerns. When global concerns dominate, support for regional cooperative security and respect for regional rules take a backseat to other strategic priorities. When domestic concerns dominate, U.S. offers of assistance (arms, training, resources, information) can be subject to intrusive end-use agreements which fall outside of the regime. The certification process which oversees provision of counterdrug assistance and the Leahy amendment which makes military assistance contingent on human rights performance are two examples.



37 Jervis (1999) claims that institutions, even if created by the system's most
powerful states, can erode power by giving "voice, legitimacy and forms of influence to weak or new actors" (p. 61). Institutions could restore some symmetry to U.S.-Latin American relations. This finding can also be derived from Krasner (1985), although his argument is a realist one.






69


The ever-present threat of unilateral U.S. action and the knowledge that

institutional constraints are insufficient to prevent such action undermines confidence that other states in the hemisphere have in hemispheric institutions. Eventually they will decay. But along the way, an agenda of security issues and related rules that is substantially different from U.S. power interests is possible.

By contrast, realism rejects the possibility that institutions can have significant effects and denies them any independent impact on state behavior. Even if institutions facilitate agreements and make information available, that does not mean agreements will actually be consummated or respected when interests clash with regime norms. In the ultimate instance, regimes cannot guarantee survival. In a self-help system, what matters is a state's relative position. Created in the interests of their most powerful members, institutions will not constrain behavior and will be abandoned when no longer convenient.

What happens when we move from the system level and look at preference formation within member states? This is the focus of the "reflective" version of institutionalism, reviewed only briefly here, since we will not return to it until the conclusion of the study when we consider the limitations of systemic approaches. Keohane (1989) offers this summary of the reflective approach: "Institutions do not merely reflect the preferences and power of the units constituting them; the institutions themselves shape those preferences and that power. Institutions are therefore constitutive of actors as well as vice versa" (Keohane, 1989, p. 162). "Reflective" approaches encompass "constructivist theory" (Checkel, 1998), "critical theory" (Mearsheimer, 1994/5) and "grotian theory" (Krasner, 1983). These labels in turn lump together a wide






70


variety of approaches. Wendt (1995), for instance, identifies "postmodernists (Ashely, Walker), constructivists (Adler, Kratochwal, Ruggie ... Katzenstein), neo-Marxists (Cox, Gill), feminists (Petersen, Sylvester) and others" (Wendt, 1995, pp. 71-72) as belonging to critical international relations theory. The approaches are united by a belief that structures are sociological constructs that shape actors' identities and interests, rather than just behavior.

Hasenclever, et al. (1997) usefully divide the reflective approach into weak and strong cognitivists. Weak cognitivists believe that institutions can change preferences through learning gleaned from the experience of cooperation. Strong cognitivists believe that the institutions themselves are deeply embedded constructs that form actors' identities.38 Both of these versions of institutionalism delve into the ideas that statesmen use to form policies regarding security competition and cooperation. New ideas can cause normative change and lead to corresponding changes in behavior (Goldstein and Keohane, 1993). Thus, the penetration of new ideas pertaining to appropriate roles for the armed forces under democratic regimes could give rise to strong proscriptions on use of force and restrictions on military activity.39


38 See March and Olson (1998) for a recent review of this approach.

39 This argument is tangentially related to the debate over the effect of democracy on interstate conflict. The observation that democratic dyads tend not to fight one another has been thoroughly studied, although there is still debate as to whether common interests or common polities explain the phenomenon (see Chan, 1984, and Farber and Gowa, 1995, for opposing views). Within the camp of those that believe democracy makes a difference, there is disagreement over whether democratic norms of peaceful resolution or democratic mechanisms of leader accountability explain the phenomenon (Starr, 1997). Recent work on democratization shows that it may positively correlate to conflict (Mansfield and Snyder, 1995).






71


Weak cognitivists hold that institutions contribute to this process both by

providing "opportunities for contacts which may contribute to learning" (Nye, 1991, p. 157) and through dissemination of new information which changes preference formation (Keohane and Martin, 1994/5). Haas (1964) found that successful integration in Europe was partly due to a learning process experienced by technocrats. Success in one area "spilled over" into others as elites expanded the process of economic integration. Adler (1988) looked at the power of ideas and learning by technocratic elites in the area of nuclear power policy in Argentina and Brazil.

Hasenclever, et al. (1997) describe an interesting extension of this idea, based on the work of Putnam and Zurn. Introduction of domestic institutions creates a two-level game actors within the state apparatus bargain with each other and with counterparts outside the state simultaneously. This relates to the "second image" and "second image reversed" logic that frequently appears in the security literature. For instance, Mares and Bernstein (1999) argue that Argentina's high level of participation in peacekeeping operations is promoted by the government as a way to provide the military with an external mission. Laurance (1996) blames slow progress in arms control in Latin America to the pressure that the military, anxious to protect its budget, exerts on the process. Jervis (1999) notes that arms control agreements favor "doves," strengthening their hand in internal negotiations.

There is a vast body of literature related to this theme in Latin America, especially dealing with the two domestic institutions we are most concerned with the military and civilian sectors of government. The literature on corporatism, authoritarianism and transitions and consolidations to democracy are well developed in the field of Latin





72


American studies. Clearly the more general literature on the relationship of democracy and war, the role of domestic institutions in other areas of international relations (particularly trade) and studies linking military culture and strategic doctrine are also valuable. These issues will be discussed further in the concluding chapter.

Returning to the neoliberal institutionalism, we can now readdress the issues posed by the cooperation dilemmas outlined in Table 2-1. We will look to see if institutions facilitate cooperative behavior in each of the behavior clusters without regard to the institutional politics described for asymmetrical situations. Behavior there is driven by the twin dynamics of reassurance and institutional balancing and not necessarily a search for the absolute gains promised by cooperation, thus institutional design may be only partially effective in addressing cooperative dilemmas.

Since military transparency can expose free riding, cheating and relative gains

problems, it is fundamental to institutional attempts to promote security cooperation. The challenge is to verify military information through various mechanisms, designing reporting formats that can be easily cross-checked and building an extensive network of contacts that produces significant information flows. Institutions can promote reductions in defense postures by addressing distributive gains through safeguards, and trade-offs in types of reductions. But, as we shall see, there are no hemispheric or regional regimes that include these measures as yet. Relative gains are normally not addressed. Will that matter?40 With regard to burden sharing, again transparency exposes cheating and


40 These kinds of rules, which have been present in the international trade regime for many years, include safeguards (related to balance of payment crises) and exclusions (agriculture, textiles, etc.). See Hasenclever, et al. (1997).






73


relative gains problems. Free riders may enjoy benefits of enhanced security but are

denied benefits of interaction. These points are summarized in the table below.

Table 2-2: Cooperative Problems Regime Solution.

Security Absolute Gain Regime Solution
Behavior
Transparency Better security Reliance on reciprocal calculation avoids verification. worst case analysis -- importance of reciprocal visits/exchanges.
-- designing formats that can be easily cross checked.


Reduction of More resources Transparency exposes free
Defense Postures available for riding, cheating and relative economic gains problems. development and other goals Institutions can address distributive gains through safeguards, and trade-offs in types of reductions. Regional Burden More efficient Transparency exposes
Sharing for allocation of cheating and relative gains Regional Security security resources problems. Free riders may enjoy benefits of enhanced security but are denied benefits of interaction.


Institutions can promote equitable sharing of costs and benefits, such as rotation of leadership roles, cost sharing for exercises, negotiation of side benefits.






74


Contending Hypotheses

Hemispheric Level

At this point, we are ready to summarize contending hypotheses for security cooperation from the two theoretical frameworks discussed above realism and institutionalism. We will formalize hypotheses that address organizational characteristics, rule creation and security behavior in each of the clusters examined. In this section, we focus only on the hemispheric level. Subregional hypotheses are developed in the following section.

As we shall see in more detail in Chapters 3 and 4, there is no denying the

substantial institution building process in the area of hemispheric security that has been underway since the advent of WWII. Old hemispheric security organizations have been substantially renovated and several new ones have been created during that period. But are they sufficiently resourced by member states to effectively facilitate negotiations and carry out the functions of regime management? Realist thinkers doubt that possibility, except where such an arrangement might serve the interests of the most powerful states.

A realist hypothesis would make resourcing contingent on an agenda consistent with the new, post-Cold War security agenda of the United States. Thus institutions should be largely underwritten by the United States and generally supportive of the U.S. security agenda, notwithstanding small deviations that do not threaten core U.S. security concerns. Other countries are likely to participate only in areas of common interests.

Institutionalism may not differ markedly from this prediction, since no

hemispheric institution or regime could be expected to resist U.S. influence. However, even under conditions of marked power asymmetry, institutions may demonstrate some






75

autonomy. Through issue linkage and the legal principle of formal equality of members, institutions work to reduce asymmetry and promote a more representational agenda that may confound large power interests.

Sharper distinctions are possible with regard to the rules that these institutions promote. Grieco (1988a) offers a set a contending hypotheses for international agreements (rules), built around the problem of relative gains. These hypotheses can be fairly easily adopted to our study of patterns of security cooperation in the Americas.

In the first instance, Grieco notes that under realism, states will in general eschew cooperative arrangements with binding commitments that present high exit costs. This should be particularly true in the area of security, where relative gains problems are most acute. Thus realism would expect that hemispheric rules regarding transparency, verification, defense postures and burden sharing would be voluntary and non-intrusive. In the absence of reliable verification, the effectiveness of reciprocity will be limited.

This contrasts with agreements made between the United States and its security partners, such as the network of mostly bilateral agreements for tracking and interdicting illicit drug production and trafficking that exists between the United States and source and transit countries. As rules become more inclusive, they will lose their specificity.

Under institutionalism, theory tells us that states, given the opportunity to

negotiate, will tend to insist on binding rules, with strong verification, whenever relative gains may present an obstacle to cooperation, so that the potential absolute gains can be "locked-in." These steps are not coercive; member states are simply binding themselves to reassure partners of their cooperative intent.






76

We can derive some realist predictions for specific behaviors in

transparency/verification, defense postures and burden sharing as well. Transparency in military matters should be limited to information that is already public. Reductions in defense postures should be based on threat perception, not regime rules. And burden sharing, unless sponsored by the United States or another major power, should be extremely tentative. In states that share the kind of cross-border problems that we have dubbed transnational threats, burden sharing should be even more difficult because of the added source of conflict.

Institutionalists, by contrast, would expect that relevant and reliable information will emerge from institutional arrangements. Institutions with the administrative capacity to handle the information will add considerably to the degree of transparency in the system. The increase in transparency will allow reductions in defense postures to proceed in a reciprocal manner, where levels might be higher than threats warrant. Under these conditions, transnational security problems should attract cooperative behaviors, since absolute gains to such cooperation are relatively high. Again, transparency will insure that no state benefits unduly from such cooperation. Side negotiations facilitated by institutions can redress distributional problems that might emerge.

Finally, with regard to institutional effectiveness, realism would contend that rules will not be effective outside of the U.S. agenda. Patterns of cooperative security behavior should conform to U.S. priorities. Participation will not be inclusive, but rather concentrate on key U.S. security partners. This pattern may attain even in the absence of security institutions, since the United States will pursue its own agenda bilaterally when multilateral approaches do not advance its interests. Finally, from Mastanduno, realism





77

would predict that the United States is likely to act with restraint towards more powerful

Latin American states in order to avoid eliciting a balancing reaction.

Institutionalism expects to find rule-based behavior not behavior driven by

balancing or U.S. influence. Bolstered by expectations of convergence (supported by

binding rules and strong verification schemes) and eased by the facilitation that

institutions provide (reduced transaction costs), states will tend to follow the rules

established in organizations formed for that purpose. However, if acts of reassurance are

insufficient to maintain cooperation, then the dynamics of institutional politics will

predominate. These contending hypotheses are summarized in Table 2-3 and Table 2-4.

Table 2-3: Contending Hypotheses at the Hemispheric Level for Institutions

Dependent Realist Institutionalist
Variables Hypotheses Hypotheses
Security Given institutions as "constructed focal points" for negotiations and Institutions: absolute gains from military transparency, reduced defense postures
and burden sharing, under conditions of marked power asymmetry ...

Organizational 1. Institutional agendas will not 1. Institutional procedures and stray significantly from U.S. issue linkage will restore Features interests and will be symmetry, providing resourced only where opportunities for institutional interests coincide. Issue balancing. Agenda will be linkage will complicate multilateral, focused on issues cooperation. where absolute gains are evident.


2. Rules will not be binding and 2. Rules will be binding wherever Rule comprehensive, but rather the potential for defection voluntary and partial. exists. Creation
3. Rules will not be supported 3. States will insist on strong
by verification mechanisms verification mechanisms so that sufficient for reciprocity. reciprocity can function.






78




Table 2-4: Contending Hypotheses at the Hemispheric Level for Behavior

Dependent Realist Institutionalist
Variables Hypotheses Hypotheses
Security Given institutions with the administrative capacity to carry out the Behavior: functions of regime management (data collection, analysis and reporting) and rules promoting transparency, defense reductions and
burden sharing ....

Transparency, 4. Transparency measures will 4. Transparency will be not reveal information of considerable, providing Defense value, reporting instead relevant and reliable information that is already information. Postures, largely public.

Burden 5. Defense reductions are 5. Defense reductions will take unlikely and in any case will place where transparency Sharing reflect perceived threats. provides assurances of reciprocity.

6. Burden sharing is unlikely
unless sponsored by the 6. Burden sharing will take place United States or resulting where transparency provides from balancing and/or assurances of reciprocity. The bandwagoning. The presence presence of transnational of transnational security security problems makes problems makes cooperation cooperation more likely even less likely because of because of the added potential the added potential for for absolute gains. relative gains.



Subregional Level

What modification of these hypotheses might be necessary at the subregional

level? There is good reason to believe that the basic tenets of realism and institutionalism

will operate there with little modification. Under realism, the dynamic of security





79


competition should not be any less salient. If anything, security competition will be more intense at the subregional level where fundamental security dilemmas operate and where threats are more proximate to use Walt's terminology. Under institutionalism, the dynamic of security cooperation should also be more intense, since neighboring states are likely to share compatible security interests especially transnational security problems leading to the possibility of greater absolute gains through cooperation.

The fact that both security competition and cooperation are more intense has an

impact on subregional institutions if they are to be effective, they must be more binding in nature than their hemispheric counterparts.

There are other reasons why two theories should work well at the subregional level. The question of multiple linkages becomes intensified at the subregional level, both because neighboring countries often share transnational concerns over a wide range of issue areas and because subregional institutions are normally nested in their hemispheric counterparts. These factors taken together mean greatly amplified connectivity between states in a given subregion. Of course, subregions may differ in terms of the density of their interrelations, but this factor simply adds another testable hypothesis to a comparative analysis.

Grieco (1988a) offers an additional comparative test of the relative explanatory power of realism and institutionalism that fits nicely within the subregional framework. He bases this comparative test on the "number of partners states prefer to include in a cooperative arrangement" (Grieco, 1998a, p. 506). According to his review of the theories, advocates of institutionalism assert that a "smaller number of participants facilitates verification of compliance and sanctioning of cheaters." Thus, Grieco points






80


out, states that can choose the context of cooperative arrangements will normally prefer to limit membership. Realism, according to Grieco, predicts the opposite, since arrangements with more members provide more partners and hence more opportunities to offset relative gains advantages. For these reasons, realism would expect that cooperative arrangements at the subregional level will be even more difficult to accomplish than at the hemispheric level.41 Institutionalism predicts precisely the opposite pattern.

In sum, institutionalism expects more institution building and more binding rules at the subregional level than the regional level. Realism predicts the opposite pattern. In any case, both paradigms are operative at this level of analysis.

What then about the effectiveness of rules regarding security cooperation?

Institutionalism predicts that security behavior will be rule-based, following the logic laid out in the previous section. Transparency will be critical because of its important role in communication. Reductions in defense postures and burden sharing are likewise important because of the absolute gains they promise. Realism, on the other hand, expects that patterns of security behavior will follow the logic of balancing; cooperation will be limited to informal ententes formed to counter subregional threats. We can approximate the predicted pattern through balance of threat analysis for each subregion.





41 "Defensive realists" who, according to Jervis (1999), adopt many of the standard "cooperation under anarchy" strategies, would disagree with Grieco's formulation. The security dilemma/collection action literature highlights the difficulties of large alliances, since detecting cheaters and enforcing obligations is difficult with large groups. But Grieco is not focusing here on cheating, but rather on the relative gains problem, which is more intense in smaller groups.






81


What of U.S. power? The United States is still an important security externality that affects subregional security behavior. But it may be possible to abstract some of that impact from the security equation by adopting a comparative approach that selects subregions differing in terms of their relationship with the United States. In which context is the United States essentially one more player in a subregional security order, a potential ally for balancing conflict? Where does it act as a security manager mediating, shaping and preventing conflict?

The answers to these questions will depend on the security issues important to the United States and the national capacity it can mobilize for influence.

A hegemonic power that provides some public goods may, as Lake (1993) points out, engage in benevolent behavior that is tolerant of free riders or coercive behavior that attempts to extract payments. Lake predicts that coercive behavior is a function of the resources available for coercion and the likelihood that coercion will be effective. In this study, we are less concerned with outright coercion than with influence attempts, which may range from shaping the security environment to direct bilateral negotiations on specific issues. Regardless, we can assume that more resources are available for security issues high on the U.S. agenda. We can also assume that those resources will be more effectively employed in regions with less autonomy relative to the United States.

For example, the United States may be able to induce regional cooperation by supporting verification mechanisms (arguably in Central American demobilization), providing significant side-benefits (perhaps in the case of the Peru-Ecuador peace accord), or sanctioning uncooperative behaviors (as in decertification of Colombia). In






82


other regions, states may be able to choose whether or not to cooperate with the U.S. agenda.42

Regions more distant from U.S. power and security concerns and more developed in terms of national power should retain some measure of autonomy, since relative capabilities are less asymmetrical and intervention is less likely. In fact, such subregions may enter into agreements that allow them to resist U.S. influence. Examples addressed later include Brazilian resistance to U.S. attempts to influence peacekeeping training and the reluctance of the Andean countries to enter into a multinational, operational agreement with the United States to combat production and trafficking in illicit narcotics. But even less autonomous regions can demonstrate some independence. For instance, Nicaragua's participation in almost all aspects of Central American security cooperation demonstrates independence from the United States, which until very recently followed a policy of limited military-to-military contact with Nicaragua.

Despite this last example, it seems evident then that the subregional security order in Central America will more closely approximate a situation of hegemonic control due to the region's proximity to the United States and its relatively lower level of development. Countries of the Southern Cone, on the other hand, operate with relative independence. The Andean countries fall somewhere between these two extremes. In any case, if a





42 They will have two calculations to make: (1) does "cooperation" provide absolute gains (given unreliability of the United States as a partner and potential incompatibility of interests) and (2) will neighbors will enjoy relative gains through their own cooperation with the United States.






83

particular security issue is not high on the U.S. agenda, then resources are not likely to be available for influence attempts.

With the end of a threat of nuclear proliferation in the Southern Cone, the main U.S. strategic concerns in Latin America are threats posed by illicit trafficking in drugs and general social instability in nearby regions (causing illegal immigration or providing a haven for terrorist/guerrilla groups). These threats intensify as they become more proximate.

Aside from the sharpened distinctions between contending hypotheses, moving to the subregional level offers a key methodological advantage. It creates more case studies and allows use of the comparative method. The comparative method allows us to gain analytical traction and arrive at a more robust case for our contending paradigms. I plan to deal with the "small-N" problem by focusing on the comparable cases cases that are similar enough to allow for good comparison, but that differ in important ways that affect the relative gains problem. 43

The subregional cases considered in this study the Southern Cone, Central

America and the Andean Ridge certainly do not exhaust the possibilities for scholarly research. There are other regions that merit examination: the Rio de la Plata subsystem; the Amazon region, the Caribbean, North America; the Pacific Rim countries, etc.



43 Collier's (1991) update of Lipjhart's (1971) influential essay on the
comparative method is useful. On the "small N, many variables" problem, the literature offers two approaches: (1) Most Different Cases: gaining analytical traction by selecting disparate cases to isolate explanatory variables and (2) Most Similar Cases: gaining analytical traction by selecting similar cases so that other variables are "canceled" out, leaving the explanatory variable to explain differing outcomes.






84


Nevertheless, a focus on the Southern Cone, Central America and the Andean Ridge should suffice to provide a valid test for contending hypotheses derived from realism and institutionalism.

These cases correspond to geographical regions with long histories of security cooperation and conflict. All have been conflictual at various times; all have border problems and militarized disputes; all have sought ways to resolve those conflicts cooperatively. They evidence an established balance of power and their member countries have used power both demonstrations and applications of force to maintain that balance. This balance also allows us to depart from an assumption of stability, and eliminates the methodological problem of cataloguing power indices for realist arguments.

All of these subregions are attempting to address regional security issues within the context of democratic orders, with civilian control of armed forces, albeit in varying degrees. All are developing states, albeit at widely disparate levels of development. All are dealing with the implications of a new global economic order and have adopted, in varying degrees, a similar developmental model, although populism bubbles beneath the surface. All inhabit a hemisphere dominated by the United States, although U.S. impact on the security policies other states varies. This difference, of course, is central to the research design.

The subregions also differ in terms of their principal security threats. This

difference is welcome, because it allows a more robust test of the contending hypotheses. The Central America countries are beset by transnational threats (drug trafficking, organized crime, illegal immigration) that overlay the standard security dilemma between






85


states in the region. This is largely the case with the Andean countries as well, although relative power concerns may overshadow transnational threats in some border regions, such as the Cenapa region along the Peru-Ecuador border. In the Southern Cone, there are few transnational considerations, aside from smuggling of contraband in the tri-border area of Argentina, Brazil and Paraguay. The primary security problem is the traditional one of maintaining a reasonable balance of power.44 These variations in security problems can be handled by our existing theories. All require some sort of security cooperation for their solution. They differ only in the details of the absolute and relative gains that result from that security cooperation. Thus, the differences can be expressed in specific research questions derived from the basic theories. Table 2-5 and Table 2-6 summarize the hypotheses developed here for the subregional cases.

















44 Military spending in Latin America has historically been concentrated on the Southern Cone countries of Argentina, Brazil and Chile. During the arms build-ups of the 1980s, these three countries accounted for almost two-thirds of the total military expenditures in the region (Varas, 1985).






86


Table 2-5: Contending Hypotheses at the Subregional Level for Institutions

Dependent Realist Institutionalist Variables Hypotheses Hypotheses
Security Given institutions as "constructed focal points" for negotiations and Institutions: absolute gains from military transparency, reduced defense postures
and burden sharing, in subregional security systems with general
power parity ...

Organizational 1. Institutions will be under 1. Institutions will be resourced by resourced unless sponsored members. Agenda will be Features by United States, in which multilateral, focused on issues case their agendas will not where absolute gains are evident. stray significantly from U.S. Linkages will enhance iteration interests. Issue linkage will and increase benefits of complicate cooperation. cooperation and costs of defection.
U.S. influence varies by subregion.


2. Rules will be less binding at 2. Rules will be more binding at Rule the subregional level than the the subregional level than the hemispheric level. hemispheric level. Creation
3. Verification mechanisms will 3. Verification mechanisms will be weaker at subregional be stronger at subregional than than hemispheric level, hemispheric level to ensure reciprocity.






87


Table 2- 6: Contending Hypotheses at the Subregional Level for Behavior

Dependent Realist Institutionalist
Variables Hypotheses Hypotheses
Security Given institutions with the administrative capacity to carry out the Behavior: functions of regime management (data collection, analysis and reporting) and rules promoting transparency, defense reductions and
burden sharing ....

Transparency, 4. Generally, transparency 4. Transparency will be measures will not reveal considerable, providing Defense information of value, but relevant and reliable some information sharing is information, so that security Postures, likely between states in a cooperation can take place to balancing entente. reap absolute gains. Burden
5. Defense reductions are
Sharing unlikely and in any case will 5. Defense reductions will take reflect perceived threats. place where transparency and verification provide assurances
of reciprocity.
6. Burden sharing is unlikely
unless sponsored by the
United States or resulting 6. Burden sharing will take place from balancing and/or where transparency and bandwagoning. Security verification provide assurances cooperation should reinforce of reciprocity. The presence of ententes and sharpen rivalries, transnational security problems The presence of transnational makes cooperation more likely security problems makes because of the added potential cooperation even less likely for absolute gains. because of the added potential
for relative gains.




Operationalization

Careful consideration of the power and interests of the states in the system and

comparison of those factors with the design of security institutions and content of security

rules will illuminate the relative strengths and weaknesses of the two theories. National






88


interests are developed primarily from official sources. Sources for U.S. interests include published data such as the U.S. National Strategy, National Military Strategy, the Quadrennial Defense Review, and other informative sources. Latin American interests are revealed through official documents and debates in institutional forums, which often present well-developed national positions on a wide variety of security issues.

Regional military capabilities can be objectively determined through a number of open source conflict profiles and balance of power studies. Although precise measurement over time is difficult, rough profiles suffice to bring out balancing/bandwagoning proclivities. Presence of enduring border disputes offers evidence of aggressive intent. This mix of interests and capabilities provides one explanation for patterns of security cooperation.

Institutions provide another. Observable characteristics include (1) degree of administrative capacity to support multilateral decision-making and track compliance;

(2) issue agenda, particularly as compared to U.S. security interests; and (3) issue linkages, especially nesting of security institutions within larger structures. The rules produced by security organizations can be examined as to their degree of bindingness and ease of verification. Sources for these factors derive from official reports and minutes of meetings in institutions of IASS. Substantial new material from the archives of the IADB, the OAS and the Defense Ministerials is introduced for this purpose. Staff reports from the IADB on exercises, security studies and defense plans are also key to tracing the evolution of institutions. Research monographs from senior Latin American officers assigned to the Inter-American Defense College also add new perspectives on that






89


subject. Careful reading of this material provides definitive information on the resources,

agenda and rules of the security institutions studied. Table 2-7, below, summarizes this

discussion.


Table 2-7: Principal Indicators and Sources for Cooperative Security Institutions

Dependent Factor Indicators Source Variable
Administrative Forums for negotiation (conferences, Documents from Organi- Capacity working groups, commissions, entities indicated zations committees, etc.); decision-making procedures; mechanisms to track compliance, evaluate results and report to members; enforcement
mechanisms.
Issue Published agendas and work program Documents from
Agenda Summits,
Ministerials, OAS
and IADB meetings.
Issue Linkages Linkage in formal exposition of Documents from security rules; state bargaining Summits, positions Ministerials, OAS and IADB meetings.
Rules Bindingness Language of agreement Published agreements
Verification Specific mechanism (reciprocal visits, Documents from publication of data) Summits, Ministerials, OAS
and IADB meetings.



Regime effectiveness, as opposed to institutional evolution, is more amenable to

quantitative approaches based on key indicators of security behavior. Table 2.8,

"Principal Indicators and Sources for Cooperative Security Behavior," summarizes this

approach.

Indicators of military transparency include participation in the OAS/IADB CSBM

inventory and the UN arms and defense expenditure registries; publication of defense






90


"white papers" or other official documents outlining strategy (intent), doctrine (orientation), force structure (capability), and training/maintenance (readiness); advance notification of exercises; and exchanges/visits at military schools, installations and exercises; and weapons demonstrations.

None of these indicators is perfect. Measures reported to the OAS/IADB in the CSBM inventories are often difficult to verify. Common methodology for comparing defense expenditures has not been implemented in the region. The UN arms registries do not address arms production and in any case do not include categories for many small caliber weapons of importance to the region. Few defense white papers have been released, perhaps because of the difficulties of coordinating inter-agency positions in recently established democracies that are still institutionalizing new civil-military relations. Exercise notification is still spotty. Finally, exchange programs are often not designed to enhance transparency; they are frequently a tool for influence or deterrence. Furthermore, exchange officers are sometimes excluded from participating in sensitive defense matters.

Defense postures are somewhat easier to determine objectively. Good studies on military capabilities and arms production and transfers are available. Trends in defense postures during the 1990s can be established. But military readiness is more elusive; it relies on factors as disparate as a state's recruitment model, morale in the armed forces and national will. Data in these areas is incomplete or unavailable.

Indicators of burden sharing behavior suffer from some of these same problems. Not all of the measures listed in Table 2.8 can be effectively verified. Furthermore, subjective judgments must be made regarding the impact of some of the indicators on






91


actual security postures and their effectiveness in addressing common security risks. Such judgments can rarely be wholly objective.

Even where data is available, rigorous quantitative analysis may not be possible. In the case of the CSBM inventories, for example, while geographical and thematic patterns are readily discernable, quantitative comparisons between states may not be possible due to problems with the data as it has been compiled to date. The 1995 inventory of CSBMs, for instance, prepared by the IADB, suffers from a series of limitations that preclude any accurate quantitative analysis. In the first place, little more than half of the members of the IADB participated. Second, those that participated did not follow any standard format for categorization of CSBMs. Third, CSBMs were not standardized in terms of size or duration, leading to crippling aggregation problems. Fourth, the data reported was not verified, and bilateral events reported by one country were often not reflected in the partner country's submission. Due to lack of standardization, countries made their own decisions about what measures to report. These issues are slowly being address in the reporting formats, but the data submitted since 1995 do not reflect a marked improvement.

Despite these methodological problems, the cluster of indicators for transparency, burden sharing and defense postures combine to give weight to the analysis of security cooperation. A picture of security cooperation emerges which can evaluated against power and interests, on the one hand, and institutional rules on the other.






92


Table 2-8: Principal Indicators and Sources for Cooperative Security Behavior

Factor Indicators Source Observation
Military Participation in OAS IADB inventories These inventories and Trans- and IADB inventories of (95,96,97), OAS registers were designed to be parency declared CSBM data inventories (96, 97, as non-intrusive as possible 98), in order to make participation Participation in UN UN Arms Transfer universal without the burden
Arms Transfer registries Registers (92-97) of verification. However,
Participation in UN UNDCA annual there is an informal
Defense Expenditure reports of use of verification mechanism in the
Reporting Standard Instrument sense that the uncoordinated for Reporting individual submissions can be cross-checked after they are publicly disclosed, which provides a disincentive to cheating.
Public release to all Published white
countries of "white papers (US, Brazil,
papers" or other official Chile, Canada);
documents outlining national positions as strategy (intent), presented in official doctrine (orientation), statements at OAS, force structure summits, and other
(capability), and conferences;
training/maintenance support/nonsupport
(readiness). for cooperative
measures proposed
at the IADB/C and
OAS.
Advance notification of As reported in Announcement of exercises
exercises IADB inventories in multilateral forum (IADB) (95,96,97), OAS or through unilateral inventories (96, 97, procedures. 98),
Exchanges/visits at As reported in Note that these are reciprocal,
military schools, IADB inventories bilateral (occasionally
installations, exercises, (95,96,97), OAS multilateral) programs, but
and weapons inventories (96, 97, they are excellent indicators
demonstrations 98), of willingness to be open, since capabilities, readiness and doctrinal training are all subjected to direct
observation. Exchanges in operational units and
doctrinal schools are serious business.






93


Table 2-8 -- continued

Factor Indicator Source Observation
Defense Arms Production Authoritative Postures Arms transfers military analysis and
Arms control arms control
agreements entered into documents Janes, SIPRI, IISS,
UNCDA, ACDA,
etc.
Burden Coordination to solve As reported in While coordination does not Sharing technical problems IADB inventories necessarily represent a strong
protocols for border (95,96,97), OAS commitment to cooperative
situations inventories (96, 97, security, the conduct of
98), actual operations and training Coordination to solve As reported in exercises between countries technical problems IADB inventories does, because of what those
protocols for safe (95,96,97), OAS activities reveal about
transport by ground, air inventories (96, 97, national doctrine, readiness
and sea 98), and capabilities.
Bilateral and As reported in Furthermore, such activity
multilateral exercises IADB inventories can promote interoperability (95,96,97), OAS development of common inventories (96, 97, doctrine, common equipment, 98), training and logistics. In the Bilateral and As reported in final analysis, it can also lead
multilateral IADB inventories to military specialization a
humanitarian operations (95,96,97), OAS division of labor that makes inventories (96, 97, multilateral security 98), operations more effective, but Bilateral and As reported in may leave countries much multilateral IADB inventories more vulnerable; Uruguay peacekeeping operations (95,96,97), OAS specializes in Peacekeeping inventories (96, 97, Operations, for instance. 98), Other examples include Utilization of common Data from developing force structures
training centers (e.g.: institutions based on light aircraft and School of the Americas, involved. riverine craft for counterCenter for Hemispheric drug operations or focusing Defense Studies, the on humanitarian operations
Argentine Peacekeeping such as demining.
Operations Center, the
Central American Police
Academy)














CHAPTER 3
ASYMMETRY AND SECURITY

Introduction

This chapter deals with the first of our case studies, which provides an example of the dynamics of institutional security politics at the hemispheric level. Security cooperation in the Americas begins in earnest with the onset of WWII, tracing a trajectory that rises during the war years to a highpoint in the 1950s. At its peak, U.S. military influence reached astounding proportions. U.S. bilateral military relations boasted basing rights, military assistance programs and logistics support agreements with key countries throughout the hemisphere. U.S.-led multilateral initiatives resulted in the creation of hemispheric security institutions, the adoption of common defense plans and the employment of an Inter-American Peace Force. This level of influence declined during the difficult 1960s to a turning point arguably at the Buenos Aires conference convened to reform the OAS Charter in 1967. Attempts to bring the Inter-American Defense Board under the Charter and create a permanent inter-American force were turned back. Security multilateralism would not be revived until the 1990s.

This was not simply the inevitable result of the institutional politics of security in an asymmetrical regional order. In the end, U.S. inattention, diversion of interest, and aversion to the authoritarianism taking root on the continent, coupled with the impact of economic development in Latin America, contributed more to the loss of U.S. influence




94






95


than institutional politics of hemispheric security. Still, institutional politics had an important impact that is well worth an investigation.

U.S. policy makers designing hemispheric defense plans during WWII were aware of the difficulties inherent in a multilateral security policy. But the options of purely bilateral and unilateral policies were expensive, and in a theater with few core interests, the United States was not prepared to muster the resources necessary for such a course of action. Making the case for multilateralism required convincing Latin American states that Axis countries posed a concrete threat to the entire region. Until the Japanese attack on Pearl Harbor consensus on that point was not overwhelming. This is especially true of the Southern Cone countries, where German immigrants were an important presence. Nevertheless, over time the United States was able to win the interest of the other states of the region, working both bilaterally and multilaterally to establish a system of defense for the hemisphere.

However, from the outset, there was a reluctance to organize hemispheric security around a permanent military organization, such the IADB, which Latin American states assumed would be powerfully influenced by the United States. But institutional balancing lay dormant as long as mutual interests were served by the institution. That changed once the United States redefined the threat from external to internal. U.S. pressure on Latin American states in Caracas (1954) to make the presence of domestic communist movements a trigger for invoking the Rio Treaty stirred resentment and prompted the institutional balancing that ultimately undid the post-WWII multilateral security system.






96


After the Cuban Revolution, the United States was able to proceed on a

multilateral track only through increasing support to multilateralism (creation of new institutions) and by virtue of the well-intentioned Alliance for Progress and military assistance programs (providing side benefits). Hyper-sovereign Mexico, a newly recalcitrant Brazil (before 1964) and unpredictable Argentina were already challenging the U.S. Cold War security agenda. When the United States settled into a pattern of interventionism along its southern flank, the game was up. By that time, its influence had waned and restoration of the hemispheric security organizations it had largely created was probably out of reach.

But the dynamic of institutional politics is only one piece of security cooperation and conflict in the region. We need to begin with a look at the enduring rivalries that shape military balancing in the region. This chapter continues with a look at the Latin American balance of power system, then overlays the hemispheric system that the U.S. ushered in during the years leading up to WWII. The main purpose here is to establish the context for the study of cooperative security in the 1990s, which is taken up in Chapters 4 and 5.

The Role of Power in Hemispheric Relations Early Patterns

In the Americas, as in other regions, power has been the ultimate arbiter of international relations. Alternatives have been hard to find. For instance, the role of international law in Latin American security affairs has been a curious one, more often honored in the breach than not. As Mares (1997) notes, "between 1826 and 1889 at least




Full Text
157
Summary
The consistent fault line that divides the hemisphere the U.S. security agenda vs.
concerns of sovereignty from other countries of the region means that security
institutions rest on a shaky foundation. Under these conditions, realism expects that Latin
American states will under-invest in hemispheric institutions and hemispheric security
institutions are unlikely to function unless sponsored by the United States and subservient
to U.S. security interests.
What has emerged from our study of the evolution of security institutions? First,
that an expanding network of new and remade organizations is remarkably active in the
security field. For the most part, they overlap membership and are mutual reinforcing.
The renewal process at the OAS led to the creation of a Committee on Hemispheric
Security, built along the same fault line that the Board has had to straddle for its entire
existence. The challenges encountered by the Committee seem familiar. But the
Committee overcame its critics and became an active focal point in the OAS for security
issues. The Board evolved away from its planning mission to assume a supporting role to
the Committee. Creation of new organizations outside of the OAS for the most part
furthered collaborative approaches. It is difficult to describe this process of institution
creation as under-investment.
The second conclusions is that while almost all of the organizations relied on
substantial US support, their evolution did not always reflect U.S. preferences.
Institutional balancing prevented the full incorporation of the IADB under the OAS.
Furthermore, the security priorities that emerged did not always match U.S. concerns;


281
All states are also signatories of the Ottawa Convention and are committed to
destroying stockpiled antipersonnel mines. Nicaragua destroyed its first tranch of 5,000
antipersonnel mines in 1999, leaving approximately 15,000 reported mines still
warehoused. Operations are expected to be complete within the four years stipulated by
the convention. Honduras has also begun destruction of its mines.
Modernization programs in the armed forces of the states of the subregion are
generally in compliance with these restrictions. El Salvadors Plan Arce guides its
current modernization effort, which is primarily focused on changing the organizational
structure and professionalizing its officer corps (El Salvador, 1998c) and not on the
acquisition of new weapons or the formation of new combat units. Of course,
organizational change is an important avenue to enhancing capability in Latin America
(Resende-Santos, 1996).
Honduras is undergoing a major restructuring of its armed forces to bring them
more effectively under the newly appointed civilian defense minister. The military is
losing its police functions, its independent resource base, and ready source of manpower
with the end of conscription. Similar steps are underway in Nicaragua. Guatemala's
reform has been delayed by the protracted conflict there and the 1999 rejection by
plebiscite of a package of reforms that included changes in the military. But the trend
there is also towards a reduced role of the armed forces in determining defense policy a
trend that is supportive of the rules regarding the implementation of a regime of
"democratic security."
The United States, which maintains a forward presence (Joint Task Force Bravo)
at Honduran Air Force Base Enrique Soto Cano, is still an important factor in the


23
States to dominate a process of "securitization"19 that defines hemispheric threats (Inter-
American Dialogue, 1997). The asymmetry in power relations driven home by the
1989 invasion of Panama to address the results of a "narco-regime" make hemispheric
consensus difficult to arrange around these sorts of issues.
The "internationalization of politics" is another feature of the impact of
globalization (Muoz, 1996b). Its first order effect the awakening of ethnicities and
nationalities in Eastern Europe did not have a major impact in Latin America, where the
business of nation building is largely complete. But new norms in the area of human
rights and democracy found ready acceptance in the region, which was just completing its
transition from authoritarianism as the 1990s began. With democracy as a unifying
theme and the absence of "alliance handicaps" stemming from ideological differences
(Liska, 1962), the possibilities for hemispheric convergence seemed bright.
But as applied to security relations, the picture is more complex. In the early
1990s, ground work was laid for the "collective defense of democracy": OAS Resolution
1080 adopted in Santiago (1991) and the Washington Protocol amending the Charter
strengthened automatic responses within the OAS to ensure a strong reaction in the event
that democratic processes were interrupted in a member state. But as demonstrated
during the Haitian crisis, the political will to violate sovereignty in the defense of
democracy proved impossible to muster (Maingot, 1996). Mexican opposition to U.S.
policy in this incident was strong. Mexican policy remains anchored in the principle of
19 This is the term employed by Buzan, Waever, and de Wilde (1998) to explain
how an economic, social or cultural issue becomes identified as a threat within a given
regional security complex.


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
t- N- 'fa'*
Terry L. McCoy, Chairman
Professor of Political Science
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Professor of Geography
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jj/ca+HP' Am#
M. ann Brown
A&ociate Professor of Political Science
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Political Science
This dissertation was submitted to the Graduate Faculty of the Department of
Political Science in the College of Liberal Arts and Sciences and to the Graduate School
and was accepted as partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
December 1999
Dean, Graduate School


84
Nevertheless, a focus on the Southern Cone, Central America and the Andean Ridge
should suffice to provide a valid test for contending hypotheses derived from realism and
institutionalism.
These cases correspond to geographical regions with long histories of security
cooperation and conflict. All have been conflictual at various times; all have border
problems and militarized disputes; all have sought ways to resolve those conflicts
cooperatively. They evidence an established balance of power and their member
countries have used power both demonstrations and applications of force to maintain
that balance. This balance also allows us to depart from an assumption of stability, and
eliminates the methodological problem of cataloguing power indices for realist
arguments.
All of these subregions are attempting to address regional security issues within
the context of democratic orders, with civilian control of armed forces, albeit in varying
degrees. All are developing states, albeit at widely disparate levels of development. All
are dealing with the implications of a new global economic order and have adopted, in
varying degrees, a similar developmental model, although populism bubbles beneath the
surface. All inhabit a hemisphere dominated by the United States, although U.S. impact
on the security policies other states varies. This difference, of course, is central to the
research design.
The subregions also differ in terms of their principal security threats. This
difference is welcome, because it allows a more robust test of the contending hypotheses.
The Central America countries are beset by transnational threats (drug trafficking,
organized crime, illegal immigration) that overlay the standard security dilemma between


341
Child, Jack. 1996. Peacekeeping, Confidence-Building. In Richard L. Millett and
Michael Gold-Bliss, eds., Beyond Praetorianism: The Latin American Military in
Transition. Coral Gables, FL: University of Miami North-South Center, 11-36.
Christensen, Thomas J. and Jack Snyder. 1990. Chain Gangs and Passed Bucks:
Predicting Alliance Patterns in Multipolarity. International Organization 44 (2):
137-68.
Claude, Inis. 1962. Power and International Relations. New York: Random House.
Cohen, Benjamin J. 1983. "Balance of Payments Financing: Evolution of a Regime.
In Stephen Krasner, ed., International Regimes. Ithaca, NY: Cornell University
Press.
Coll, Albert. 1997. "United States Strategic Interests in Latin America; An
Assessment." Journal of Interamerican Studies and World Affairs 39 (1): 45-58.
Collier, David. 1991. The Comparative Method. In Dankwart A. Rustow and Kenneth
Paul Erickson, eds., Comparative Political Dynamics: Global Research
Perspectives. New York: HarperCollins, 7-31.
Colombia and Venezuela. 1998. Joint Presentation at the Third Defense Ministerial of
the Americas. Cartegena de Indias, Colombia.
Committee on Hemispheric Security. 1997. "Comments and Observations by the
Governments of the Member States on CSBM Adopted in the Declaration of
Santiago." CP/CSH-32/96 add.4-a. Washington DC: Organization of American
States.
Conferencia de las Fuerzas Armadas Centroamericanas. 1998. Presentation at the Third
Defense Ministerial of Americas. Cartegena de Indias, Colombia.
"Confidentiality accord with U.S. military viewed." La Nacin (Buenas Aires, internet
version), 13 October 1998, Foreign Broadcast Information System (on line), 13
October 1998.
Conn, Stetson and Bryon Fairchild. 1989. The Western Hemisphere: The Framework of
Hemisphere Defense. Office of the Chief of Military History. Washington, DC:
U.S. Government Printing Office.
Connell-Smith, Gordon. 1966. The Inter-American System. Oxford: Oxford University
Press.
Connell-Smith, Gordon. 1968. "The Inter-American System: Problems of Peace and
Security in the Western Hemisphere." In Robert W. Gregg, ed., International


50
Table 2-1- continued
Security
Behavior
Absolute
Gain
Collective Good
Problem and
Non-Cooperative
Behavior
Cheating
Problem and
Non-Cooperative
Behavior
Relative Gains
Problem and
Non-
Cooperative
Behavior
Regional
More
Regional security
Increased
Some states
Burden
efficient
can approximate a
vulnerability if
may gain more
Sharing
allocation of
collective good.
states to do not
from interaction
for
security
uphold
and some
Regional
resources
commitments.
members may
Security
gain more from
increased
Strong incentive
regional
to free ride.
security or
solutions to
States may not
transnational
cooperate or may
problems.
offset
vulnerabilities
Cooperation in
through increase in
many areas may
defense posture or
not be possible.
seeking alliances
(balancing.)
Thus, under the assumption of realism, the logic of collective action and the
security dilemma demonstrate that sustained security cooperation is difficult to achieve.
And yet we know that realism admits some forms of security cooperation. To better
understand what cooperative security behavior might emerge under realism, it is
necessary to look at the questions of sensitivity to relative gains and the dynamics of
alliance formation.
Grieco (1988b) develops the concept of sensitivity to relative gains (his "k"
factor), arguing that, among other criteria, states faced with increasing levels of threat, an


297
commanders, occur more frequently in the border zone. Most impressively, Colombia
and Venezuela established an Ad Hoc Commission on Frontier Incidents and an
organization known as COMBIFRON (Comisin Binacional Fronteriza Colombo-
Venezolan). The commission has actively intervened to address alleged border
violations by security forces.
At the third Defense Ministerial, held in Cartagena de Indias in 1998, the
Colombian and Venezuelan delegates made a joint presentation on COMBIFRON
(Colombia and Venezuela, 1998) which demonstrates a sophisticated approach to
transnational security issues in the subregion. The presentation noted that the shared
border, one of the most dynamic frontiers of the continent, had suffered serious security
problems since the beginning of the 1990s principally due to the presence of cross-
border "narcoguerrilla" activity. Solutions to this problem were hampered by lack of
communication and understanding between the military forces and police of both
countries.
In August, 1996, the defense ministers decided to create a bilateral commission to
promote direct relations between border units, allow for the exchange of information and
coordinate of operations. Responsibility for carrying out these initiatives fell to
COMBIFRON.
COMBIFRON is based on two principal documents: 1) a standing order which
prescribes its mission, composition and functions and 2) a standing operating procedure
which establishes the norms for continual cooperation between military commanders of
border units, from the platoon level up.


330
What is also clear is that El Salvador, again in contrast to the other states of the
subregion, has adopted a comprehensive set of measures in the area of military
transparency. El Salvador participates in the UN registries, submits comprehensive lists
of CSBMs to the IADB and the OAS, and alone in the subregion has published a defense
"white book" that describes its force structure, defense policy and defense budget. El
Salvador hosted a UN peace forum, a regional symposium of war college directors and
the second OAS conference on CSBMs. It has been a subregional leader in this field and
its high level of participation is unmatched by its neighbors. The military has
aggressively supported this effort, almost in competition with is foreign policy
counterparts. El Salvador maintains a strong presence at the IADB and a close affiliation
with the OAS delegate assigned to security matters.
The Chilean case is less straightforward, but suggestive nonetheless. Frohman
(1996) argues that Chile's transition although managed internally by Pinochet was
internationalized in the sense that the civilian leadership had extensive international
contacts, primarily through the links established by exiles during the authoritarian years.
A scholarly tradition of peace and security studies has become an important source of
thinking on security matters in the region. Work on CSBMs from the Facultad
Latinoamericana de Ciencies Sociales has provided academic credibility to the OAS
meetings of experts, conferences and committee work. A FLACSO scholar made a major
presentation at the third Defense Ministerial of the Americas on the next steps for
CSBMs in the region.
Not surprisingly, Chile has been at the forefront of military transparency in the
Southern Cone, as well as the region as a whole. But it is not so much that Chile's


92
Table 2-8: Principal Indicators and Sources for Cooperative Security Behavior
Factor
Indicators
Source
Observation
Military
Trans
parency
Participation in OAS
and IADB inventories of
declared CSBM data
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
These inventories and
registers were designed to be
as non-intrusive as possible
in order to make participation
universal without the burden
of verification. However,
there is an informal
verification mechanism in the
sense that the uncoordinated
individual submissions can
be cross-checked after they
are publicly disclosed, which
provides a disincentive to
cheating.
Participation in UN
Arms Transfer registries
UN Arms Transfer
Registers (92-97)
Participation in UN
Defense Expenditure
Reporting
UNDCA annual
reports of use of
Standard Instrument
for Reporting
Public release to all
countries of "white
papers" or other official
documents outlining
strategy (intent),
doctrine (orientation),
force structure
(capability), and
training/maintenance
(readiness).
Published white
papers (US, Brazil,
Chile, Canada);
national positions as
presented in official
statements at OAS,
summits, and other
conferences;
support/nonsupport
for cooperative
measures proposed
at the IADB/C and
OAS.
Advance notification of
exercises
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
Announcement of exercises
in multilateral forum (IADB)
or through unilateral
procedures.
Exchanges/visits at
military schools,
installations, exercises,
and weapons
demonstrations
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
Note that these are reciprocal,
bilateral (occasionally
multilateral) programs, but
they are excellent indicators
of willingness to be open,
since capabilities, readiness
and doctrinal training are all
subjected to direct
observation. Exchanges in
operational units and
doctrinal schools are serious
business.


189
DMAs and the CSBM Declarations have lent support to the conferences of the armed
forces, an important venue for enhancing interoperability.
Third, rules for transparency are significantly more dense and concrete than in the
areas of burden sharing and defense postures. In fact, transparency rules are far more
intrusive than the information exchange systems associated with collective security. The
rules reach into domestic organizations, aiming to restructure relationships between the
militaries of the hemisphere, and between military and civilian sectors of society. We
will return to this observation in the concluding chapter.
Fourth, there is ample evidence to point to a complicated environment of
overlapping regimes that provides a wide umbrella for security issues. Security issues are
clearly linked to efforts to strengthen democratic institutions, for instance, as well as
respect for human rights. A complicating factor, however, is the numerically large
presence of Caribbean states eager to link security issues with economic development.
One of their main vulnerabilities, they argue, is an undiversified economic base. This
attempted linkage has probably weakened attempts to bring more security concerns under
the purview of hemispheric institutions. No explicit links between security and economic
issues have been endorsed all attempts to include economic development as a part of
the security regime have been stymied, principally but not exclusively by the United
States.
What has not changed? First, the fault-line of U.S. versus Latin American
interests has not disappeared. The tension that derives from the power asymmetry in the
region is certainly still operative forcing institutional compromises and constraining
rule making, especially in the area of military burden sharing where the United States has


307
Conclusions
With the exception of the Andean subregion, the evidence provided by these cases
supports an institutionalist explanation more readily than one based on realism. While
there is some balancing behavior present, the strength of subregional institutions, the
density of their rule making and the generally compliant security behavior demonstrate
that subregional security regimes work. They work most effectively when they are nested
with other regimes that reinforce their prescriptions for behavior.
Let us review the major findings in more detail. In the area of institution building,
it is clear that far more activity here is present than realism can explain. To begin,
subregional institutions are generally stronger, more prevalent and more effective than
their hemispheric counterparts a finding that validates the institutionalist approach.
Secondly, they enjoy broad legitimacy, are reinforced in other subregional forums and
represent important investments in time and personnel from their member states. They
do not always have the resources to carry out their ambitious plans of action Central
America suffers in this regard but their impact is growing, not waning.
Rules are frequently binding at the subregional level, a marked difference from
what we discovered at the hemispheric level. This has progressed furthest among states
that share security problems, especially complex transnational issues. Again, this finding
is not consistent with realism. There is a problem, however, with inclusiveness, even in
the reduced context of subregional security orders. Binding rules with strong verification
were possible only among the trilateral states of Argentina, Brazil and Uruguay in the
Southern Cone; between the four states of Central America with military forces; and
between Colombia and Venezuela and between Peru and Ecuador in the Andean Ridge.


51
historical adversary and a relative decline in power will have a high sensitivity to relative
gains. Threat, as laid out by Walt (1987), includes the systemic elements of capability,
proximity, and state of technology (offense-defense), as well as the classic realist factor
of aggressive intent. This last element enriches Waltz's systemic approach with a unit-
level variable, at the cost of some loss of parsimony. Nevertheless, to the extent that the
presence of active border disputes can be taken as an indicator of aggressive intent, it is
relatively easy to measure and incorporate into our analysis.
Thus, in bounded subregions composed of proximate states with a history of
conflict, enduring border disputes and recent changes in power distribution, relative gains
dilemmas should be difficult to surmount. This will be especially so during times of
technological and organizational change which favor offensive operations, such as the
current "revolution in military affairs," based on unprecedented mobility, precision
weaponry and information technologies (Mares, 1996/7; National Defense University,
1995 and 1996). Conversely, in subregions with a pacific history, low threat levels and
stable power relations, relative gains sensitivity will be lower and some cooperation may
be possible, using standard "cooperation under anarchy" techniques (Oye, 1986).
Grieco (1988a) also argues that high levels of economic interaction in these
situations will tend to make relative gains problems more severe, assuming economic
advantages can be transformed into military capabilities (Grieco, 1988b). Taken together,
these factors help explain patterns of security cooperation and conflict between states.


324
Pittman (1988) studied the impact of democratization on disputes in the Southern
Cone and concluded that its effect on most subregional conflicts was negligible.^ He
found that the governments in the Southern Cone pursue their own national interests and
geopolitical goals regardless of the type of regime in power (Pittman, 1988, p. 51).^
Van Klaveren (1996) agrees that regime type is a poor predictor of changes in strategic
orientation towards a states principal partners. Mares (1997) is blunt in his rejection of
this approach.
As Norden's (1998) review of recent scholarship on consolidation of democracy
reveals, there is still a lingering concern over the remaining traces of military autonomy
in the region.^ But the newest research on how that might affect prospects for peace in
the region has proven inconclusive. As Hirst (1998) notes, there is a consensus that the
3 In the case of Bolivian aspirations to gain a salida al mar, the last serious
negotiations occurred from 1975-1978, years coinciding with authoritarian governments.
Pittman (1988) concludes that the impact of democratization on this dispute must be
regarded as nil (p. 34). With regard to the Argentina-Chilean dispute over the islands in
the Beagle Channel, and by extension, control over the sea space and claims in
Antarctica, he finds that democratization in Argentina may have been a factor in
successful resolution, but was by no means the only or principal cause.
4 It is important to note that others have challenged this assertion, at least as it
applies to Argentina-Brazilian relations. In the same volume that published Pittmans
(1988) conclusions, Russell (1988) makes a powerful argument that the emergence of
democracy in Argentina greatly facilitated agreements with Brazil that led to cooperation
in the Rio de la Plata region. However, Russells counter-argument can be questioned.
Cooperation in this region predated the return of democracy to the region the Corpus-
Itaip agreement between Argentina, Brazil and Paraguay, a landmark agreement to
jointly develop hydro-electric resources, was signed October 19, 1979. Russell treats this
agreement as an exception.
^ Recent study on the issue includes research on emerging democracy in Eastern
Europe and the newly independent Soviet bloc states (Lovell and Albright, 1997).


265
took place once civil wars had ended, and a commitment by international community to
support stability in the subregion.
Institutionalization: Organizations and Rules
Involvement by the Central American republics in the hemispheric security
organizations is considerable, and has grown during the 1990s. All states in the region
are members of the OAS, the IADB, and the conferences of the armed forces. For the
most part, all have participated in the presidential summit process and the DMAs. Also,
the countries of the region have been strong supporters of the OAS convention on
trafficking in small arms and explosives.
There are noteworthy differences, however. El Salvadors participation has
included a willingness to host several important hemispheric conferences of particular
note are the second conference of defense university directors in 1997 and the follow-up
conference on CSBMs in 1998, both held in San Salvador. In addition, El Salvador
hosted an important regional conference with UNESCO, the Armed Forces Forum for a
Culture of Peace and Nonviolence, addressed later in this chapter.
By contrast, both Honduras and Nicaragua have been less active. Nicaraguan
participation in the armed forces conferences was only renewed in 1997, after lapsing
during the Sandinista era. Nicaragua, although a founding member of the IADB, was
absent for most of the 1990s. 14 in addition, Nicaragua did not have a representative at
the 1998 DMA, although its nonparticipation was due to the intensity of reconstruction
14 The Nicaraguan delegate departed in 1989 once it became clear that its
neighboring states intended to declare Nicaragua as a threat to the inter-American system
during the Boards strategic estimate process.


220
With this approach, we expect to explain the significant differences in cooperative
security behavior that emerged in these subregions from the analysis in the previous
chapter. If puzzles still persist, we will turn to alternative explanations in the conclusion
of this study.
Southern Cone
"Subregional Integration"
Introduction
This section begins with an analysis of the context of security competition and
cooperation to establish a baseline for analysis. Following that examination, we address
the organizations and rules and related behavior in the areas of transparency/verification,
burden sharing and defense posture.
The historical pattern of security competition and cooperation in the subregion
was comprehensively reviewed in Chapter 3. Competition between Argentina and Brazil
over the Rio de la Plata; between Argentina and Chile over Patagonia, Antarctica and the
South Atlantic sea lanes; and between Chile and Peru over the Arica desert area stand out
as the major dynamics driving security behavior in the Southern Cone. The subregional
balance of power that emerged in the 19lh century has been sustained for most of this
century by threats and counter-threats between the principal antagonists and by informal
ententes between Brazil and Chile and between Argentina and Peru.
Historically, this subregion offers a clear case of the power-constraining-power
mode of conflict management, marked by distrust and lack of cooperation outside of
strategic balancing. The major security risk in such an environment is the threat of
militarized disputes and arms escalation. Another concern is the threat that one of the


296
Institutionalization: Organizations and Rules
All the countries in the subregion demonstrate a strong commitment to
hemispheric organizations primarily the OAS Committee on Hemispheric Security, the
Inter-American Defense Board and College and the DMAs. This is most marked in
Colombia, which has maintained an important presence in the Board and College.
Colombia held leadership positions in both organizations during the 1990s, endowed a
chair at the College, commanded the MARMINCA headquarters in the Central American
demining effort and hosted the third DMA. In all of those organizations, Colombia
maintained a posture clearly supportive of democracy and human rights, as we shall see
in more detail later.
The principal problem in the subregion, terrorist acts by guerrilla units funded
largely by the drug trade, has elicited some cooperative action. Peru hosted the inter-
American terrorism conference that led to the convention described in Chapter 4. There
is also strong cooperation in legal regime established by CICAD, again as presented in
Chapter 4.
Subregional institutions and rule making in the Andean Ridge suffer by
comparison with the other regions. However, two bilateral cases stand out cooperation
between Colombia and Venezuela over transnational concerns and the recently signed
peace agreement between Ecuador and Peru.
As a result of a 1991 agreement to reactivate border coordination, which had lain
relatively dormant since an earlier agreement in 1978, Colombia and Venezuela began an
extensive phase of institution building for cooperative security. Meetings occur twice a
year between the chiefs of each of the armed forces. Lower level meetings, between local


260
Nevertheless, the legacy of U.S. involvement in the region and a tendency to look
to U.S. assistance when domestic resources are insufficient mean that the United States
continues to be an important security "externality" in the subregion (Mares, 1997). The
presence of the United States can spur cooperative activity. For instance, as the threat of
Soviet sponsored intervention declined, the United States took a more supportive role to
Central American initiatives and helped to underwrite a number of cooperative endeavors
which strengthen cooperative security. 12 But competition for U.S. assistance can also
trigger conflict between states of the region.
In addition to this competitive dynamic, the subregion also has a history of
security cooperation and a legacy of security institutions. The impressive institution
building involving the security and defense systems of Central America dates back many
years, embodied in CONDECA in the 1960s, as reviewed in Chapter 4. But the most
significant institution building occurred in the 1980s, with the onset of the peace process
initiated in 1983 with the first meeting of the Contadora group (Mexico, Panama,
Colombia and Venezuela). That first meeting did not develop specific military measures
between the countries of Central America, but it did issue an appeal "to engage in
dialogue and negotiation so as to reduce tension and lay the foundations for a permanent
atmosphere of peaceful coexistence and mutual respect among states" (Child, 1996). The
immediate concern of these countries was to forestall direct U.S. intervention in Central
America (Romero, 1998). The second Contadora meeting, also in 1983, produced the
* 2 The evolution of the treaty and the important role that CSBMs played are ably
reviewed by Child (1996).


286
related exercise, denominated Coastal Forces 98, included Guatemala and El Salvador,
along with the United States (El Salvador 1998 CSBM Submission).
More explicit international counterdrug coordination (beyond U.S. bilateral
arrangements) has not been reported or is unavailable. A series of subregional meetings,
which have taken place annually since 1991, have recommended the adoption of several
concrete mechanisms, ranging from coordinated air and ship control measures to
intelligence sharing, demonstrate a commitment to a subregional approach that, to date,
has apparently not bom fruit. Cooperation is still primarily limited to bilateral programs
with the United States and information sharing networks among the regional police
forces. (Guatemala 1997 CSBM Submission).
Other examples of independent activity, consistent with the institutional focus on
criminal activity, include combined naval operations in the Gulf of Fonseca by El
Salvador, Honduras and Nicaragua (El Salvador 1996 CSBM Submission). Cooperation
in the Gulf is important since it has been the site of much security competition between
the countries that border that body of water. In addition, there have been a number of
coordinated actions between Honduras and Nicaragua on their common border.
Honduras reported (1996 CSBM Submission) four patrols with Nicaragua in 1995 to
protect populations, fight crime, protect coffee, and avoid fishing problems. Honduras
also conducted an airborne exercise "Francisco Morazon" which included regional
participants (El Salvador 1998 CSBM Submission).
In addition to these contacts between the militaries of the region, substantial
police activity is reported as declared data under individual CSBM submissions. For
instance, Guatemala reports a series of meetings in the area of vehicle theft that


32
design research questions that ably weigh the available evidence. In cases where
institutions appear to succeed, success might in fact be due to institutional effectiveness
or U.S. support. On the other hand, the failure of an institution to constrain the United
States says little about the impact of institutions on state behavior in less extreme cases of
power asymmetry.
One way to escape this complication is to drop down one level of analysis to the
subregional level. Within the geographical confines of Central America, the Andean
Ridge and the Southern Cone power is distributed more evenly. At this level, sharp
distinctions appear in the behavior patterns predicted by realism and institutionalism.
This becomes evident in the analysis of the subregional case studies in Chapter 5. At
this point however, it is useful to consider what a cooperative security regime might
consist of, in order to better focus discussion of the two theoretical paradigms.
Cooperative Security Regimes
The purpose of this section is to describe an ideal type that represents a potential
end point for the developing set of security institutions in the region. We begin with a
discussion of cooperative security, and then develop the constituent parts of what would
make up a cooperative security regime. The end result of this discussion is a set of
variables that can be operationalized and used to measure the degree to which such a
regime might be evolving in the hemisphere.
To develop the concept of cooperation security, we will rely on the work of Jarme
Nolan (1994), Antonia and Abram Chayes (1994), Joseph Nye (1991), and Ivo Daadler
(1992). According to Nolan et al. (1994), cooperative security is designed to ensure that


358
Russett, Bruce, John R. Oneal, and David R. Davis. 1998. The Third Leg of the
Kantian Tripod for Peace: International Organizations and Militarized Disputes
1950-85. International Organization 52 (3): 441-467.
"Salvadoran Defense Minister Denies Tension at Border," La Prensa Grfica (San
Salvador), 1 July 1998, Foreign Broadcast Information Service, 2 July 1998.
Sanderson, Steven E. 1992. The Politics of Trade in Latin American Development.
Stanford, CA: Stanford University Press.
Schelling, Thomas. 1963. The Strategy of Conflict. New York: Oxford University
Press.
Schmitter, Phillipe C. 1971. Interest Conflict and Political Change in Brazil. Stanford,
CA: Stanford University Press.
"Senate Approves Armed Forces Modernization Law." Buenos Aires TELAM (Buenos
Aires), 19 March 1998, Foreign Broadcast Information Service (on line), 20 March
1998.
Selcher, Wayne A. 19992. Brazil and the Southern Cone Subsystem. In G. Pope
Atkins, ed., South America into the 1990s: Evolving International Relations in a
New Era. Boulder, CO: Westview, 87-120.
Sereseres, Caesar D. 1998. Central American Regional Security. In Jorge I.
Dominguez, ed., International Security and Democracy: Latin America and the
Caribbean in the Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh
Press, 211-220.
Skidmore, Thomas E. 1988. The Politics of Military Rule in Brazil, 1964-1985. New
York: Oxford University Press.
Skocpol, Theda. 1985. "Bringing the State Back In: Strategies and Analysis in Current
Research." In P. Evans, D. Rueschemeyer and T. Skocpol, eds., Bringing the State
Back In. New York: Cambridge University Press, 3-37.
Snidal, Duncan. 1985. The Limits of Hegemonic Stability Theory. International
Organization 39: 579-614.
Snidal, Duncan. 1991. International Cooperation Among Relative Gain Maximizers.
International Studies Quarterly 35: 387-402.


170
Table 4-3: Comparison of the Santiago and San Salvador Declarations
Sector
Santiago -1995
San Salvador -1998
Observations
Trans
parency
Prior notification of
military exercises.
No new
developments.
Exchange of information
regarding arms transfers
and military
expenditures.
Stimulate studies to
establish a common
methodology to compare
defense expenditures in
the hemisphere.
Request that information
be provided to the OAS.
Beginning process
to refine UN
registries.
Exchange of information
regarding defense
doctrine and policy.
Adds detail to the
documents: information
on organization,
structure, size and
composition of armed
forces.
Adds specificity.
Verifi
cation
Exchanges of
observers/students and
visits to military
exercises and
installations.
No new
developments.
Burden
Sharing
Activities to prevent
incidents and increase
security for transport by
land, sea and air.
Develop a program of
cooperation to deal with
sea transport of nuclear
waste.
Adds new
dimension
Cooperative programs for
disaster relief prevention
and remediation.
No new
developments.
Improved
communications in
situations involving
border problems.
Identify and develop
activities that promote
cooperation between
neighboring countries in
their border regions.
Add cooperation to
injunction to
develop
communications
systems.
High-level meetings on
the special concerns of
small island states.
Support small island state
security, including
economic, financial and
environmental concerns.
Adds new
dimensions.


CHAPTER 1
COOPERATIVE SECURITY IN THE AMERICAS
Principal Themes
This is a study of the patterns of security cooperation and competition in Latin
America since the end of the Cold War. Its focus is on changes in security behavior in
the 1990s, or the process of moving beyond praetorianism and the doctrines of
counterinsurgency" (Child, 1996, p. 11), which have characterized the hemispheres Cold
War alliance system for the past half century, to a new mode of conflict management
based on security cooperation and military transparency. The theoretical tools used to
study this process derive from realist and institutionalist traditions in international
relations, the first based on power and the second on international organizations.
Many observers argue that a new security regime is emerging in the Western
Hemisphere, based on an understanding that the new threats to hemispheric security are
largely transnational in character and require interstate cooperation. Consensus on how
to jointly address these new security challenges remains elusive. But despite the absence
of central direction, an astounding array of transparency measures and cooperative
endeavors in the security field has arisen to capture the imagination of policy makers
throughout the region. The procedures and mechanisms to ensure a secure, prosperous
and democratic region are in evidence all around us only awaiting final agreement on
organizational architecture to transform disparate intentions into cohesive, functional,
and comprehensive security arrangements (Downes, 1996, p. 8).
1


53
Grieco does not believe that relative gains are a significant problem in strongly
asymmetric power situations (Hasenclever, et al., 1997). To gain a better idea of what
patterns of security cooperation and conflict might be under that condition, we return to
Waltz. The main variable left after Waltz's depiction of the international system is the
distribution of capabilities within that system. Waltz focused on the different behaviors
expected under conditions of bipolarity and multipolarity.24 Bipolarity was deemed
more stable, because of the clarity that such a system offers and the subsequent
predictability of great power behavior. Interventions by the United States in Central
American and the Caribbean during the Cold War are consistent with this theoretical
model.
Great powers in a bipolar world also engage in cooperative behavior with smaller
states, as great power competition for influence and positional advantage penetrates local
conflicts throughout the globe. Cooperative arrangements entail some level of risk, since
a prospective partner could defect once conditions change, but great powers can accept
risk in their relations with smaller states since local, tactical reversals do not affect the
overall balance of power.25 Smaller states do not have this luxury (Waltz, 1979, p. 194),
and hence would be unlikely to cooperate with each other in the absence of great power
leadership. Hence, the Cold War pattern of security behavior was marked by security
24 Unipolar systems were not specifically addressed by Waltz. But we can infer
from his discussions on the disadvantages of world government that he would have
predicted them to be unstable because a powerful center invites revolution.
25 The longer time horizon of large states with little direct threat to survival from
smaller states is an argument that appears in numerous works. See Stein (1990, Chapter
4, esp. 109-110) for a representative argument.


271
The belief that the armed forces should play an integral role in the regional model
of democratic security has taken shape in a series of meetings that have occurred under
the auspices of another regional institution -the Central American Military Forum for a
Culture of Peace. In two region-wide conferences under the auspices of UNESCO, the
first in San Salvador (1996) and the second in Guatemala (1998), the forum has promoted
the involvement of the armed forces in the peaceful resolution of conflict, the promotion
of human rights and support of sustainable development through a program of civic
works.
The meetings defined a culture of peace and nonviolence as:
a set of ethical values, practices, customs and norms of behavior based on
respect of life, human dignity, law, rejection of violence, equality of men
and women and supportive of principles of democracy, justice, solidarity,
tolerance of domestic and subregional groups regardless of ethnic,
religious, cultural, social of ideological differences. (CACAF, 1998, p. 12)
Clearly, this set of values is strongly linked to the democracy regime described in Chapter
4.
In addition to these subregional institutions, a network of largely bilateral
agreements buttresses efforts to forge cooperative security rules. Three examples stand
out: (1) an agreement of cooperation and professional exchange between the Nicaraguan
Army and the armed forces of El Salvador in Managua, signed by the defense leaders in
1995; (2) the formation of a binational commission comprised of El Salvador and
Honduras to deal with the findings of the International Court of Justice border ruling; and
(3) an agreement reached late in 1990 between Honduras and Nicaragua to conduct


102
outlet to the Pacific.^ But the major problem facing Chile was an expanding and
prosperous Argentina (Burr, 1955, p. 56). Argentina marshaled support from other
countries, while Chile looked for potential allies in Ecuador and Colombia, who were at
odds with Peru over the territory in the Amazon suddenly of strategic interest due to the
rubber boom.8
Aside from this military-diplomatic maneuvering, the Argentina-Chilean rivalry
generated an expensive arms race. The heavy drain of resources ultimately convinced the
two countries to enter into the Pactos de Mayo of 1902, solidifying the balance of power
system in South America. At its heart lay rivalries between Brazil and Argentina,
Argentina and Chile, and Chile and Peru. The independence of the smaller states of
Bolivia, Paraguay, Uruguay and Ecuador were also fundamental features of the system.
What of the other regions of Latin America Central America and the Caribbean
Basin? Unfortunately, Burr did not address those rigorously, although he does claim that
"the inter-relations of the Central American states provided a classic example of the
operation of the balance of power principle" (Burr, 1955, p. 39). In Central America, this
dynamic came into operation after resolution of the regional civil wars in 1906 and 1907,
7 Interestingly, it was in part Bolivia's land-locked status that encouraged it to
initiate conflict with Paraguay in the Chaco, which might have provided access to the
navigable river system to the Atlantic. True to the balancing pattern, Chile and
Argentina, as an extension of their own security competition, manipulated action in both
countries during that conflict (Kelly, 1997).
8 Chile sold outmoded arms and munitions to Ecuador, arranged for academic
exchanges in Quitos military academy and sent a delegation to explore cooperation with
Bogota. Chile also invited Paraguayan youths to enter Chilean military and naval
academies (Burr, 1955). Also, Martz (1997) covers Chile's early influence with the
Colombian military.


70
variety of approaches. Wendt (1995), for instance, identifies postmodernists (Ashely,
Walker), constructivists (Adler, Kratochwal, Ruggie ... Katzenstein), neo-Marxists (Cox,
Gill), feminists (Petersen, Sylvester) and others (Wendt, 1995, pp. 71-72) as belonging
to critical international relations theory. The approaches are united by a belief that
structures are sociological constructs that shape actors identities and interests, rather than
just behavior.
Hasenclever, et al. (1997) usefully divide the reflective approach into weak and
strong cognitivists. Weak cognitivists believe that institutions can change preferences
through learning gleaned from the experience of cooperation. Strong cognitivists believe
that the institutions themselves are deeply embedded constructs that form actors'
identities.3 8 Both of these versions of institutionalism delve into the ideas that statesmen
use to form policies regarding security competition and cooperation. New ideas can
cause normative change and lead to corresponding changes in behavior (Goldstein and
Keohane, 1993). Thus, the penetration of new ideas pertaining to appropriate roles for the
armed forces under democratic regimes could give rise to strong proscriptions on use of
force and restrictions on military activity.39
38 See March and Olson (1998) for a recent review of this approach.
39 This argument is tangentially related to the debate over the effect of democracy
on interstate conflict. The observation that democratic dyads tend not to fight one
another has been thoroughly studied, although there is still debate as to whether common
interests or common polities explain the phenomenon (see Chan, 1984, and Farber and
Gowa, 1995, for opposing views). Within the camp of those that believe democracy
makes a difference, there is disagreement over whether democratic norms of peaceful
resolution or democratic mechanisms of leader accountability explain the phenomenon
(Starr, 1997). Recent work on democratization shows that it may positively correlate to
conflict (Mansfield and Snyder, 1995).


327
regime type and "military economics" (military expenditures and arms production) was
weak (p. 201). They did find, however, that "arms production has a higher level of
correlation with the degree of political power exercised by the armed forces than with
budgets" (p. 225), pointing to a need to understanding military prerogatives regardless of
regime type.
Theory building in this area is incipient; little work has been directed at
specifying links between the two levels in the area of security and in determining when
those links might matter. But there is much related work: a vast body of literature that
addresses the impact of domestic institutions on foreign policy on the one hand, and the
bargaining process that goes on between civilian and military on the other hand. Building
bridges between these two schools of thought might prove productive.
After Skocpol (1985) triumphantly announced the "return of the state" a new
debate on the merits of the institutional approach to understanding international relations
exploded in the academic literature (Almond, 1988). The new institutionalists borrowed
from the work of March and Olson (1984), among others, to argue that institutions at the
domestic level contributed to a better understanding of state preferences. This new
paradigm was consistent with productive work already ongoing in a wide variety of fields
and found immediate application in understanding how foreign economic policy was
determined (Lake, 1988; Ikenberry, et al., 1988). Latin Americanists used the
institutionalist focus to analyze development strategies (Adler, 1988), revolution (Dos
Santos, 1970), bargaining with multinational corporations (Evans, 1992), bureaucratic
authoritarianism (O'Donnell, 1973; Stepan, 1978) and regime change, both breakdowns
of democracy (Linz and Stepan, 1978) and the bargaining between civilian and military


153
the critical role of the armed forces in defense matters, affirm the norm of civilian control
of the military, advocate increased transparency of defense matters, adopt confidence and
security building measures, and promote greater defense cooperation in peacekeeping
operations and the fight against narcoterrorism.27
In the working groups these themes were developed in detail. Several
observations stand out. Transparency measures specifically addressed included the UN
arms transfer registry, communication in border areas and prior announcement of bilateral
and multilateral exercises. Burden sharing themes specifically addressed included
peacekeeping, demining, natural disasters and the fight against narcotrafficking -
although the extent of military involvement in the drug fight was the subject of debate
and disagreement. Reductions in defense postures was not a popular theme in fact the
common sentiment expressed was the necessity to maintain defense establishments to
ensure national security but the generalized reductions that had been ongoing for
several years were recognized, and explicitly connected to progress in achieving military
transparency.
U.S. involvement was critical to this initiative and lent special support in the area
of transparency. The U.S. major policy initiative was the presentation of its defense
white paper the annual Defense Department report to Congress. By 1999, five other
countries in the hemisphere had published white papers (Argentina, Brazil, Canada, Chile
27 a term which clearly denotes a link between the drug trade and guerrilla
movements a position that still excites heated debate among some, but is increasingly
accepted.


184
Although military burden sharing in humanitarian areas (demining and natural
disasters) has been accepted, efforts to include burden sharing measures in other areas
have been much more conflicted. For instance, peacekeeping has been kept out of OAS
resolutions because of the contention of some countries that this issue belongs to the
United Nations and not with a hemispheric security regime. This is the oft stated position
of the Mexican delegation at the OAS a point made when an OAS peacekeeping force
was considered for Haiti and Guatemala. The memories of the U.S.-led intervention in
the Dominican Republic still shape the debate in the OAS.
Peacekeeping operations by member nations have been encouraged in other
forums. Both presidential summits and the first two DMAs include discussions of
peacekeeping. The general trend in these discussions, however, leaves it clear that a
limited supporting role is envisioned for hemispheric security organizations providing a
venue for exchanges of information and "lessons learned." There is no support for an
institution that might mirror the work of the United Nations at the hemispheric level..
Thus, there is no specific set of rules at the hemispheric level that coordinates
peacekeeping operations to facilitate burden sharing between countries of the region. All
that exists is vague support for peacekeeping operations within the UN framework. This
lack of specificity clashes with the long tradition of involvement by many countries in the
region.
If at times the debate over hemispheric rules in the area of peacekeeping has been
strident, counterdrug operations have proven to be even more conflictive. Numerous
OAS resolutions have identified illegal drugs as a threat to hemispheric security and have
called for cooperation in the fight against drugs. In fact, a full-blown counterdrug regime


174
3. practice restraint in arms transfers to prevent excessive and destabilizing build
ups
4. prevent transfers of arms to groups seeking to destabilize governments and
such groups safe haven
This resolution includes an important qualifier that recognizes the need to take into
account "the specific characteristics of the each region and in accordance with the
principle of undiminished security at the lowest level of armaments. In recognition of
the regional context that affects arms build-ups, there are no specific provisions in these
resolutions. However, within subregions, there is strong evidence that arms control can
work through hemispheric institutions.
In 1996, General Assembly Resolution 1415, Cooperation for Hemispheric
Security again called on member states to initiate a process of consultation to limit and
control conventional weapons. At the hemispheric level, however, only one weapons
system has provided a near consensus target for reduction: anti-personnel landmines.
This is in spite of, and perhaps in part because of, opposition of the United States to the
landmine ban.
At the initiative of Canada, the General Assembly began in 1994 a series of
resolutions on eliminating landmines. Resolution 1411 in 1996 called for the creation of
a landmine free zone in the Western Hemisphere, which was reiterated in 1997 with
Resolution 1496. Of course this effort was part of the global campaign, which
culminated in Ottawa in 1997 with the adoption of a convention to prohibit the use,
production and stockpiling of anti-personnel landmines. All members of the OAS, with
the exception of the United States, are signatories to the Ottawa convention.


154
and El Salvador). Many others had made comprehensive policy statements at
conferences sponsored by the hemispheric institutions we are reviewing.
The Bariloche and Cartagena DMAs were less comprehensive. The Bariloche
DMA took place in Argentina in 1996. The Bariloche Declaration expresses support for
consolidation of democracy, the need to cooperate in security matters, continued support
for CSBMs and an endorsement of peacekeeping operations. Transparency measures
singled out for mention included expanded military-to-military contacts, support for the
UN arms transfer registry and exchange of information in the conferences of the armed
forces.
But the major focus was the issue of civilian empowerment, training civilians to
be responsible stewards of their armed forces. The Inter-American Defense College and
the U.S. Center for Hemispheric Defense Studies (CHDS) were cited as vehicles to help
train civilians in defense-related matters. The CHDS, now into its second year of classes,
has growing participation from throughout the hemisphere. However, it was initially
viewed with skepticism by many countries. Some observations at the DMA and later at
the IADB contained little veiled critiques of U.S. attempts to advance its security agenda
through yet another security institution, hemispheric in mission, if not in organizational
character. Initial plans in the United States to eventually multilateralize the organization
have apparently been dropped.
The third DMA took place in Cartagena in November 1998. Although no major
multilateral agreements were reached at the meeting, the regions defense ministers did
agree to continue working towards military transparency and cooperation. The defense
ministers heard from many of the subregions of the Americas where security transparency


3
The field of security studies is divided as well. Realism remains the dominant
paradigm for understanding patterns of security behavior in this hemisphere, relying on
the concepts of power, national interest, and strategic balancing. The contender is
institutionalism, which posits that institutions can facilitate communication, forge
cooperative commitments, and encourage rule-based behaviors that go well beyond mere
balancing.
The basic research design borrows from Grieco (1988a), who develops a set of
contending hypotheses, derived from realism and neoliberal institutionalism, which look
both at the emergence of security institutions and the security behavior predicted by each
theory. Both theories expect states to act in their own best interest. But, can states
cooperate to reap the absolute gains implied by lower levels of conflict, reduced defense
expenditures, and burden sharing to address transnational threats? The realist approach
ranks the likelihood of significant security cooperation as very low, since states are more
concerned with relative than absolute gains and, in any case, lack credible assurance that
neighbors will respect rules restricting the use of force. Institutionalism offers more hope
- the dilemmas of distributive gains and defection can be at least partially ameliorated
through international institutions.
With hypotheses derived from these two approaches, I examine security
cooperation separately at the hemispheric and subregional levels, looking at the evolution
of security institutions and observable patterns of security behavior relevant to those
institutions. Comparison of the empirical record with explanations offered by system-
level theories based on realism and institutionalism shows the latter to offer a more
reliable framework for understanding security behavior in the region.


217
In burden sharing, the rough correspondence of regime and behavior breaks
down. There may not be as much cooperation to address transnational threats as the
United States would like to see, but there is certainly more activity than regime rules
establish. Cooperation in this area responds more closely to the U.S. security agenda
than to the cooperative security regime.
These mixed results cannot be said to offer definitive support for either theoretical
approach. The clearly visible trend of subregionalization offers a clue of where to go
next. Perhaps by unlocking the CSBM communities identified earlier, we can see what
makes security cooperation tick in the hemisphere. At the subregional level, we can
better control for the impact of the United States and allow intra-regional rivalries to
emerge. We can also can find more common ground on which to build cooperation or
conflict. In short, we can escape a number of frustrations that have made it difficult to
fully understand the patterns visible at the hemispheric level.


47
Brazil, Argentina and Chile (three countries who together account for 90% of intra-Latin
America arms transfers). Other relative gains problems associated with defense
reductions are present as well. Some types of reduction (mechanized arms, aircraft, etc.)
will fall more heavily on states with modernized forces, while others (troop ceilings,
limits on maneuvers) may primarily affect states with larger, but less technologically
advanced forces. Some states may rely more heavily on defensive systems (such as
landmines or anti-aircraft missile systems) than others. Thus reductions in some areas
may not be possible, or may be offset by increases in other areas. The net effect would be
marginal reductions, at best.
The issue of regional burden sharing raises a different question. To what extent
will a state compromise its security autonomy for advantages gained by cooperation,
accepting increased vulnerability in exchange for gains associated with efficient
allocation of regional resources to address common problems?
Regional burden sharing allows all states to use their defense assets more
efficiently. But some states will benefit more than others by the reduction of regional
threats. Clearly a state without a coastline receives no immediate benefit from
agreements to share sea-patrolling duties. In cases of more complex, transnational
threats, such as illicit trafficking in narcotics, calculating relative gains is much more
difficult. How are the benefits of reduced drug trafficking distributed among producer,
transit and consumer states?
Another relative gains problem relates to the training advantages of taking part in
combined operations. These exercises and operations require intense interactions that
provide access to state-of-the art military techniques and technology, perhaps most


79
competition should not be any less salient. If anything, security competition will be more
intense at the subregional level where fundamental security dilemmas operate and where
threats are more proximate to use Walt's terminology. Under institutionalism, the
dynamic of security cooperation should also be more intense, since neighboring states are
likely to share compatible security interests especially transnational security problems -
leading to the possibility of greater absolute gains through cooperation.
The fact that both security competition and cooperation are more intense has an
impact on subregional institutions if they are to be effective, they must be more binding
in nature than their hemispheric counterparts.
There are other reasons why two theories should work well at the subregional
level. The question of multiple linkages becomes intensified at the subregional level,
both because neighboring countries often share transnational concerns over a wide range
of issue areas and because subregional institutions are normally nested in their
hemispheric counterparts. These factors taken together mean greatly amplified
connectivity between states in a given subregion. Of course, subregions may differ in
terms of the density of their interrelations, but this factor simply adds another testable
hypothesis to a comparative analysis.
Grieco (1988a) offers an additional comparative test of the relative explanatory
power of realism and institutionalism that fits nicely within the subregional framework.
He bases this comparative test on the number of partners states prefer to include in a
cooperative arrangement (Grieco, 1998a, p. 506). According to his review of the
theories, advocates of institutionalism assert that a smaller number of participants
facilitates verification of compliance and sanctioning of cheaters. Thus, Grieco points


62
Neoliberal institutionalism, operating at the system level, accepts the neorealist
assertions that states are the most important actors of the international system and that
they act rationally in their interactions with each other. However, the assumption that the
distribution of capabilities is the most important system-level variable is sharply
questioned. Neoliberal institutionalism holds that deeply embedded institutions are as
fundamental to world politics as are the power resources of units (Keohane, 1989, pp. 8-
9). Institutions change the context in which states operate (a game theoretic notion) by
facilitating communications so that agreements can be made (a functional notion), by
providing information to reassure states that agreements are being respected, by
institutionalizing norms so that expectations converge around rule-based behaviors, and
by linking compliance to other areas of concern to states so that the costs of defection are
raised.
These institutionalized settings facilitate cooperation in two primary ways. First,
they reduce uncertainty by providing information that can be considered in a states
decision making process. Second, they reduce the transaction costs of entering into
negotiations by providing an arena for communication.
Institutionalists hold that even in the absence of changes in the international
distribution of power or the preferences of actors, information can profoundly affect
international political behavior" (Keohane, 1986, p. 19). In the area of security
cooperation, information is especially valuable because it avoids worst-case analysis
(Keohane and Martin, 1995; Jervis, 1986). In fact, Keohane (1984) sees the provision of
high quality information as a main function of these institutions.


96
After the Cuban Revolution, the United States was able to proceed on a
multilateral track only through increasing support to multilateralism (creation of new
institutions) and by virtue of the well-intentioned Alliance for Progress and military
assistance programs (providing side benefits). Hyper-sovereign Mexico, a newly
recalcitrant Brazil (before 1964) and unpredictable Argentina were already challenging
the U.S. Cold War security agenda. When the United States settled into a pattern of
interventionism along its southern flank, the game was up. By that time, its influence had
waned and restoration of the hemispheric security organizations it had largely created was
probably out of reach.
But the dynamic of institutional politics is only one piece of security cooperation
and conflict in the region. We need to begin with a look at the enduring rivalries that
shape military balancing in the region. This chapter continues with a look at the Latin
American balance of power system, then overlays the hemispheric system that the U.S.
ushered in during the years leading up to WWII. The main purpose here is to establish
the context for the study of cooperative security in the 1990s, which is taken up in
Chapters 4 and 5.
The Role of Power in Hemispheric Relations
Early Patterns
In the Americas, as in other regions, power has been the ultimate arbiter of
international relations. Alternatives have been hard to find. For instance, the role of
international law in Latin American security affairs has been a curious one, more often
honored in the breach than not. As Mares (1997) notes, "between 1826 and 1889 at least


106
to institutionalize the system were not productive.^ Ad hoc mediation often did not
resolve conflicts and in fact sometimes prolonged them, as belligerents were manipulated
by guarantor nations looking to advance their own interests. Mecham (1962) argues
that this was the case with mediation efforts undertaken by Argentina and Chile to resolve
the Chaco War between Bolivia and Paraguay efforts which spanned entire length of
one of the bloodiest conflicts in the history of the region. Peace was concluded only after
the belligerents were exhausted, and on the victors terms.
The advent of WWII dramatically altered the tone of inter-American cooperation.
The United States worked to ensure the inclusion of Latin American countries in the war
effort, both as staging areas for U.S. defense forces and as active partners providing
combat units (e.g., the Brazilian Foreign Expeditionary Force). Before and during the
war, military missions from the United States visited their Latin American counterparts to
develope consensus approaches to defense, win basing rights for U.S. forces, administer
rearmament programs and establish joint training programs, information exchange
systems, and standardization agreements. Comprehensive pacts were concluded with
12 The process was ultimately completed with the establishment of The American
Treaty on Pacific Settlement (Pact of Bogota) negotiated in 1948 to establish a
mechanism for peaceful resolution of conflict. The Pact stands as the classic example in
the region of institutional overreaching (Mecham, 1962). Representing the culmination
of years of negotiations in successive meetings of the regions foreign ministers, the
treaty requires peaceful resolutions of all disputes and establishes a comprehensive
mechanism for that purpose. In the event that the good offices of a mediator do not
suffice to reach a peaceful resolution, disputes are to be forwarded to the International
Court of Justice. Because of the features of compulsory referral and binding
commitment, it has never been fully ratified. Perhaps more successful was the Inter-
American Peace Committee, which took up a number of conflictual situations. But after
it was reorganized during the 1967 OAS reform, not one case was brought before it
(Baena Soares, 1994).


108
Nevertheless, the United States would find in the Board a convenient venue for
expressing a commitment to multilateralism, which would became an important tool of
assurance in the years immediately after the war.
Concerned that they had been shut out of the Dumbarton Oaks discussions, Latin
American states were anxious that the inter-American System be preserved in the coming
post-war order (Baena Soares, 1994). There was broad acceptance of the system of
consultation, developed as a result of U.S. prodding, as well as in other
accomplishments. 14 ¡n return for Latin American support at the San Francisco
conference, the United States agreed to preserve the regional system (Mecham, 1962).
Gradually, the three important institutions of the IASS took form: the Inter American
Treaty for Reciprocal Assistance the Rio Treaty (1947), the OAS (1948) and the IADB
(1942).
As work on a new collective security arrangement began, the United States
intended to marginalize Argentina, now under Juan Perons leadership, based on its
limited contribution during the war. At the insistence of the other Latin American states,
however, the United States made concessions that permitted Argentine participation
14 The American Conference for the Maintenance of Peace (1936) called for such
a system, which was approved at the 8th International Conference of American States in
Lima (1938).


54
competition between the United States and the Soviet Union and security cooperation
within each allied camp.
A strong current in the theoretical literature develops this point further. Olson
(1964) had already shown that under special conditions the provision of collective goods
may be possible notwithstanding strong incentives to free ride. If one state, or a small
group of states, determined that the provision of such a good provided private benefits
that outweighed the costs of provision, then a collective good could be produced. The
great powers would bear the costs.
Working from this framework to address problems of international political
economy, Krasner (1976) and Gilpen (1987) accomplished much theory-building in the
area of hegemonic stability and produced strong hypotheses and a compelling research
agenda that proved very popular for a time. Under hegemonic stability theory, a large
power establishes rules for cooperative behavior for its own benefit and bears the costs of
maintaining those rules. This approach is consistent with Walts view that great powers
can accept the costs of short-term tactical reversals.
In recent years, this approach has suffered a decline, in part as a result of its own
overreaching. As Lake (1993) points out, a great power or small group of great powers
may not necessarily choose to exert its influence. In any case, great power behavior can
be either benevolent (tolerant of free riding behavior) or coercive (payment extracting).
Coercive behavior will depend partly on the resources available for coercion and the
likelihood that coercion will be effective.
Even if coercion can be cost effective, however, it does not follow that a hegemon
will adopt coercive practices. Mastanduno (1997) demonstrates that a hegemon might


233
established a routine annual meeting between the two high commands of their armed
forces.
In light of this strong institutionalization between Argentina and Chile and Chile
and Peru and the even more significant activity between Argentina, Brazil and the
buffer states of Uruguay and Paraguay it is instructive to note the relative absence of
institutionalization of security relations between countries once considered informal allies
in the balance of power game of the Southern Cone. Argentina and Peru conduct bilateral
meetings, as do Brazil and Chile (Brazil 1997 CSBM Submission), but by comparison
these institutions have seen relatively little evolution. Institutionalism in the Southern
Cone does not follow traditional power axes, but rather cuts across them.
There are three additional points to be made. First, the subregional security
organizations function independently of U.S. influence. The United States is not a
member or sponsor of any of the organizations nor privy to their operations. Second,
security organizations have evolved which directly address the principal security
problems of the Southern Cone, despite the potential for unequal distributive gains. The
Strategy Symposiums aim to overcome security competition between states in the
Southern Cone and bind the region with a common strategic purpose. The Trilateral
Border Commission addresses the transnational security threats along the common border
of Argentina, Brazil and Paraguay. Third, the centrality of Argentina in almost all of
these institutions offers strong evidence that Argentina has opted for an institutional route
that leads to security cooperation, despite its relative disadvantage in terms of military
capability.


242
With the important exception of progress eliminating the threat of weapons of
mass destruction from the Southern Cone, there have been few important regional arms
control agreements in the region. In fact, as we shall see, military modernization has
proceeded apace, especially in Brazil (Dominguez, 1998) although at levels below those
seen in the 1970s and early 1980s.
However, this does not mean that there have not been significant developments in
this area, such as increased contacts along the border, redeployment of forces away from
border regions, development of doctrine that de-emphasizes the prospects of subregional
conflict and changes in training priorities that emphasize non-confrontational roles for the
armed forces of the region.
Defense "white" books have been generated by Argentina, Brazil and Chile.
Chile reported this measure in 1995, briefing the delegates at the CSBM conference in
Santiago on the contents of the book and informing them that, as a further measure to
enhance transparency with Argentina, Chilean officials had explained the book and
process to their Argentine counterparts (Committee on Hemispheric Security, 1997).
The document itself contains details on the budget, the budget process and
defense doctrine. That doctrine does not contain any startling new approaches to
security, still basing national strategy on the potential for conflict citing "hypotheses of
conflict" as the starting point for the design of defense forces. Chilean doctrine does
accept that the possibility of conflict is remote at this time and that "hypotheses of
collaboration" also exist (Chile, 1998, p. 37). However, the stated Chilean view on the
potential of CSBMs to foment such collaboration and overcome the possibility of future
conflict reveals a strictly realist conception of international relations. The purpose of


9
empirical settings are examined. Chapter 2 also develops a working definition o
cooperative security, with factors that can be operationalized for the detailed analysis to
come. Institutional factors relate to organizational assets (administrative capacity, issue
agendas and linkages) and behavioral rules (degree of bindingness and verifiability).
Cooperative security behavior is measured in terms of military transparency, reductions
in defense postures and combined security operations (burden sharing).
An historic case study the institutional politics of the IASS during WWII and its
aftermath is the subject of Chapter 3. A convergence of interests, an uncomplicated
security agenda, and the interests of United States and its ability to offer assurances and
side benefits to key member states help explain the institutional design of the IASS and
the cooperative security behavior characteristic of that epoch. When the context changed
in ways that complicated interstate relations, divergent interests led to a process of
institutional decay.5 Institutional balancing replaced institutional bandwagoning.^
This historical review offers a case study of the enduring political processes present in
hemispheric security relations, showing when cooperation is likely and when it is likely
to fail.
5 This general finding is consistent with Desch (1998), as well as a number of
other cooperation pessimists. But I do not argue that hemispheric institutions are
simply a tool of the United States, as Desch does. The bargaining dynamics are
substantially more complex.
^ Corrales and Feinberg (1999, p. 8) employ this term, which they credit to
Hurrell. U.S. acts of reassurance helped maintain "institutional bandwagoning" towards a
Free Trade Area of the Americas in the early 1990s. But with the loss of fast track
authority and a cooling of enthusiasm after the Mexican peso crisis, dynamism has
shifted to subregional agreements.


TABLE OF CONTENTS
page
ABSTRACT iv
CHAPTERS
1. COOPERATIVE SECURITY IN THE AMERICAS 1
Principal Themes 1
Recent Trends in Hemispheric Security 11
The New Politics of Cooperative Security 20
2. THEORY 30
Introduction 30
Cooperative Security Regimes 32
Contending Paradigms 39
Contending Hypotheses 74
Operationalization 87
3. ASYMMETRY AND SECURITY 94
Introduction 94
The Role of Power in Hemispheric Relations 96
Balancing at the Subregional Level 125
Conclusion 129
4. HEMISPHERIC COOPERATIVE SECURITY INSTITUTIONS 134
Introduction 134
Institutional Evolution 136
Regime Rules 158
Regime Effectiveness 197
5. SUBREGIONAL COOPERATIVE SECURITY INSTITUTIONS 218
Introduction 218
Southern Cone 220
Central America 258
ii


115
transportation, telecommunications systems, war materiel, military organization and
training (including language instruction at defense academies); military exchanges; and
multilateral meetings of the high commands of member services (Mecham, 1962). In
fact, many of these steps were ultimately implemented with U.S. sponsorship.
Partly because of domestic pressure in many of the countries where U.S. bases
had been established, the United States agreed to close them down after the war, even in
the face of growing security competition with the Soviet Union. This measure offered an
important display of reassurance to Latin American countries. The United States
bolstered Latin American sovereignty and demonstrated a commitment to
multilateralism. The Board, not the United States, would provide the basis for a future
system of bases should it be called for.
U.S. influence remained high during the early years of the Cold War. In response
to the Korean War, U.S. Military Assistance Programs (MAP) were implemented to build
up regional military capacity for hemispheric defense. Military assistance agreements
with key countries were accompanied with secret plans that pledged U.S. materiel support
in exchange for complementary military operations (Grossi, 1997). Some countries in the
region responded with extra-hemispheric assistance to U.S.-led efforts during the Cold
War; worth special mention is the deployment of combat units from Colombia to the
Korean Conflict (Martz, 1997). The United States enjoyed a "special relationship" with
Brazil; in fact, some observers characterized Brazilian foreign policy in this epoch as
"sub-imperialistic," a surrogate for U.S. power in the Southern Cone (Selcher, 1992;
Kelly, 1997).


61
analysis, four basic forms emerge: structural, which links regime creation and
maintenance to hegemonic power (Gilpin, 1987); functional, which claims that regimes
play a facilitating role that reduces the transaction costs associated with formulating
agreements (Keohane, 1984); game theoretic, where regimes help structure the context in
which states make strategic choices in pursuit of their preferences (Stein, 1990); and
cognitive, which is strongly related to the conditioning role that norms/ideas have on
behavior (Ruggie, 1983 and Haas, 1964).
Institutionalism, as currently employed (Keohane, 1989), contains strong traces of
this prior work. Its principal characteristic is its view that institutions have independent
causal significance rather than simply reflecting power relationships.32 Thus, hegemonic
stability theorists, who see institutions simply as creations of powerful states, are not
liberal institutionalists. However, the functional notion that institutions can help states
reduce costs associated with forging cooperative agreements, as well as the game
theoretic concept that institutions affect pay-offs for cooperation and defection, are
central to what Keohane (1989, 166) calls the rationalistic study of institutionalism,
commonly referred to a neoliberal institutionalism. Cognitive regime theorists have
contributed to what Keohane refers to as reflective institutionalism, which is distinct
from neoliberalism in that it operates below the system level, where unit-level processes
(such as the formation of preferences) are important variables.
32 As Jervis (1999) notes, it is not enough to show that cooperation and the
presence of institutions are correlated, as they most certainly are (see work of Russett,
Oneal, and Davis, 1998). Institutionalism must demonstrate causality.


267
expenditures and holdings, including the presence of foreign military personnel;
invitations to observe military maneuvers and participate in combined operations;
strengthening communications between border units; prior reporting of military
maneuvers; and a program of visits to military installations (Guatemala 1997 CSBM
Submission).
The signing of the Central American Democratic Security Treaty in December 15,
1995 in Honduras, represented the culmination of years of effort in confidence building in
that sub-region. Its many commitments reflected the large number of CSBMs already in
place. Honduras ratified all parts 13 December 1996 and the other countries of the region
followed suit in subsequent years. The Treaty includes a renunciation of force, support
for peaceful resolution of crises, and a recognition that regional security is interrelated -
no state should adopt a military posture which threatens its neighbors (Guzmn, 1996).
The Treatys aims are high it purports to create a new model of regional
security, that is single, integral and indivisible. It includes as basic premises civilian
control of the military; maintenance of reasonable balance of forces; security of citizens
and their property; the elimination of poverty; the promotion of sustainable development;
the eradication of violence, corruption, impunity, narco-activity, arms trafficking,
organized crime and environment deterioration (El Salvador, 1998c).
What makes the Treaty more than empty rhetoric are its detailed provisions for
achieving these lofty aims. The specific provisions related to the armed forces are
summarized below, as reported by CSBM submissions from El Salvador, Guatemala and
Honduras.


177
The lack of rules and procedures for conventional arms at the hemispheric level is
probably because reasonable balances already exist in most subregions and because
further reductions can only be negotiated in subregional contexts hence the
recommendation for states to engage their neighbors directly on this issue. Here again,
the evidence points to the importance of the subregional context.
Another important component of defense postures encompasses the rules
prescribing appropriate roles for the armed forces under democratic governments. In
point of fact, these rules emanated not directly from the security regime, but rather from
what might be considered a democracy regime.
Of all the others areas of institution building and rule making in the hemisphere
that might be suitable for regime analysis, support for democracy in the hemisphere
comes closest to security cooperation in terms of the depth and breadth of the institutions
and rules involved. This effort is much more complete than those associated with
economic integration, sustainable development, or any of the other areas of potential
cooperation considered in the inter-American system. In many ways, the effort to
support democracy is stronger than the cooperative security regime described above.
While there exists no over-arching, binding commitment to foster security cooperation
between states, the democracy regime has such an instrument the 1993 Washington
Protocol to the OAS. This protocol requires that states whose democratic process is
interrupted to be considered for expulsion from the OAS. In addition, the human rights
component of the democracy regime includes a Court with broad verification


232
based on a supplemental agreement signed in 1994 (Chile 1996 CSBM Submission).
That same year, the high commands of the Chilean and Argentina Air Forces signed an
agreement to institutionalize contacts. In 1995, the Navy of Chile and Subsecretary of
Politics and Strategy in Argentina set up a mechanism to promote contacts between the
naval forces of the two countries (Chile 1996 CSBM Submission). There is still no
equivalent agreement between the two armies.
In 1994, both countries collaborated to host the OAS Meeting of Experts in
Buenos Aires. And in 1995, in anticipation of the OAS conference on CSBMs in
Santiago, Chile and Argentina founded a Standing Committee on Chile-Argentina
Security (COMPERSEG), established by the foreign ministers, the defense ministers and
armed forces chiefs (Argentina 1998 CSBM Submission). The committee held its first
meeting in May 1996, with the goal of strengthening security cooperation in areas of
common interest. (Argentina 1996 CSBM Submission). These high-level meetings led to
a variety of cooperative endeavors, including joint exercises between military units and
the development of a common methodology to compare military expenditures (Argentina
and Chile, 1998).
Chiles increasingly institutionalized military relations with Argentina have been
characterized as an interconsultation system," consisting of encounters between the Joint
Chiefs of Staff of the Armed Forces, the National Defense Staff, and the High Commands
of the Argentine and Chilean Armed Forces, as well as nongovernmental organizations
(Villagra, 1998).
Chile has also greatly increased contacts with Peru based on a 1991 agreement to
deal with issues pending from the 1921 treaty between the two countries. Peru and Chile


264
subregion. 13 The realist paradigm would expect that these factors would complicate
efforts to coordinate security. Institutionalism would expect the opposite result, since the
potential absolute gains from coordination and cooperation are high.
These transnational problems require integrated, multinational solutions.
Institutions should help by promoting cooperative security approaches to deal with illegal
cross border flows, by promoting the sharing of information, pooling of resources and
coordination of security operations. In fact, a security regime, dubbed "democratic
security" by participating states in the subregion has arisen which emphasizes cooperative
approaches to enhance citizen security. As we shall see, the regime demonstrates
extensive institution building to facilitate negotiations and counts on a number of binding
rules that prescribe certain behaviors. Administrative capacity to maintain this regime is
weak, however, leaving open the question of how effective these rules might be.
There also exist relative gains dilemmas that may further constrain the
effectiveness of the new security institutions. To the extent that economic dynamism
returns to the region once citizen security is enhanced, relative gains will be enjoyed by
the more dynamic states, particularly Guatemala but also El Salvador.
Nevertheless, there were good reasons at the beginning of 1990s to expect that
institutions might take hold: the common political will to leave behind the devastating
experience of the 1980s, the essentially status quo nature of the states in the subregion,
the common commitment to economic integration, the dramatic defense down-sizing that
13 See Torres (1998) on the recent tension between Nicaragua and Costa Rica
over the illegal immigration.


219
Institutionalism expects to see a different pattern. Subregional institutions that
supplement the hemispheric regime will strengthen mechanisms in areas that will yield
significant absolute gains. If these subregional institutions are properly resourced, strong,
binding rules will be negotiated, especially in areas where relative gains dilemmas may
make realization of absolute gains difficult. Where linkages are available, they will add
iteration and assist in reaching agreements. Finally, if strong transparency mechanisms
are present and reciprocation is facilitated, rule-based cooperative behavior will be
evident.
The case studies provide an excellent proving ground for the theories and should
give us greater confidence in the robustness of the conclusions. We will also have an
opportunity to employ the comparative method to see in what ways the variables of U.S.
presence, type of security problems, and degree of regime nesting affect the evolution and
effectiveness of subregional security regimes.
As hypothesized at the end of Chapter 2, the Southern Cone should show more
autonomy, vis--vis the United States, than either Central American and the Andean
Ridge. With regard to security problems, we will see that the Southern Cone is
characterized primarily by the traditional security dilemma that leads to state-directed
militarized disputes and arms races, while security in the Andean Ridge and Central
America is affected by transnational security problems as well as interstate rivalry. There
is linkage or regime nesting present in both the Southern Cone (principally linkage with
Mercosur) and Central America (principally with democratization). Little regime nesting
is evident in the Andean Ridge.


109
(Mecham, 1962). Once a full member of the new system, Argentine would become the
first state to assume the mantel of institutional balancer.15
The Inter-American Conference on Problems of War and Peace in the Act of
Chapultepec (1945) had affirmed the principle of collective responsibility for the security
of the Western Hemisphere, as well as the collective use of force as one mechanism to
preserve that security. The conference led directly to the Inter-American Treaty for
Reciprocal Assistance, signed in Rio de Janeiro in 1947. The Rio Treaty provides the
machinery to maintain peace and security in the hemisphere, setting up a mechanism to
bring the collective weight of its signatories to bear on a situation of aggression.
Some view the treaty as a classic collective security arrangement intended to
redress acts of aggression between states. For the hemisphere, it is indeed a remarkable
instrument, featuring automatic referral mechanisms, detailed voting procedures and
binding obligations (Mecham, 1962). Under the provisions of the treaty, the United
States can be obligated to break off diplomatic relations and sever all economic relations
with an aggressor state when two-thirds of the members so decide (Council on Foreign
Relations, 1962, p. 29).
U.S. acceptance of this provision was an important concession, an act of
assurance that may have forestalled an Argentine effort to limit the Treatys definition of
15 This is the major thesis of Grossi (1997). President Juan Pern even attempted
at one point in the 1950s to establish an Argentina-Brazil-Chile bloc that would project
Argentine power in the aftermath of the U.S.-U.S.S.R war he saw as inevitable. The bloc
failed, as did most other overt balancing attempts against the United States. Cuba would
be the only country to form an opposing military alliance and that step would require a
near-nuclear war (Mares, 1997).


319
Table 6-3 -- continued
Characteristics
Southern Cone
Central America
Andean Ridge
Coopera
tive
Security
Behavior
Trans
parency
Cuts across
rivalries.
Reveals sensitive
information.
Low levels of
implementation.
El Salvador is the
outlyer.
Colombia is transparent
due to
internationalization of
civil war.
Defense
Postures
Very little more
concerned with
power projection.
Downsizing and
strengthening of
civilian control is
ongoing.
Some saber rattling
No downsizing except
that associated with
MOMEP in Cenapa
region.
Burden
Sharing
Peacekeeping
operations
strong.
US-Argentine
connection.
Little
competition for
buffer states.
Dependent: occurs
with U.S. assistance.
Colombia-Venezuela
well developed.
Failure to realize
absolute gains.
Comparison of the subregional case studies shows that the United States did
influence cooperative security in those areas as predicted. In Central America, the U.S.
role approximated that of a security manager, while in the Southern Cone, the U.S. served
primarily as a security partner and/or rival. Comparison of the subregions in terms of
their security dilemmas shows that transnational problems may be more difficult to
resolve than traditional security challenges. Finally, comparing nesting/linkages across
the case studies shows them to be critically important in broadening the prospects for
iteration and the scope for absolute gains. Cooperative security did not always emerge in
an orderly fashion grassroots dynamism produced an unruly set of individual
measures in some cases. But if the structures and behaviors were not always perfectly
hierarchical, they nevertheless formed a coherent whole that facilitated cooperative
security behaviors in important ways.


314
and rationalization of hemispheric security institutions and preclude the creation of more
binding rules with strong verification.
Table 6-2 summarizes the key observations for security organizations, rules and
behavior at the hemispheric level in the 1990s. The institutional design suffers from
tension over the role of the IADB in the OAS, creating a division between the political
and technical-military bodies of the system. This division reflects the unwillingness of
states in the region to create a powerful regional security body that could serve U.S.
interests; it also reflects uncertainty about the role of the military in the security context
of the 1990s. This uncertainty appears in the make-up of the hemispheric security
agenda, which although multilateral and comprehensive, does not enjoy consensus on
core security threats or the role of the armed forces. This is especially true for burden
sharing. Although numerous rules exist at the hemispheric level related to burden sharing
to address transnational threats (drugs, terrorism, smuggling of small arms), few include a
military component.
Because of the divided structure and the lack of a consensus agenda, few rules are
binding and verifiable. Transparency is an exception, due primarily to its strong linkage
to the democracy regime. Finally, in the area of security behavior, three important
observations stand out. First, transparency is fairly wide spread. Second, substantial
burden sharing and some reductions in defense posture are taking place despite the
absence of strong hemispheric rules. Third, all the behavior clusters reveal geographic
patterns that center on specific subregions. This evidence led us to examine the three
subregional case studies.


345
Farber, Henry S. and Joanne Gowa. 1995. Polities and Peace. International Security
20(2): 123-146.
Farer, Tom, ed. 1996. Beyond Sovereignty: Collectively Defending Democracy in the
Americas. Baltimore, MD: Johns Hopkins University Press.
"Foreign Ministers Spar on Border Land Mine Issue." La Paz Radio Cadena Nacional, 4
December 1997, Foreign Broadcast Information Service (on line), 4 December 1997.
Franko-Jones, Patrice. 1987/8 "Public-Private Partnership: Lessons from the Brazilian
Armaments Industry." Journal of Interamerican Studies and World Affairs 29: 41-
68.
Frohman, Alicia. 1996. "Chile: External Actors and the Transition to Democracy." In
Tom Farer, ed., Beyond Sovereignty: Collectively Defending Democracy in the
Americas. Baltimore, MD: Johns Hopkins University Press, 238-256.
Gilpin, Robert. 1987. The Political Economy of International Relations. Princeton, NJ:
Princeton University Press.
Glaser, Charles L. 1997. The Security Dilemma Revisited. World Politics (October)
50(1): 171-201.
Godson, Roy and Phil Williams (1998). "Strengthening Cooperation against
Transnational Crime." Survival 40 (3): 66-88.
Goldstein, Judith and Robert O. Keohane, eds. 1993. Ideas and Foreign Policy: Beliefs,
Institutions, and Political Change. Ithaca, NY: Cornell University Press.
Gonzlez Glvez. 1999. Hacia un Nuevo Concepto de Seguidad Hemifrica.
Presentation at the Second Regional Conference on Confidence and Security
Building Measures. San Salvador: Mexican Foreign Service (Mimeograph).
"Government Denies Military Intervention Plan for Mercosur." Agenda Estado (Sao
Paulo), 29 July 1997, Foreign Broadcast Information Service (on line), 29 July 1997.
Gowa, Joanna. 1989. Bipolarity, Multipolarity and Free Trade. American Political
Science Quarterly 83 (4): 1245-56.
Grieco, Joseph. 1988a. Anarchy and the Limits of Cooperation: A Realist Critique of the
Newest Liberal Institutionalism. International Organization 42 (3): 485-507.
Grieco, Joseph. 1988b. Anarchy and the Limits of Cooperation: Realist Theory and the
Problem of International Cooperation. Journal of World Politics 50: 600-624.


360
Stepan, Alfred. 1988. Rethinking Military Politics: Brazil and the Southern Cone.
Princeton, NJ: Princeton University Press.
Tate, Trevor M. 1990. Regime Building in the Non-Proliferation System. Journal of
Peace Research 27 (4): 399-414.
Tavares, M. C. 1985. "A Retomada da Hegemona Norte-americana." Revista de
Economic Poltica 5 (2): 5-15.
Tenneriello, Bonni, with Geoff Thale and Richard L. Millett. 1996. "Unfinished
Business: Military Reform and Peace Process in El Salvador and Guatemala." In
Richard L. Millett and Michael Gold-Bliss, eds., Beyond Praetorianism: The Latin
American Military in Transition. Miami: University of Miami North-South Center,
181-205.
Tesn, Fernando R. 1996. "Changing Perceptions of Domestic Jurisdiction and
Intervention." In Tom Farer, ed., Beyond Sovereignty: Collectively Defending
Democracy in the Americas. Baltimore, MD: Johns Hopkins University Press, 29-
51.
Torres, Fernando Zeledn. 1998. Security, Agenda, and Military Balance in Central
America. In Jorge I. Dominguez, ed., International Security and Democracy: Latin
America and the Caribbean in the Post-Cold War Era. Pittsburgh, PA: University
of Pittsburgh Press, 222-245.
United States Arms Control and Disarmament Agency. Various years. World Military
Expenditures and Arms Transfers. Washington, DC: United States Arms Control
and Disarmament Agency.
United States National Strategy of Engagement and Enlargement. 1995. Washington,
D.C.: The White House.
"Uruguayan, Argentine Armies Carry Out Joint Maneuvers." Buenos Aires Telam
(Buenos Aires), 11 June 1997, Foreign Broadcast Information Service (on line), 12
June 1997.
Vaky, Virn P. 1993. The Organization of American States and Multilateralism in the
Americas. In Virn P. Vaky and Heraldo Muoz, eds., The Future of the
Organization of American States. New York: Twentieth Century Fund, 1-65.
Van Klaveren, Alberto. 1992. Latin America and the International Political System.
Jonathan Hartlyn, Lars Schoultz and Augusto Varas, eds., The United States and
Latin America in the 1990s: Beyond the Cold War. Chapel Hill, NC: University of
North Carolina Press, 21-45.


191
factors that allowed the negotiation of arms limitations and, later, the development of
military cooperation that has led to coalition operations out of theater.
Third, there are few formal mechanisms for verification and hence little sign that
reciprocity is functioning. These three points cast doubt on the institutional explanation.
Let us review each behavior cluster in more detail. There is a sort of happy
coincidence between the countries of the hemisphere in the area of military transparency
and not surprisingly a dense network of rules has been created. Verification mechanisms
are not impressive however. There are no binding measures outside of the third party
verification schemes present in nuclear accords and in some peacekeeping activities.
Both of these are special cases. There is little evidence at the hemispheric level that
verification operates effectively to overcome cooperative dilemmas in other cases.
Does this mean that verification is impossible? Not necessarily. Some declared
data such as arms transfers and CSBM reports can be cross-checked. In other cases,
comparison with open sources is possible as with arms transfers and defense
expenditures.36 Alternatively, the network of personnel exchanges, invitations to
observe exercises and visits to installations can offer reasonably strong, bilateral
verification of defense capabilities. Although this network is not yet organized in such a
way as to provide a comprehensive view of compliance, it does provide a set of snapshots
of military activity, routinely reported back to participating states. But these exchanges
are not new; they were originally promoted by the United States during the Cold War
36 See Laurance (p. 350) for discussion on self-verification and comparison to
public data. Also, Anthony (p. 121-122).


122
eliminate the overlapping jurisdictions claimed by the various security organizations and
subject them to centralized control have all failed. This issue has been guaranteed to
spark institutional balancing.
After Buenos Aires, the Board began to loose relevance. Over time some military
delegates at the Board became less connected to the formulation of defense strategy in
their home countries, as civilian authorities increasingly assumed that responsibility.
Budgets for defense activities, both at home and in international institutions, came under
constant attack. While over its history the Board produced more than one hundred
distinct organs to deal with matters of a single nature, with the objective of consolidating
the peace and security of the Continent: 1. The Advisory Defense Committee, created by
the IX International Conference of American States in 1948, within the organization
itself. To date it has never been convoked. 2. The Inter-American Defense Board,
created in 1942 by Resolution XXXIX of the Third Meeting of the Ministers of Foreign
Affairs, with a direct relationship with the American Governments and with no
organizational ties with the Organization of American States, and which to this date has
produced many studies and recommendation for consideration by the Governments; and
3. The Special Consultative Committee on Security, created by Resolution II of the
Eighth Meeting of Consultation of Ministers of Foreign Relations, in January 1962,
responsible to the Council of the Organization, which has produced reports for the
consideration of that organ ... There exists a duplication of organizations that are not
linked to each other in dealing with the topics of continental defense, which makes
difficult the accomplishment of their objectives. Moreover, in this way, defense matters
are not incorporated with the economic, social, and political problems that are the
concern of the Organization of American States, and whose co-relationship is essential
because they bear on the essence of the principles and objectives of the Organization
...Asa consequence of the foregoing, the Second Special Inter-American Conference, in
the first point of the "Act of Rio de Janeiro" states that it is necessary to endow the Inter-
American System with new dynamism and to avoid duplications and conflict of
jurisdiction among its organs... In the second point, it defines the need to modify the
functional structure of the organization of the States, as defined in the existing Charter"
(IADB, 1966).


275
programs to host exchange officers from Latin American in the regular military schools
and the School of Americas. In other instances, regional forums provide a setting to
develop personal contacts that lead to informal coordination between leaders of all levels
(El Salvador, 1997).
It is difficult to find a pattern in these contacts, beyond the high degree of
Salvadoran participation. There is little to indicate that underlying rivalries are driving
this verification pattern. What is evident, however, is the disconnect between the dense
network of organizations and rules and the actual pattern of behavior reported/observed.
Rules are not fully implemented or consistently respected. This may indicate a relative
lack of resources within the subregional institutions to perform the administrative tasks of
regime maintenance. The fact that verification is accepted when made possible by
external organizations supports this conjecture.
Of course, the disconnect may also indicate that security behavior is less than
responsive to the rules designed to guide transparency and verification. Let us now turn
to another behavior cluster to see what other evidence will bear on this subregions
pattern of security behavior.
Defense Posture
In the 1990s, huge reductions in troop strength in El Salvador, Guatemala and
Nicaragua, pursuant to the peace accords that ended the civil wars in those countries,
reduced the risk of militarized disputes between states to the lowest level in many
decades. These reductions were not simultaneous (in fact, reductions in Guatemala are
still ongoing), indicating that states were able to lower defense postures despite
transitional imbalances and relative gains dilemmas. To some extent, the reductions took


20
There is substantial evidence that elements of this vision have taken root in the
hemisphere. Institutionalized cooperation is emerging at all levels locally,
subregionally and regionally. Its constituent parts overlap in places and leave voids in
others, and mechanisms of coordination between levels are underdeveloped. Despite
these limitations, this network of security institutions is becoming an important factor in
security relations between states of the region and may be directly responsible for much
of the security cooperation observed in the 1990s. This study is designed to evaluate that
challenge to the realist paradigm.
The New Politics of Cooperative Security
Just as regional trade integration has slowed at the hemispheric level, giving way
to subregional pacts, security cooperation demonstrates a similar trend. The evidence
shows that cooperative security generally coalesces at the subregional level, especially
where security can be nested in other cooperative regimes, such as Mercosur (Mercado
Comn Sur) in the Southern Cone and the peace process in Central America. But the
security institutions that have emerged at the hemispheric level as a result of this process
are far from fulfilling the vision of Patio Mayer. Understanding of how the renewal
process actually unfolded in the IASS requires a theoretical model that can encompass the
dynamics of power, interest and institutional behavior.
The U.S.-led intervention in Haiti (1994) and the shock of the Ecuador-Peru
conflict (1995) confronted states in the region with two guideposts to the future: the first,
an example of outcomes likely under an effective hemispheric security regime and the
second, an example of outcomes likely under a failed regime. Latin American states, for
the most part, were not satisfied with either scenario. On the one hand, Latin American


283
one dedicated to peacekeeping training and the other to humanitarian operations, make up
the bulk of that activity (El Salvador 1996 CSBM Submission).
The "Fuerzas Aliadas Peacekeeping Operations" exercise has taken place in the
region since 1996. The first exercise in the series, held at the U.S. headquarters of Joint
Task Force Bravo in Honduras, included participation by El Salvador, Guatemala,
Honduras and the United States. The 1997 exercise, based at a Salvadoran facility
renovated by the United States, added participation by the Caribbean countries of
Barbados and Trinidad and Tobago.
These exercises are based on Bosnia-like scenarios, supported by computer
simulation, which train military officers from the participating countries in peacekeeping
operations by confronting them with a number of realist situations, ranging from initial
deployment to an area under a UN mandate to violent conflicts between warring ethnic
factions. The participating officers assume command and staff responsibilities in
fictional units operating under the special representative of the UN secretary-general.
The simulation exercise itself lasts about a week and is preceded by a week of
training in peacekeeping operations, with classes conducted by experts from the
international community. In addition, a distinguished visitors program runs concurrently
to the training and includes forums for the exchange of national experiences in the area of
peacekeeping, generally presented by senior defense officials from the participating
countries.
El Salvador, Honduras and Guatemala all have some operational peacekeeping
experience as well. El Salvador participated in MINURSA, in Western Sahara. Belize,
Guatemala and Honduras participated with military contingents in the peacekeeping


12
for the region. Mares counts in the twentieth century alone more than two hundred cases
where states either threatened or used military force or were the subject of such threats
or force" (Mares, 1997, p. 195). While interventions involving the United States account
for a significant portion of this number, most involve militarized disputes between
neighboring states of Latin America that bear the same characteristics of conflict
everywhere.
The twentieth century disputes that Mares (1997) cites are taken largely from the
Militarized Interstate Disputes (MID) database. While many fall short of war, as defined
by the standard 1,000 battlefield deaths, at least eight conflicts qualify by that criteria: the
Second Central American War, 1906; Third Central American War, 1907; Dominican
Republic-Haiti, 1937; Bolivia-Paraguay, 1931-35; Peru-Ecuador, 1939-1941; El
Salvador-Honduras, 1969; Argentina-Great Britain, 1982; and Peru-Ecuador, 1995.
Other disputes which were not counted came very close to the cutoff criteria. The Peru-
Colombia conflict in Leticia in 1932, for instance, resulted in 852 deaths.
Against this historical backdrop of violence and mistrust, developments in the
field of security cooperation in the 1990s do indeed present an empirical contrast. In the
Southern Cone, Argentina and Chile have participated in joint exercises involving their
naval and air forces deployed in the Beagle channel, the object of a near war in 1978.
The two countries also sent ground forces to work under Brazilian command on the
border of Peru and Ecuador as part of a military observer mission. And in 1999,
Presidents Carlos Menem and Eduardo Frei, in a reenactment of the 1899 abrazo which
led to the Pactos de Mayo in 1902, signed a declaration in the austral port of Ushaiua


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EL7JNNU5M_KTZTFX INGEST_TIME 2014-11-17T21:57:56Z PACKAGE AA00026384_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


248
exercises as well (Argentina 1996 CSBM Submission). That fact that these exercises
have been successful is testimony to the extensive contacts all three countries have
pursued in the area of aviation and air defense.
However, Brazil does face a growing problem along its Amazon border with
Colombia. The Brazilian military began to recognize that military forces would have to
be involved in a solution to the drug trafficking problem as a result of a 1991 skirmish
between a raiding party of the Fuerzas Armadas Revolucionarias de Colombia (FARC)
and a small Brazilian military outpost (Pinheiro, 1995). Over the course of the decade,
the military has become more involved involvement which has required cooperation
with the Colombian military in conflictive situations.^ This activity demonstrates that
Brazilian objections to hemispheric initiatives suggesting the use of military forces in the
counterdrug effort are based on a policy of resisting U.S. control, not disagreement with
the concept.
With regard to humanitarian operations, Argentina and Chile conduct search and
rescue exercises and related communications exercises (Argentina 1997 CSBM
Submission). Until recently, however, these were board exercises using staff officers and
gaming simulations. In September 1997, the exercise went "live" with joint maneuvers
by air and naval forces of the two countries ("Confidentiality Accord," 1998). In
addition, Brazil and Chile have conducted combined training for crews of transport
10 In November, 1998, 800 FARC attacked a Colombian police garrison 20 miles
from the Brazilian border. The FARC controlled the airfield, forcing Colombian
reinforcements to land in Brazil. There is informal coordination between security forces
of Brazil and Colombia, but no official agreement for this kind of support (Marcella and
Schulz, 1999).


213
As can be seen from the data, Argentina, Brazil and Uruguay emerge as the leaders in
numbers of CSBMs reported. A second tier emerges which includes the countries of
North America, Chile and Peru. The rest of the Andean countries plus most of Central
America make up a third tier.
However a more sophisticated approach is possible that provides a clearer picture
of the geographic pattern. Participation by dyads can be calculated regardless of whether
a given measure is bilateral or multilateral. This approach is useful because it overcomes
some of the problems with lack of thresholds in the reporting format. For instance, the
inventories contain numerous measures with the United States that do not appear in the
U.S. report. U.S. partners report a total of 186 CSBMs with the United States, while the
total number of measures reported by the United States is only 53. The United States is
the only country that underreports in this manner.
When dyads are built from the raw data, we can portray them graphically on a
map of the region. Figure 4-2 does just that. It is clear from this portrayal of the
information that the three countries of Argentina, Brazil and Uruguay are indeed very
active players. But the United States is the principal partner of all three of those
countries. Furthermore, clear CSBM communities emerge from the data: the Southern
Cone, Central America (excluding Costa Rica) and the Eastern Caribbean states. There
is also a somewhat disjointed community among the Andean countries, centered on Peru
and Colombia. But the Peru-Ecuador dyad should be considered as a separate case. All
of the CSBMs reported there dealt with attempts to maintain a cease-fire in the Cenapa
region. When that dyad is extracted, a weak Andean Community centers on Colombia.


BIOGRAPHICAL SKETCH
William B. Fullerton, a 1981 graduate of the United States Military Academy, is a
lieutenant colonel in the U.S. Army with extensive experience in Latin America. His past
assignments include tours in Argentina, Brazil, El Salvador and Panama. In addition, he
served as the executive officer to the Chairman of the Inter-American Defense Board
from 1997-1999 and an associate professor of international relations at the Military
Academy. His current assignment is as the Joint Engineer Officer, Joint Staff.
LTC Fullerton attended the Catholic University of Rio de Janeiro as an Olmsted
Scholar from 1988-1989. He completed a Masters of Arts in International Relations and
a Master of Science in Civil Engineering at the University of Florida from 1990-1991.
He was awarded the degree of Doctor of Philosophy by the University of Florida in 1999.
LTC Fullerton speaks Spanish and Portuguese.
LTC Fullerton is married and has four children. He resides with his family in
Bethesda, Maryland.
363


52
From Walt (1987), based on his theory of alliance formation,2! we would expect
to see balancing behavior predominate states will seek partners to rally against states
they perceive as most threatening. Balancing patterns should be readily observable in the
behavior clusters developed to explain the make-up of cooperative security regimes:
transparency/verification, defense postures and burden sharing.
In his empirical work, Walt deals with the broader concept of alignment, of
which alliances are a subset. Alignments need not be formalized (Snyder, 1990). The
tightness of such arrangements is the subject of much debate, but the presence of a
common enemy is probably the most important factor. 22 of course, high levels of threat
may not always induce external balancing; arming, or internal balancing is also a
rational response to threat (Morrow, 1993). In any case, alliances are still subject to the
same problems inherent in less formal forms of security cooperation.23
21 Walt, studying alliance formation in the Middle East, found that regional
balancing and bandwagoning were not closely related to superpower penetration even
during the Cold War. He determined that security behavior can best be explained by the
dynamic of competitive balancing. The study of military conflict and cooperation in the
Middle East remains attractive to security analysts. For a recent look at regional security
cooperation during the early 1990s, see Jentleson and Kage (1998), who find that realist
factors best explain the trend.
22 See Christensen and Snyder (1990) and Weitsman (1997) on the tightness of
alliances.
23 Snyder (1990) describes the alliance games as burden sharing (concern that
costs are borne equally), abandonment (fears that commitments will not be upheld when a
states security is threatened) and entrapment (fears that binding commitments will draw
a state into a conflict it would have rather avoided).


211
Countries of Central America and the Caribbean have also participated in
peacekeeping operations, most significantly in Haiti. Honduras, Guatemala and a
CARICOM Contingent rotated battalion-sized units under U.S. sponsorship to take part
in the Haitian peacekeeping effort. Because each follow-on rotational unit assumed
duties in the same sector, these countries were required to cooperate among themselves
for turnover procedures. Clearly, they also cooperated closely with the United States.
Statistically, burden sharing activities make up a small percentage of the CSBMs
reported, but their significance is considerable, as demonstrated by the detailed
descriptions above. This level of activity cannot be explained by the hemispheric security
regime, which is relatively weak, except in the areas of border contacts, transportation
coordination and humanitarian operations. Some of those priorities emerge, but other
activity cannot be effectively unexplained by the presence of a hemispheric regime.
The presence of the United States is palpable in many of the hemispheric
activities and some of the subregional ones. Support from the U.S. Southern Command
is vital to the Central American demining program, MOMEP, and many of the
subregional exercises which address natural disaster relief, peacekeeping and counterdrug
operations. Since these exercises are now largely multinational, they carry great weight in
the reporting. Consequently, it is not surprising to see the priorities of the United States
in this area show up in the data.
But, there is significant activity taking place both outside of U.S. programs and
outside of the hemispheric security regime. Much of it is subregional in nature, reflecting
a willingness of neighboring countries to work together to solve common security
problems. We will examine this puzzle more closely in Chapter 5.


208
dedicated to developing interoperability between countries in support of combined
operations. Conference systems exist for each of the armed services: the Inter-American
Naval Conference (CNI); the Conference of American Armies (CAA) and the System of
Cooperation for the Air Forces of the Americas (SICOFA). The conferences encompass
an extended cycle of technical meetings. These meetings are held all across the
hemisphere and conclude with a conference of service chiefs held on an annual or
biannual basis.40
Each of the conference cycles is organized around a common theme, announced
by member states at the conclusion of the previous cycle. The service chiefs debate that
theme at the capstone conference and vote on the adoption of related agreements
developed at the technical meetings. While some of the agreements are administrative
and internal to the organizations, most are attempts to enhance interoperability by
standardizing training, equipment and doctrine.
It is instructive to note how the focus of mandatory themes has changed over time.
Established at the initiative of the United States at the end of the 1950s, the conferences
initially focused on the threat of communism. Themes during the 1990s now deal with
the importance of civilian control of the armed forces, appropriate roles for the military
and support for democratic institutions. This agenda often looks closer to that of the
democracy regime than one supporting burden sharing.
40 See Kresho (1987) for a history of the Conference of American Armies. The
Naval and Air Force conferences are similar, although perhaps more technical in nature.


60
based on a set of neoclassical economic principles that demonstrate that
global utility is maximized by the free flow of goods ... a basic norm ...
that tariff and nontariff barriers should be reduced and ultimately
eliminated ... specific rules and decision-making procedures ... spelled
out in the General Agreement on Tariffs and Trade. 29 (p.4)
While most of regime-based analysis belongs to the field of international political
economy, the approach has a strong tradition in environmental studies and international
law. Theoretical work in the area of security regimes is much less common, partly in
recognition of the power of realism to predict interstate security behavior, although
Keohane and Martin (1995), among others, express the hope that regime analysis will
gradually invade security studies (p. 44).30 Security regimes have been discovered
in the area of strategic nuclear weapons reduction, the European military order,
prevention of nuclear war, and nuclear non-proliferation.31
Efforts to categorize the theoretical insights developed within the institutionalist
paradigm are legion. Haggard and Simmons, working specifically with the concept of
international regimes, organize variants by their theoretical component. From their
29 The application of institutionalist theory to explain international trade behavior
is perhaps more complete than in any other area of interstate relations. The scholarly
debate on the topic emerged in the context of the declining U.S. economic hegemony in
the 1970s. Institutionalists argued that what they saw to be an enduring practice of free
trade despite the U.S. decline was linked to the presence of institutions and norms.
Critics disagreed both with the institutionalist characterization of trading practices as
largely free and with the claim that such behavior was in any case due to the impact of
institutions. See Keohane (1984).
30 Axelrod and Keohane (1986) advocate a single framework for military-
security and political-economic issues (Axelrod and Keohane, 1986, p. 227). See Jervis
(1983) for the classic explanation of why security is different.
31See Mueller (1993), Tate (1990) and Brzoska (1992).


320
Contributions to Theory
Aside from a general endorsement of institutionalist approaches, this study makes
a number of potentially useful contributions to international relations theory. First, the
study helps us to better understand security cooperation in asymmetric orders. Analysis
of the interplay between the United States and Latin American countries in the context of
hemispheric security organizations contributes both to institutionalism, through enhanced
understanding of institutional political processes identified by Keohane and Nye (1989),
and to realism, especially in the area of hegemonic regime theory.
Keohane and Nye developed a new understanding of institutional politics in their
examination of cooperation under situations that they described as complex
interdependence (1989). Those situations approximated what Deutsch (1957) would
consider a political security community: a system of states with high levels of economic
interaction, information exchange and an understanding that use or threat of force within
the community was not legitimate. These conditions do not entail in Latin America and
yet the processes identified as belonging to complex interdependence are operable.
Inclusive decision-making procedures restore symmetry. Organizational forums allow
smaller states to voice their preferences. Linkages reinforce multilateralism. Contrary to
Krasner (1991), power cannot always be used effectively to dictate agendas and
determine players. Institutional balancing is possible.
When the United States chooses to promote its agenda through multilateral
institutions, the process by which power is translated into outcomes becomes essential to
our understanding of international relations. Institutionalism with its focus on
organizational structure and rules is much better equipped to understanding this process


8
renewed willingness to work with multilateral institutions. As an historical point of
comparison, previous security institutions of the hemisphere are covered in Chapter 3.
The lack of a fixed end date for the study complicates efforts to draw hard
conclusions about the patterns of security cooperation and the effect of regional and
subregional institutions and rules regarding the use of force. For instance, there is little
that can be said about the robustness of the new institutions and rules in the absence of a
significant extra-hemispheric challenge or change in power distribution globally or
regionally. Such changes are unlikely in the short term. But we can still analyze how
institutions emerge, how they reinforce norms and create rules and how they affect
behavior patterns.
The topical focus is on the use of the armed forces in the foreign policies of states
in the Western Hemisphere. Thus issues of transparency in military matters, reciprocal
verification of defense postures, arms restraint and security cooperation (involving
militaries), as well as conflict, are within the scope of the study. Security cooperation
that might be present between police forces or other branches of the government is only
tangentially discussed. Likewise, domestic aspects regarding the role of the armed forces
within states are considered only briefly.
This chapter continues with a short discussion of recent developments in security
relations in light of the end of the Cold War and the onset of globalization, followed by a
more fully developed exposition of the main argument of this investigation. Chapter 2
discusses the theoretical literature in order to develop theories and hypotheses. These
hypotheses are then situated in a research design that will test their relative explanatory
power. Specific research questions will be developed in later chapters, after the requisite


318
accord is comprehensive in scope, with links to economic development and resource
exploitation that may prove key to maintaining peace.
Table 6-3 summarizes the subregional case studies in terms of the key
characteristics identified by theory.
Table 6-3: Characteristics of Subregional Security Regimes
Characteristics
Southern Cone
Central America
Andean Ridge
Security Context
Emergence of
Mercosur.
Argentine drop in
relative power.
Absolute gains in
defense postures.
Formal integration
movement.
Reductions under
external control.
Absolute gains in
burden sharing.
Colombia-Venezuelan
economic cooperation.
Transnational threats.
Absolute gains in all
areas.
Organi
zational
Features
Forum
No central
forum;
numerous
overlapping
organizations.
Hierarchical,
centralized.
COMBIFRON for
Colombia-Venezuela.
MOMEP for Ecuador-
Peru.
Agenda
Emerging agenda
on subregional
security.
Emerging agenda on
problems of
organized and
common criminality.
No consensus on
subregional security
issues except
Colombia-Venezuela.
Link
ages
Strong linkages
to Mercosur and
hemispheric
security
institution.
Directly linked to
other areas of
integration through
SICA.
Economic linkages
between Colombia-
Venezuela and
Ecuador-Peru (through
peace accord).
Rule
Creation
Binding
-ness
Rules more
binding at
subregional level:
e.g. defense
expenditures and
border
deployments.
Much more binding
than hemispheric
counterpart (CA
Democratic Security
Treaty).
Colombia-Venezuela
much stronger than
hemispheric institution.
Ecuador-Peru bound
via outside mediation.
Verifi
ability
Deep network of
military contacts.
Strong U.S.
presence.
International
verification very
strong.
Underdeveloped
subregionally.
Colombia-Venezuela
much stronger than
hemispheric institution.
Ecuador-Peru bound
through MOMEP.


238
(Cuevas and Nakata, 1997). The content of these meetings is generally focused on the
development of new mechanisms to enhance transparency and build mutual confidence.
However, there are instances of deeper security cooperation. According to Chile's 1996
CSBM report, the Argentine and Chilean Air Forces signed an agreement in 1994 on
cooperation for natural disaster preparedness. It is interesting to note the relative absence
of contacts between the two armies ("Chilean-Argentine Joint Exercises," 1997), although
agreements have been reached on proposed exercises (Villagra, 1997).
Personnel exchanges between Argentina and Chile have been less frequent than
the training and cultural and information exchanges mentioned above, but programs are
apparently growing in the context of the recent rapprochement between the two countries.
For instance, Argentina and Chile have participated in a reciprocal exchange of professors
at the Air War College since 1996 (Argentina 1997 CSBM Submission). Of these
exchange programs, perhaps the most significant is the exchange taking place at the
Antarctic bases of each country's Air Force: Chiles Presidente Frei and Argentinas
Marambio (Argentina 1997 CSBM Submission). This is an impressive show of
confidence building in light of the overlapping claims that both countries have made to
territory in Antarctica.
Chilean-Brazilian contacts have increased since 1996, with the initiation of a
program of ship visits, fighter pilot exchanges and joint simulations training (Brazil 1996
CSBM Submission). Brazilian and Chilean naval forces have also participated in joint
operations ("Cardoso discusses regional issues," 1997) and their armies have discussed
the possibility of conducting joint training on armored vehicles (Brazil 1997 CSBM


336
In the absence of a unifying military threat, states are unlikely to agree on rules
that prescribe how to train, maintain and employ their armed forces. The United States,
supporter and shaper of hemispheric security institutions, ultimately finds itself frustrated
by what it sees as lost potential to cooperatively address transnational threats with
military means if necessary. When the frustration boils over unilateral action is a
possible outcome. A better approach is promotion of multilateral responses stepped
down below the hemispheric level, where neighboring countries can agree on what
constitutes a clear and present security problem. As we have seen institutions can then
facilitate coordinated security action.
The hemisphere is at peace and that peace will best be kept by a multilevel set of
security institutions that compliment each other, allowing space for subregional initiatives
without becoming paralyzed by the old cleavages of U.S.-Latin American security
relations. This recommendation, supported by the results of our study, is not new. In the
early 1980s, Varas (1983) argued for a variety of institutional forums, both within and
outside the OAS, to support crisis resolution and promote military transparency. At the
beginning of this decade, Poitras (1990) observed that new multilateral efforts to resolve
Latin American conflicts could be constructed around specific subregions or issues areas,
including arms control and proliferation, the containment of domestic conflict and
confidence building measures (p. 126). More recently, Mares (1997) saw promise in a
"multitiered and integrated security structure" that might "facilitate working out
disagreements locally first, and, as necessary, subregionally, regionally, and globally" (p.
216).


348
Jervis, Robert. 1983. Security Regimes. In Stephen Krasner, ed., International
Regimes. Ithaca, NY: Cornell University Press, 173-194.
Jervis, Robert. 1986. From Balance to Concert: A Study of International Security
Cooperation. In Kenneth A. Oye, ed., Cooperation under Anarchy. Princeton, NJ
Princeton University Press.
Jervis, Robert. 1999. Realism, Neoliberalism, and Cooperation. International
Security 24 (1): 42-63.
Jose, James R. An Inter-American Peace Force within the OAS. Metuchen, NJ:
Scarecrow.
"Justice Minister on Tri-Border Area Security Projects." Folha de Sad Paula (Sao Paulo,
internet version), 27 November 1997, Foreign Broadcast Information Service (on
line), 27 November 1997.
Kacowicz, Arie M. 1997. "Peru vs. Colombia and Senegal vs. Mauritania: Mixed
Dyads and '"Negative Peace.'" In Miriam Fendius Elman, ed., Paths to Peace: Is
Democracy the Answer? Cambridge, MA: MIT Press, 335-369.
Kelly, Philip. 1997. Checkboards and Shatterbelts: The Geopolitics of South America.
Austin: University of Texas Press.
Kelly, Philip and Jack Child. 1988. Geopolitics of the Southern Cone and Antarctica.
Boulder, CO: Lynne Rienner.
Keohane, Robert O. 1984. After Hegemony, Cooperation and Discord in the World
Political Economy. Princeton, NY: Princeton University Press.
Keohane, Robert O., ed. 1986. Neorealism and its Critics. New York: Columbia
University Press.
Keohane, Robert O. 1989. International Institutions and State Power. Boulder, CO:
Westview.
Keohane, Robert O. and Lisa L. Martin. 1995. "The Promise of Institutionalist Theory."
International Security 20(1): 39-51.
Keohane, Robert O. and Joseph S. Nye. 1989. Power and Interdependence. 2nd ed.
New York: HarperCollins.
Kindleberger, Charles P. 1973. The World in Depression, 1929-1939. Berkeley, CA:
University of California Press.


315
Table 6-2: Characteristics of the Hemispheric Security Regime
Characteristics
Observation
Organizational
Features
Forum
Many overlapping organizations, many with
US support.
No centralized system.
Fractured at top between the IADB and OAS.
Agenda
Multilateral issue agenda, but many security
issues have no agreed military component.
Transnational threats problematic.
Many instances of institutional balancing.
Linkages
Democracy crosscuts cooperative security
regime.
Economic linkages underdeveloped at
hemispheric level.
Rule
Creation
Bindingness
Largely non-binding and voluntary.
Deference to subregional contexts.
Verifiability
Weak network of contacts only.
Formats gradually being designed to facilitate
cross-checking.
Cooperative
Security
Behavior
Transparency
Well developed; strongest behavior cluster;
linked to democracy.
Defense
Postures
Weak: only weapons of mass destruction and
anti-personnel landmines.
Deference to subregional contexts.
Burden Sharing
Extremely contentious.
Much U.S. support.
Numerous examples at subregional level.
The hemispheric security institution clearly falls well short of the ideal type the
cooperative security regime laid out in Chapter 2. If the shortfall had followed patterns
expected by realism, then we would have to conclude that realism provides a better
explanation than institutionalism. But, in fact, the organization and rules of the IASS
allowed institutional balancing to take place, which thwarted U.S. attempts to steer the
multilateral security agenda and the evolution of the security institutions themselves,
despite periodic acts of reassurance on the part of the United States.


35
development of common military doctrine, establishment of a common military education
system, and participation in joint training operations. There are discussions in some
subregions regarding procedures to create multinational units, share logistics and
purchase common hardware, and adopt unified command structures.
How can we make sense of these disparate elements?
The recent work in the area of cooperative security shows that there are several
important features to consider.4 First, a cooperative security regime should have the
administrative capacity to facilitate negotiation of rules, track compliance (through data
collection), evaluate results (through data analysis) and report to members. Second a
cooperative security regime should have binding rules regarding use of force. Use of
force may only be employed as a last resort (after exhausting all other means), must be
applied multilaterally (except in the case of self-defense), and can only be used in support
of multilateral, consensus objectives (Nolan et al.,1994). Clearly, there must be prior
agreement on what sorts of crises should invoke a military sanction. Third, rules must be
produced through an inclusive decision making procedure. The Chayes (1994) assert that
a regime is inclusive to the extent that the states affected by it have a reasonable
opportunity to participate in its processes and it does not unfairly discriminate against any
of its members (p. 72). These two general requirements for rules that they be binding
and produced via an inclusive decision making process are important characteristics
that should be present regardless of the particular aspect of security cooperation
considered.
4 This derives from Nolan (1994), the Chayes (1994) and Daadler (1992).


285
Guatemala among member countries of the CACAF. This event had also been
incorporated into the CACAF annual plan of action (El Salvador, 1998b).
Worth noting is the ongoing demining operation in Central America, the Mission
for the Removal of Mines in Central America, known by its Spanish acronym
MARMINCA. MARMINCA functions under the auspices of the OAS and the IADB.
Since it is not principally a subregional organization, it is not covered in detail here.
Operations are ongoing in Costa Rica, Guatemala, Honduras, and Nicaragua, the
operations are conducted by the militaries of each countries, supervised by an
international staff from MARMINCA and supported by international donations. Despite
the fact that the operations are separate, deminers from the recipient countries interact at
periodic training seminars and exercises, which take place primarily in Honduras. There
has not been, to date, much intraregional cooperation in this operation. However, El
Salvador, which eliminated its anti-personnel mine problem earlier in the decade, provide
supervisors to the program, currently stationed in Honduras with responsibilities that
include supervision of the Honduran sappers removing mines along the border with
Nicaragua.
Other important regional exercises, reported by Honduras in its 1996 CSBM
submission, include a riverine, counterdrug exercise sponsored by the U.S. Southern
Command. In 1995, the exercise included Honduras, Guatemala and El Salvador,
operating under U.S. sponsorship. In 1997, this exercise included El Salvador, Belize,
Guatemala, Honduras and Nicaragua. According to Salvadoran reports, the 1997 exercise
was based in both El Salvador and Honduras, and featured combined operations in the
waterways included in the Gulf of Fonseca system, a potential flash point for conflict. A


16
threat to the United States, other than the transnational challenges of drug trafficking,
organized crime and terrorism. 13
These threats represent the dark underside of globalization and interdependence
(Godson and Williams, 1998, p. 66). The same processes that accelerate the flow of
goods, people, ideas and money through the global economy, facilitate criminal
transnational networks, mass immigration flows, and volatile capital movements (Muoz,
1996). To the extent that these threats affect the security of the states, they can be
considered with realist tools.
Transnational threats may not be core survival issues for the United States, but
they do have an impact on security. The threats to security caused by cross-border flows
of contraband (principally drugs), people (immigration and terrorism) and other
phenomena (environmental degradation, natural disasters, etc.) affect the U.S. territory
both directly (drugs, illegal immigration and terrorism) and indirectly (by undermining
newly consolidated democracies and creating an environment of instability which
exacerbates transnational threats). Since unilateral action against such threats is not cost-
effective, the United States might rationally seek ways to influence behavior through
bilateral agreements or the creation of security institutions that foster cooperation and
See Perry (1997), Wilhelm (1998). General Wilhelm commanded the U.S.
Southern Command in the late 1990s. There are numerous official publications that
provide detailed lists of transnational threats of concern to the United States. Often cited
are the transnational threats of drugs, terrorism, trafficking in arms, migrant flows, and
internal instability. The Department of Defense, "United States Security Strategy for the
Americas" (1995), offers such a similar catalog. See Einaudi (1996/7), Lowenthal (1992)
and Van Klaveren (1992) for good treatments of the topic. Villagran de Leon (1993)
offers a review of OAS activity in these areas.


274
rights course in Guatemala; attendance by Nicaraguan officials at the Salvadoran
Command and General Staff College; and Salvadoran participation in a Nicaraguan civil
defense demonstration and the Honduran Soldier Day parade.
There are few other reports of concrete measures of military contacts and
verification, although Honduran and Guatemalan reports include many police-related
contacts, reflecting concerns about contraband flows on the border.
There are no reported instances of prior notification of exercises or prior
consultation on modernization plans between the armed forces of the region, although
these rules are well established and supposedly covered by oversight institutions
reviewed in previously. El Salvador reports that it has established a mechanism for prior
reporting of exercises (El Salvador 1997 CSBM Submission).
What is confirmed by the other CSBM submissions in the subregion is a pattern
of transparency that is markedly different from the contacts between various military
organizations that characterize the Southern Cone. These contacts are more frequently
between non-military security organizations, focused on common criminality, such as
auto theft and cross border incursions by armed bands that conduct robberies and
kidnappings.
Clearly there are many meetings and forums across a very wide spectrum of
activity (Honduras 1997 and Guatemala 1997 CSBM Submission). In addition, there is a
relatively unstructured form of casual communication between border commanders of
Honduran, Nicaraguan and Salvadoran units (Bennett, 1996). But these contacts tend to
be of an informal nature, based on personal relationships established in common
schooling experiences often in the United States under the auspices of the military


128
showed that even a strong defensive position (occupation of the islands) offered little
advantage against technologically superior forces.
Gupta's (1997) account of Brazil's reaction to the war is worth citing in detail:
The post-Falklands land threat was seen as being a surprise attack by the
enemy to grab as much territory as possible and to then negotiate a
settlement. It was decided that Brazilian Armored Cavalry should be
moved out of Rio to the south of the country bordering Argentina. A
mechanized cavalry unit was to be moved along the border with Uruguay.
Further, the impact of the Falklands War was also felt in military planning.
It became necessary to plan for deterring an external power from
projecting military force onto the Latin American mainland. It also
become apparent that the Brazilian armed forces, for so long hampered by
small budgets and used to countering an internal threat, were outclassed in
the event of a future conflict with technologically advanced states. All
three branches, therefore, came up with plans for modernizing their
arsenals." (p. 146)
This interpretation is supported by Varas (1985).
The most recent open conflict in the region, the Peru-Ecuador war of 1995,
featured the use of advanced fighters for the first time in the history of intra-Latin
American conflict (Mares, 1996/7). If we have indeed entered an offense-dominated
world, then security cooperation should be even more difficult for the region. 40
40 Since the relative decline of U.S. influence in the hemisphere and a rise in
regional tensions seem to correlate, some have argued for causality. According to Mares
(1997), military capability and informal ententes between Argentina-Bolivia-Peru and
Brazil-Chile were stable elements in the defense-dominated world that existed in Latin
American until the late 1970s" (Mares, 1997, p. 210). At that point U.S. "unilateral arms
control policy via military assistance programs faltered. In the 1970s, Latin American
countries turned to international market to purchase tanks, aircraft and missile systems
and developed indigenous arms industries" (Mares, 1997, p. 212). This period saw a
rapid doubling in military expenditures (Varas, 1985).


135
process? Are verification mechanisms sufficient to overcome the cooperative dilemmas
of cheating, uncertainty and relative gains? More specifically, does the pattern of rule
making in the areas of transparency, defense posture and burden sharing reflect
institutional politics or an effort to capture absolute gains which otherwise might be
unattainable due to dilemmas of cooperation? Finally, are security rules linked to other
issue areas in ways that strengthen their prescriptions for behavior or in ways that simply
exacerbate institutional politics?
Third, what pattern of cooperative security behavior emerges from the evidence?
Is the pattern power-based or rule-based? Does cooperative behavior respond to the
promise of absolute gains in the area of security or simply reflect the interest on the part
of individual states to maximize relative gains? Is information reported via transparency
measures significant? Do defense postures reflect the impact of regime rules? Is burden
sharing taking place, especially in the area of transnational security threats?
In this chapter, as we shall see, a hemispheric security regime does emerge from
the analysis, but it is incomplete, deeply divided especially in the areas of burden sharing
and defensive military postures, and not likely a direct source of cooperative behavior.
The regime is not simply a reflection of power, rather it is a child of institutional politics.
Liberal institutionalism offers an explanatory model that adds to analysis derived solely
from realism.
The cooperative security behavior present in the hemisphere corresponds
completely neither with the pattern expected by power considerations, nor with rule based
behavior. The evidence points to important variations in geographic subregions.
Differences in the degree of transparency primarily, but also in terms of defense postures


74
Contending Hypotheses
Hemispheric Level
At this point, we are ready to summarize contending hypotheses for security
cooperation from the two theoretical frameworks discussed above realism and
institutionalism. We will formalize hypotheses that address organizational
characteristics, rule creation and security behavior in each of the clusters examined. In
this section, we focus only on the hemispheric level. Subregional hypotheses are
developed in the following section.
As we shall see in more detail in Chapters 3 and 4, there is no denying the
substantial institution building process in the area of hemispheric security that has been
underway since the advent of WWII. Old hemispheric security organizations have been
substantially renovated and several new ones have been created during that period. But
are they sufficiently resourced by member states to effectively facilitate negotiations and
carry out the functions of regime management? Realist thinkers doubt that possibility,
except where such an arrangement might serve the interests of the most powerful states.
A realist hypothesis would make resourcing contingent on an agenda consistent
with the new, post-Cold War security agenda of the United States. Thus institutions
should be largely underwritten by the United States and generally supportive of the U.S.
security agenda, notwithstanding small deviations that do not threaten core U.S. security
concerns. Other countries are likely to participate only in areas of common interests.
Institutionalism may not differ markedly from this prediction, since no
hemispheric institution or regime could be expected to resist U.S. influence. However,
even under conditions of marked power asymmetry, institutions may demonstrate some


313
Table 6-1: Historical Evolution of Hemispheric Security
Period
Key Event
WWII and
aftermath
1942: IADB ESTABLISHED Oldest multilateral regional security
body. Promotes interoperability and begins planning for continental
defense.
1945: ACT OF CHAPULTEPEC Reaffirms the principle of
collective responsibility for the security of the hemisphere.
1947: INTER-AMERICAN TREATY OF RECIPROCAL
ASSISTANCE Provides the machinery to maintain peace and
security in the hemisphere. Provides for the Advisory Defense
Council.
1948: NINTH CONFERENCE OF AMERICAN STATES -
Establishes the OAS and confirms charter of Inter-American Defense
Board.
Cold War
1954: CARACAS DECLARATION Establishes communism as
trigger threat for the Rio Treaty.
1962: COMBINED QUARANTINE FORCE Collective action
during Cuban missile crisis.
1965: INTER-AMERICAN PEACE FORCE High water mark of
hemispheric security cooperation.
1967: BUENOS AIRES CONFERENCE Effort to rationalize IASS
defeated.
Post Cold
War
1991: SANTIAGO DECLARATION-- Commitment to democracy.
1993: WASHINGTON PROTOCOL Commitment to the defense
of democracy.
1994/5: OAS HEMISPHERIC SECURITY COMMITTEE New
vision of cooperative security.
1995/8: SANTIAGO AND SAN SALVADOR CSBM
DECLARATIONS Codified rules for cooperative security.
Although we yet lack sufficient historical perspective, it appears that the 1990s
attempt to organize hemispheric security may have also passed its high water mark: the
introduction of the concept of cooperative security in 1994 and the creation of a
permanent hemispheric security committee in the OAS in 1995. The dynamics of
institutional politics under conditions of asymmetry militate against further centralization


197
performance as well.37 In short, the findings for defense reductions and burden sharing
show both movement outside of U.S. interests and resistance to U.S. priorities.
Regime Effectiveness
Introduction
The chapter now turns to the task of examining the effectiveness of the
hemispheric security regime in the 1990s. Do security rules have any real effect on
behavior? Do institutions encourage security cooperation in measurable ways? Is
cooperation more than a mere reflection of the U.S. security agenda? These research
questions, posed in the introduction to this chapter, provide the focus of our investigation
of regime effectiveness. In this section, the two theories compete for the best explanation
of patterns of cooperative security behavior.
Using the CSBM reports and other data that emanate from countries and security
institutions, we establish a pattern of compliance that can be analyzed for regime
effectiveness. As we shall see, compliance varies across time, across the principal rule-
making areas transparency/verification, burden sharing and defensive orientation and
across geographic regions. That geographic variation will be the focus of Chapter 5,
when the subregional case studies are considered.
The analysis in this section is based principally on the 1995, 1996 and 1997
inventories of CSBMs compiled by the Inter-American Defense Board, as well as some
early submissions for the year 1998. These inventories provide an overview of the
37 I do not suggest that CICAD developed this with such a strategy in mind. But
it was proposed at the OAS General Assembly (1997) by Belize, a victim of
decertification, precisely to highlight the unilateral nature of U.S. counterdrug policy.


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor Of Philosophy
THE NEW POLITICS OF COOPERATIVE SECURITY
IN THE AMERICAS
By
William B. Fullerton
December 1999
Chairman: Terry M. McCoy
Major Department: Political Science
Hypotheses derived from institutionalism are tested against the pattern of
cooperative security observed in Western Hemisphere, especially since the end of the
Cold War. The study addresses the hemispheric level and in three subregions: the
Southern Cone, the Andean Region, and Central America. Findings confirm that security
institutions have an independent effect in the areas of transparency in military matters,
reductions in defense postures, and burden sharing. State power and interest play an
important role, but power based theories alone are inadequate. At both the hemispheric
and subregional levels, the processes of institutional politics produce rules that validated
institutional hypotheses. The pattern of security behavior that emerges is generally rule
based. The study makes three contributions to theory: (1) improved understanding of
security politics in asymmetric systems; (2) the impact of relative gains on cooperation;
and (3) the relationship of transparency, democracy and cooperative security.
IV


38
informal conventions to formal intergovernmental organizations.6 Their rules prescribe
behavioral roles for actors in accordance with general norms (Keohane, 1989, p. 163).
Together, organizations, norms and rules regarding the use of military force make up
what we identify as the cooperative security regime in the hemisphere. Tracing the
evolution of these elements and comparing them to underlying power and interests of
states in the region especially along the fault line of U.S.-Latin American relations -
offer a constructive comparison of realism and institutionalism.
I treat institutions as conceptually distinct from observable behaviors in areas of
transparency and verification, defense postures and burden sharing 7 Those behaviors
demonstrate a pattern of security cooperation that can be compared to that promoted by
security institutions. Such an exercise offers a second test of the explanatory power of
the contending theoretical paradigms. This research requires, then, data that can support
discussion of the evolution of institutions vis-a-vis power and interests, as well as data
that substantiates the depth and scope of cooperative security behaviors. These are the
principal methodological challenges of the study.
The remainder of this chapter examines both the realist and institutionalist
paradigms in light of their explanatory power regarding cooperative security behavior,
first at the hemispheric level and then at the subregional level. The dilemmas of
^ See Keohane (1989) for a longer discussion of typologies and factors related to
this discussion. Nye (1991) offers a similar definition: persistent sets of norms that
shape expectations and constrain behavior (p. 155).
7 Not all forms of institutionalism adopt this perspective, as we shall see when
reviewing reflective approaches to the study of institutionalism.


230
members for the exchange of operational information regarding ship traffic, with links to
the Inter-American Naval Conference.4
Members include Argentina, Brazil, Paraguay and Uruguay. The seat of this
regional organization rotates among all member countries, with the exception of Paraguay
(Argentina 1996 CSBM Submission). Related to this institution is a Brazilian course for
organizing maritime traffic, which has included participants from several countries in the
region (Brazil 1996 CSBM Submission).
A newly created operational organization with a growing role in regional security
is the Tripartite Border Command, which rotates responsibility for policing the common
border areas of Argentina, Brazil and Paraguay every six months among the members.
This command, while primarily dedicated to police, intelligence and customs operations,
does include the military (Justice Minister, 1997). Related agreements on border
operations have been extended to Chile and Bolivia.
Of symbolic importance is the 1998 Declaration of Mercosur as a Zone of Peace
(Argentina and Chile, 1998). This declaration is remarkable because it includes Bolivia
and Chile, which have had no significant military contacts in the area of security
coordination. This agreement includes measures for binding arbitration.
The Zone of Peace declaration also includes a clause that would make the
Southern Cone a mine-free area a contentious issue between Chile and Bolivia given
4 CAMAS was part of a larger system, established by the IADB plan for the
defense of maritime traffic produced during the Cold War. Representatives from the
Interamerican Naval Conference examined the plan in the early 1990s with the intent to
update it. But the plan was too outmoded to provide a useful basis for cooperation. Still
CAMAS continues to exist as a coordinating entity.


43
enough information to convict on a less serious charge. Both are offered freedom in
exchange for evidence against their partner. If only one provides information, the other
will receive a harsh sentence. Absent reliable information and assurances that the partner
will remain silent, both prisoners are likely to offer evidence against each other leading
to imprisonment. In short, they rationally select strategies that lead to suboptimal
outcomes. 16 This problem potentially affects all three of our behavior clusters.
In the case of transparency, states that report information about their military
activities may expose critical weaknesses. If potential rivals do not also report their
activities, then the reporting state may increase its vulnerability by compliance. Thus
states may not provide information or may provide partial or misleading information.
Reduction of defense postures is also affected by the possibility of cheating.
States that do reduce their postures will increase their vulnerability if potential rivals do
not reciprocate. Thus states may not reduce their postures, or may offset those reductions
by seeking alliances engaging in competitive balancing.
Burden sharing faces a similar dilemma, to the extent that it is based on the
pooling of capabilities and specialization of roles. States that design their forces based on
the assumption that others will provide support in certain areas (sea patrols, air defense,
etc.) become vulnerable if others do not uphold their commitments. Thus states will tend
not to cooperate or may offset vulnerabilities by building military capabilities, thus
affecting the second behavior cluster. 1 7
16 See Oye (1996) for a description of the Prisoners Dilemma and related games.
17 See Snyder (1990) on burden sharing concerns.


158
this is most apparent in the resistance to organize military responses to meet transnational
threats. We will address this theme in more detail in the following section.
On balance, then, the evidence casts doubt on the realist hypotheses. There is
simply too much strength to this overlapping set of institutions (although not unified) and
too much resilience in its resistance to adopting a strong burden sharing dimension.
With that, let us proceed on to the next task, examining the evolution of regime rules.
Regime Rules
Introduction
The purpose of this section is to complete our construction of the hemispheric
security regime by specifying the rules that govern behavior in the areas of military
transparency and verification, burden sharing, and defense postures. As we consider the
evolution of these rules, produced by their associated institutions, we will look to see if
they simply reflect the processes of institutional bargaining or in fact are designed to
facilitate cooperative behavior. We will also look to see if they require binding
commitments from states and involve substantive areas. Finally, we will be alert the
degree of linkage to other issue areas.
As we shall see, the rules regarding transparency and verification include sharing
information on arms transfers, defense budgets and strategy, prior notification of
exercises, programs of visits and exchanges at military bases and other installations, and
CSBM conferences in a variety of settings. Rules regarding burden sharing are less well
developed, but include use of military assets to respond to natural disasters, coordination
to avoid and react to transportation accidents and incidents, border contacts, small island


349
Klare, Michael T. 1994. The Perspectives of the U.S. Government and Private Military
Sales. In Lars Schoultz, William C. Smith and Augusto Varas, eds., Security,
Democracy and Development in U.S.-Latin American Relations. Boulder, CO:
Lynne Rienner, 153-169.
Kolodziej, Edward A. and William Zartman. 1996. Coping with Conflict: A Global
Approach. In Edward A. Kolodziej and Roger E. Kanet, eds., Coping with Conflict
after the Cold War. Baltimore, MD: Johns Hopkins University Press, 3-32.
Krasner, Stephen D. 1976. State Power and the Structure of International Trade.
Reprinted in Robert J. Art and Robert Jervis, eds., International Politics: Anarchy,
Force, Political Economy, and Decision Making. 2nd ed. Boston: Little, Brown,
1985,340-363.
Krasner, Stephen D. 1985. Structural Conflict: The Third World Against Global
Liberalism. Berkeley: University of California Press.
Krasner, Stephen D. 1991. Global Communications and National Power: Life on the
Pareto Frontier. World Politics 43: 336-66.
Krasner, Stephen D., ed. 1983. International Regimes. Ithaca: Cornell University Press.
Krepon, Michael. 1995. A Handbook of Confidence-building Measures for Regional
Security. 2nd ed. Washington, D.C.: The Henry L. Stimson Center.
Kresho, Mark. 1987. Conferencia de Ejrcitos Americanos. Military Review 67 (8):
4-9.
Kupchan, Charles A. and Clifford A. Kupchan. 1995. "The Promise of Collective
Security." International Security 20 (1): 52-61.
Lake, David. 1988. Power, Protectionism, and Free Trade: International Sources of
U.S. Commercial Strategy, 1887-1939. Ithaca, NY: Cornell University Press.
Lake, David. 1993. Leadership, Hegemony, and the International Economy: Naked
Emperor or Tattered Monarch with Potential? International Studies Quarterly 37:
459-89.
Lake, David A. and Patrick M. Morgan, eds. 1997. Regional Orders: Building Security
in a New World. University Park, PA: Pennsylvania State University Press.
Lambom, Alan C. 1998. "Theoretical and Historical Perspectives on Collective
Security: The Intellectual Roots of Contemporary Debates about Collective Conflict
Management." In Joseph Lepgold and Thomas G. Weiss, eds., Collective Conflict


26
If the processes of globalization in the 1990s have favored subregional security
integration, what are the prospects for a hemispheric counterpart? Based on the pattern of
institutional security politics in asymmetric systems, they are poor. Varas (1992)
categorizes U.S. security behavior in the region as varying between "hegemonic control"
and "coercive control." These labels exaggerate the influence of the United States, but
nevertheless offer a key to understanding the new security politics. What Varas means by
"hegemonic control" is a U.S. policy of using regional institutions to give a multilateral
flavor to its security policies. When this course of action seems more cost-effective than
unilateral action ("coercive control"), the United States attempts to create or remake
security institutions that reflect its own interests in the region.
Under the best of circumstances, this is not easy. Because institutions tend to
restore symmetry through common decision-making procedures, opportunities for small
countries to "voice" their interests, and facilitation of issue linkage new forms of
political behavior are possible: "institutional bandwagoning" (support to U.S. initiatives)
and institutional "balancing" (resistance to U.S. initiatives). The United States is forced
into an assurance game to maintain support, requiring constant displays of multilateralism
(acquiescing to majority decisions, accepting binding obligations, refraining from
unilateralism), as well as significant side payments to potential balancers (e.g.,
Mastanduno security outlets and other benefits). This is because the U.S. commitment
to multilateralism is rarely credible on its own merits. The "shadow of the future" for
Latin American states is the certainty that the United States will defect or change the
rules of the game when necessary to secure its own interests.


113
technical meetings, intelligence meetings, as well as gatherings of the military chiefs.17
Subregional military organizations, such as the Central American Defense Council
(CONDECA), have occasionally sent representatives to IADB meetings and reported on
their activities.18
The IADB itself was established in 1942 by the Third Meeting of the Ministers of
Foreign Affairs in Rio de Janeiro, in the aftermath of the Japanese attack on Pearl Harbor
at the initiate of the U.S. Department of State.19 The Board shows its parentage its
host country is the United States, the Chairman is always a U.S. general officer and the
Secretary is a U.S. colonel.
The mission of the Board, until it was substantially altered in 1995, was to (1)
plan for hemispheric defense; (2) prepare member states to act collectively; and (3)
provide technical assistance in its area of competence. While the precise wording of the
17 The term "conferences of the armed forces" is a convenient shorthand. In fact,
there are three separate conference "systems" for the three main armed services which
exist independently of each other, all established between 1959 and 1961. In general they
consist of an annual or biennial series of technical conferences which culminate in a
Commander's Conference where agreements are collectively endorsed by the chiefs of the
armed services. Membership largely overlaps that of the IADB, with some notable
exceptions (Canada participates in the conferences but is not a member of the IADB).
111 A 1966 Board study offers a perspective on the maturing military system: "The
creation of the Central American Defense Council, the periodical meetings of the
commanders-in-chief of the armed forces of the Americas, and the holding of combined
military exercises in the Americas demonstrate the concern of the Governments with the
suitable preparation of collective aspects of the defense of the Continent" (IADB, 1966,
P-11)-
19 See Spracher and Mason (1994) for a recent history of the IADB. The IADB
yearbooks also provide good information.


270
studies in areas of common interest; (5) contribute to studies to promote sustainable
development and disaster preparedness; and (6) maintain permanent communication
with the secretary-general of SICA.
The organization consists of a superior council (comprised of the senior military
officer from each member country), a president (senior military official from the host
country), an executive committee (service chiefs from each participating country), and a
secretariat (from the host country). The superior council, which held its first meeting in
April 1998 has authority to establish directives relating to educational programs, disaster
assistance and studies that the armed forces of member countries are obligated to respect
(El Salvador 1998 CSBM Submission). CACAF has already spawned specialized
conferences some areas, as well as prompting conferences for the subregion in individual
services. In 1998, separate conferences took place in the areas of military intelligence,
demining operations, military medicine and naval operations (CACAF, 1998).
CACAF is based on the same principles established in the Democratic Security
Treaty that is the model of democratic security. Not surprisingly, the CACAF sees
the armed forces playing a critical role in that security model, especially as regards citizen
security, sustainable development and confidence building between member countries
(CACAF, 1998). Future plans include the establishment of a Central American
University for Security and Defense, studies on regional military integration, organization
of a subregional support unit for peacekeeping operations, execution of a plan of action to
implement a "culture of peace and nonviolence" (see below), and the creation of an
internet based information system (CACAF, 1998, p. 14).


149
vigorous debate over the extent of Board involvement in counterdrug activities prompted
by Mexico limited activity to attending conferences and conducting studies. By contrast,
the Board has been very involved with the demining effort in Central America, in close
collaboration with the OAS Unit for the Promotion of Democracy (UPD).20 Part of the
reason for the success is undoubtedly because this operation is not hemispheric in nature
- it is limited to a particular geographic region and the countries that provide demining
supervisors.21 Another reason may be the clear subordination of this effort to the OAS.
Fourth, the Board has become more responsive to the OAS, under the terms of
Resolution 1240. 22 Work has been productive in areas that the OAS acknowledges an
important military component: demining and confidence and security building measures
Based on a 1995 request of assistance from the OAS Permanent Council, the Board
compiled an inventory of CSBMs adopted in the hemisphere from submissions by some
member countries. This has evolved into an annual service the Board provides to the
20 The UPD was established on October 15, 1990, to support democratic
development by helping member countries strengthen their political institutions and
democratic procedures.
21 The IADB supports the UPD demining program by providing military
supervisors from throughout the Americas to monitor operations conducted by host
nation militaries in the countries of Costa Rica, Guatemala, Honduras and Nicaragua.
The headquarters for the IADB supervisors, based for most of the 1990s in Danli,
Honduras, is known by its Spanish acronym MARMINCA (Assistance Mission for the
Removal of Mines in Central America).
22 Even prior to Resolution 1240, the Board tried to demonstrate its relevance to
the OAS. In the 1990s, Board officers verified missile holdings in Nicaragua in
conjunction with CIAV and conducted an in-country evaluation of disaster preparedness
in Grenada. The verification mission was taken without the approval of the Boards
Council of Delegates, which certainly would have blocked it.


85
states in the region. This is largely the case with the Andean countries as well, although
relative power concerns may overshadow transnational threats in some border regions,
such as the Cenapa region along the Peru-Ecuador border. In the Southern Cone, there
are few transnational considerations, aside from smuggling of contraband in the tri-border
area of Argentina, Brazil and Paraguay. The primary security problem is the traditional
one of maintaining a reasonable balance of power.44
These variations in security problems can be handled by our existing theories. All require
some sort of security cooperation for their solution. They differ only in the details of the
absolute and relative gains that result from that security cooperation. Thus, the
differences can be expressed in specific research questions derived from the basic
theories. Table 2-5 and Table 2-6 summarize the hypotheses developed here for the
subregional cases.
44 Military spending in Latin America has historically been concentrated on the
Southern Cone countries of Argentina, Brazil and Chile. During the arms build-ups of
the 1980s, these three countries accounted for almost two-thirds of the total military
expenditures in the region (Varas, 1985).


200
the thematic pattern expected by both realism and institutionalism. But before
discussing this pattern further, let us look at how reporting has varied over time.
Pattern Over Time
Comparisons of the 1997 and 1995 data reflect a 15% increase in the number of
countries reporting. Of the 34 member countries of the OAS, 17 submitted inventories in
1995 (14 of them with delegates at the IADB and 3 without), amounting to a 50%
participation. In 1997, 22 countries submitted inventories (17 represented at the IADB
and 5 not), reflecting a 65% response rate. It is obvious that participation, although
growing, is still partial. However, since all major countries did submit some kind of
report, it is unlikely that more participation would skew the data significantly.
Comparison from 1995 to 1997 also shows an 87% increase in the number of
CSBMs reported.38 This most likely reflects deeper participation in the CSBM regime,
since the increase in the number of countries reporting since 1995 is not remarkable,
while there is a significant increase in the number of reported CSBMs. These figures
may be subject to minor modifications depending on the criteria used to identity them,
since the information contained in the inventory sometimes gives rise to different
interpretations.
38 a quantitative tabulation shows that 516 measures were implemented in 1995
and 982 in 1997, reflecting a significant increase of close to 87%.


322
maintenance of these institutions, power cannot determine their direction. Furthermore, it
is not enough for a hegemon to simply bear the costs of regime maintenance in order to
ensure the survival of multilateral institutions. The hegemon must also engage in acts of
reassurance to forestall balancing. Ultimately, if mutual interests are not strong, no act of
reassurance, side payment or leadership outlet will suffice to maintain the viability of
hemispheric institutions.
Second, the analysis in this study shows that the debate over relative gains
between institutionalists and realists appears overdrawn. In the case studies analyzed
here, once institutions were established to facilitate flows of relevant information between
member states, relative gains concerns rarely prevented cooperative action where such
action could realize absolute gains. Hard verification schemes and formal mechanisms
for reciprocity were not required. The numerous military contacts that exist in the
Southern Cone, for instance, produce a large volume of information and assurance that
neighbors were committed to subregional security. This has permitted extensive burden
sharing activity, despite the presence of arms modernization programs in each of the
larger states.
In areas where cooperative security behavior falls short of the ideal, the most
logical explanation does not lie with obstacles associated with relative gains, but rather
with the absence of strong mutual interests that is, the absence of common absolute
gains from cooperation. The relative lack of reductions in defense expenditures in the
Southern Cone, for instance, is best explained by the fact that the countries of the
subregion want to maintain viable military capability to project their image
internationally (through peacekeeping operations), while protecting their domestic arms


131
organization, which would doubtless serve the interests of the United States, had no place
in the inter-American system. As a compromise, the Board was allowed to exist with the
implicit agreement that the OAS would not be militarized .45
For the United States then, the Board, at least since the end of WWII, has been a
"second-best" option. It is also a "second-best" option for the countries (principally but
not exclusively Mexico) that would prefer a minimalist approach nothing more than a
mechanism that would convoke an ad hoc council when crises required, such as the
Advisory Defense Committee contained in the OAS Charter.
Between these two positions, the Board survived in 1948 as a "temporary"
compromise solution. As a compromise it frustrates both the proponents and opponents
of a permanent military organ in the hemispheric security system. Those who judge the
Board against the imagined efficiency of a permanent military staff within the Charter are
frustrated by what they see as lost potential at the Board; while those who compare the
Board to an ad hoc crisis committee convened for a specific, temporary purpose are
alarmed by its autonomous activity .46
The Board continued working to enhance interoperability through its own
standardization efforts in the area of military terminology, search and rescue, disaster
45 See Council on Foreign Relations (1962).
46 The compromise is enlivened by a paradox: the Board has survived precisely
because it lies outside of the OAS Charter, yet receives financial support from the OAS.


77
would predict that the United States is likely to act with restraint towards more powerful
Latin American states in order to avoid eliciting a balancing reaction.
Institutionalism expects to find rule-based behavior not behavior driven by
balancing or U.S. influence. Bolstered by expectations of convergence (supported by
binding rules and strong verification schemes) and eased by the facilitation that
institutions provide (reduced transaction costs), states will tend to follow the rules
established in organizations formed for that purpose. However, if acts of reassurance are
insufficient to maintain cooperation, then the dynamics of institutional politics will
predominate. These contending hypotheses are summarized in Table 2-3 and Table 2-4.
Table 2-3: Contending Hypotheses at the Hemispheric Level for Institutions
Dependent
Variables
Realist
Hypotheses
Institutionalist
Hypotheses
Security
Institutions:
Given institutions as constructed focal points for negotiations and
absolute gains from military transparency, reduced defense postures
and burden sharing, under conditions of marked power asymmetry ...
Organizational
Features
1. Institutional agendas will not
stray significantly from U.S.
interests and will be
resourced only where
interests coincide. Issue
linkage will complicate
cooperation.
1. Institutional procedures and
issue linkage will restore
symmetry, providing
opportunities for institutional
balancing. Agenda will be
multilateral, focused on issues
where absolute gains are
evident.
Rule
Creation
2. Rules will not be binding and
comprehensive, but rather
voluntary and partial.
3. Rules will not be supported
by verification mechanisms
sufficient for reciprocity.
2. Rules will be binding wherever
the potential for defection
exists.
3. States will insist on strong
verification mechanisms so that
reciprocity can function.


292
subregion as a serious transnational issue, affecting countries all along the chain of
production, refinement, transit and consumption. Solution of this security problem may
require comprehensive subregional cooperation to shut down drug trafficking, cross-
border criminality and guerrilla movements.
But in the Andean Ridge, such cooperation carries with it a more serious relative
gains dilemma. In this subregion cooperation would have to involve assistance to
Colombia to control its own territory. Absolute gains would accrue most directly to that
country. There might well be a marked tendency to shirk and allow another state to bear
the cost of containing the security problem. The high threat levels that have only recently
begun to decline between states of the subregion mean that sensitivity to relative gains
will be high. To date, there has been little progress in addressing transnational problems
at the subregional level. Bilateral cooperation between Venezuela and Colombia is
impressive, as we shall see, but has not been replicated elsewhere.
Second, this case study offers a middle-range example of U.S. influence. The
subregions relationship with the United States is far more direct than that of the Southern
Cone, but nowhere nearly as pervasive as in Central America. The U.S. security agenda
includes drug trafficking as a major priority and has pursued wide-ranging bilateral
agreements, which have sometimes led to dramatic successes, such as the dismantling of
the aerial trafficking route over Peru in the mid 1990s. The United States has also acted
to contain bilateral conflict in Ecuador and Peru in its role as one of the guarantor states.
There are, however, far fewer instances of the United States promoting subregional
cooperation here than in Central America.


251
developed into a full-scale exercise with military maneuvers over the next three years.
The 1997 exercise, which took place in southern Brazil, featured the joint deployment of
more than 2400 troops from all three countries for maneuvers which lasted ten days and
included the deployment of armored and amphibious vehicles, helicopters, and some of
the most advanced weapons systems in the region ("Mercosur Maneuvers," 1997).
Bolivia and Paraguay participated as observers ("Mercosur Member Countries," 1997).
Aside from joint participation in the Southern Cross exercise, Argentina and
Uruguay participate in a large bilateral peacekeeping exercise called Ceibo, which
alternates between countries each year. (Argentina 1997 CSBM Submission). Exercise
Ceibo aims to build joint operational capability for U.N.-sponsored peacekeeping
operations ("Argentine, Uruguayan Armies," 1998). In 1997, the exercise for the first
time was built around an integrated battalion, in which Argentine and Uruguayan officers
planned and implemented combined operations ("Uruguayan, Argentine Armies," 1997).
These exercises have led to joint deployments during actual peacekeeping
operations. The 1997 Argentine contingent to the UN mission in Cyprus included five
officers from Brazil and Uruguay who were integrated into the Argentine command of
380 peacekeepers ("Military Maneuvers," 1997). Preparation for this deployment was
conducted at the Argentine Joint Training Center for Peacekeeping Operations
established in 1995 (known as CAECOPAZ), which has taken on a growing subregional
training role ("Confidentiality Accords," 1998), accepting students from neighboring
countries.
Participation by Argentina, Brazil and Uruguay in UN peacekeeping operations
has grown spectacularly in the 1990s, including deployments by Argentine and Brazilian


124
programs.32 As a result, a generation of Chilean and Brazilian military leaders would
have no direct exposure to U.S. military doctrine.
A much debated topic is the extent of the decline of U.S. influence from its peak
in the 1950s, when the United States, equal in population to Latin America, out-produced
its neighbors seven-fold while attracting half of their exports and providing much of their
foreign investment (Lowenthal, 1992). This economic hegemony appeared to have a
political dimension as well, with Latin American countries generally accepting U.S.
leadership in hemispheric security initiatives as well as U.S. positions in the UN General
Assembly.
While economic asymmetry survived longer in Latin America than in other
regions of the world, by the late 1970s it had eroded significantly. The larger Latin
American economies developed and diversified, with per capita growth rates the
exceeded the United States.33 ¡n the 1980s, the debt crisis, aggressive policies on the
part of the Reagan administration and a reinvigorated U.S. economy combined to restore
the influence of the United States (Muoz, 1996a; Poitras, 1990; Sanderson, 1992; Van
Klaveren, 1992), although probably not to prior levels. Attempts to engage the large
South American countries were largely rebuffed and U.S. security activity was focused
32 See Varas (44, 102). It is important to note that human rights was not the only
issue that prompted the Carter administration to restrict arms transfers. The U.S. viewed
Brazils nuclear program with alarm and had interrupted commercial sales of fuel to its
reactors in 1974 (Skidmore, 1988).
33 See van Klaveren (1992). Also, Corrales and Feinberg (1999) who remind us
that during this period Venezuela helped found OPEC, Brazil developed ties to
Luzophone Africa, Argentina increased trade with the Soviet Union and Peru purchased
Soviet arms.


333
Committee on Hemispheric Security. Ironically, that institutional change, designed to
make the IADB dependent on civilian leadership, has been resisted precisely by those
who criticize the IADB for its independence. Clearly, the conflict there relates not to the
issue of civilian control but rather to the resistance on the part of balancing states to
initiatives that would rationalize and centralize the hemispheric security system.
The debate over the Board has recently been subsumed by the overarching review
of hemispheric security mandated by the 1998 Summit of the Americas. Over the next
two years, the OAS will: 1) analyze concepts of hemispheric security, 2) develop
common "management" approaches to security problems and 3) suggest ways to reform
hemispheric security institutions.^ This three-part security review provides the backdrop
to analysis of the Board's future and its relationship with the OAS.
What is the larger goal of the United States with respect to the ongoing review of
hemispheric security architecture and what role might the Board play in the new scheme?
U.S. policy on these issues will undoubtedly be framed by key principles derived from
national security strategy, presidential decision directives and other policy guidance
(Buchanan, 1998). Three of these principles are broadly recognizable in U.S. security
policy for the region. First, support for democracy the United States works to
strengthen democratic governments as a way to promote freedom, ensure human rights
^ The process derives from a 1998 Summit mandate to the OAS Committee on
Hemispheric Security "to analyze the meaning, scope, and implications of international
security concepts in the Hemisphere, with a view to developing the most appropriate
common approaches by which to manage their various aspects, including disarmament
and arms control" and to "pinpoint ways to revitalize and strengthen the institutions of the
Inter-American System related to the various aspects of Hemispheric Security" (IADC,
1998, p. 30).


103
which were fought in response to attempts by some Central American leaders to recreate
the United Provinces, dismantled by the civil wars of 1838-42.
Throughout most of the 19lh century, offensive operations, although not
impossible, were generally hampered by geographic barriers that favored defense.
Variations in geographic endowments affected the nature of conflict, though not
necessarily the balancing pattern. Resende-Santos (1996) makes the argument that Chile,
as a defensively disadvantaged state with "compact and exposed national territory,"
developed an "aggressive national security policy," based on "highly mobile, offensively
organized forces," with a "propensity to strike first" with preventative war and engage in
"offensive action and warfare on the enemys home turf' (Resende-Santos, 1996, pp. 217-
219). In contrast, a defensively advantaged state like Brazil, with its large territory and
dispersed population centers, tended to maintain "only a deterrent force" concentrating on
"defensive positioning and fortifications." Nevertheless, both Brazil and Chile balanced
against Argentina. Over time, the development of communications and transportation
infrastructure has generally muted the defensive advantages provided by geographic
boundaries.
Technologically advanced weapons also had an impact on the nature of conflict.
The Triple Alliance War of 1865-1870 between Paraguay and the combined forces of
Argentina, Brazil and Uruguay employed repeating rifles, armored warships and an
observation balloon advanced systems that were only gradually introduced in the
United States during its Civil War. Less than a decade later, the War of the Pacific
between Chile and the Peru-Bolivian alliance featured use of breech-loading artillery and
machine guns.


361
Van Klaveren, Alberto. 1996. "Understanding Latin American Foreign Policies. In
Heraldo Muoz and Joseph S. Tulchin, eds., Latin American Nations in World
Politics. Boulder, CO: Westview, 35-60.
Varas, Augusto. 1985. Militarization and the International Arms Race in Latin America.
Boulder, CO: Westview.
Varas, Augusto. 1992. "From Coercion to Partnership." In Jonathan Hartlyn, Lars
Schoultz and Augusto Varas, eds., The United States and Latin America in the
1990s: Beyond the Cold War. Chapel Hill, NC: University of North Carolina Press,
46-63.
Vieira, Gleuber. 1995. "La Variable Estratgica de el Proceso de Constitucin del
Mercosur." Ser en el 2000 8: 8-13.
Villagra, Pedro. 1998. "La Seguridad y la Defensa en el Sur del Continente." In Rut
Diamint, ed., Argentine y la Seguridad Internacional. Santiago: Facultad
Latinoamericana de Ciencias Sociales, 17-25.
Villagran de Len, Francisco. 1993. The OAS and Regional Security. Washington, DC:
United States Institute for Peace
Viner, Jacob. 1944. Peace as an Economic Problem. Reprinted in Robert J. Art and
Robert Jervis, eds., International Politics: Anarchy, Force, Political Economy, and
Decision Making. 2nd ed. Boston: Little, Brown, 1985, 291-302.
Wager, Stephen J. 1996. "The Mexican Military: The Dilemma of Functioning in a One-
Party System." In Richard L. Millett and Michael Gold-Bliss, eds., Beyond
Praetorianism: The Latin American Military in Transition. Coral Gables, FL:
University of Miami North-South Center, 103-131.
Wallensteen, Peter and Margareta Sollenberg. 1998. Armed Conflict and Regional
Conflict Complexes, 1989-1997. Journal of Peace Research 35 (5): 621-634.
Walt, Stephen M. 1987. The Origin of Alliances. Ithaca, NY: Cornell University Press.
Waltz, Kenneth M. 1979. Theory of International Politics. New York: Random House.
Weisburd, Mark A. 1997. Use of Force: The Practice of States Since World War II.
University Park, PA: Pennsylvania State University Press.
Weitsman, Patricia. 1997. Intimate Enemies: The Politics of Peacetime Alliances.
Security Studies 7(1): 156-192.


160
1238, Information on Defense Spending and Register of Conventional Arms) which
urged compliance with the UN registries and identified the purpose of the register to
promote reductions in conventional arms. In 1994, a follow-on resolution added the
provision that the information provided in the UN registries and contained in documents
on defense strategies be exchanged at the OAS and discussed among member states.
The year 1994 also saw the first General Assembly resolution entitled
Confidence and Security Building Measures in the Region. The resolution
recommended adoption of CSBMs at the appropriate level and charged the Permanent
Council with compiling a CSBM inventory a task that was later handed to the IADB.
A long series of declarations, reports, and resolutions from various commissions,
committees, and conferences of the OAS, the IADB, the Presidential Summits and the
DMAs has encouraged participation in CSBMs by member states. Just as importantly,
the effort has produced a set of standard CSBMs, developed report formats, established a
reporting calendar and identified a group of experts in the hemisphere for consultation on
CSBM issues.
The most important conference during this period of institutionalization was the
First Regional Conference of CSBMs, which took place in Santiago de Chile, in 1995.
The Santiago Declaration prioritized the need to identify the risks, threats and challenges
facing the Americas on the threshold of the next millennium" and the need to promote an
international climate of confidence, trust and peace based on cooperation.
More importantly, the Declaration contained a list of eleven specific CSBM
categories recommended for implementation by member states, building on the work


89
subject. Careful reading of this material provides definitive information on the resources,
agenda and rules of the security institutions studied. Table 2-7, below, summarizes this
discussion.
Table 2-7: Principal Indicators and Sources for Cooperative Security Institutions
Dependent
Variable
Factor
Indicators
Source
Organi
zations
Administrative
Capacity
Forums for negotiation (conferences,
working groups, commissions,
committees, etc.); decision-making
procedures; mechanisms to track
compliance, evaluate results and report
to members; enforcement
mechanisms.
Documents from
entities indicated
Issue
Agenda
Published agendas and work program
Documents from
Summits,
Ministerials, OAS
and IADB meetings.
Issue Linkages
Linkage in formal exposition of
security rules; state bargaining
positions
Documents from
Summits,
Ministerials, OAS
and IADB meetings.
Rules
Bindingness
Language of agreement
Published
agreements
Verification
Specific mechanism (reciprocal visits,
publication of data)
Documents from
Summits,
Ministerials, OAS
and IADB meetings.
Regime effectiveness, as opposed to institutional evolution, is more amenable to
quantitative approaches based on key indicators of security behavior. Table 2.8,
"Principal Indicators and Sources for Cooperative Security Behavior," summarizes this
approach.
Indicators of military transparency include participation in the OAS/IADB CSBM
inventory and the UN arms and defense expenditure registries; publication of defense


254
Dr. Eliezer Rizzo de Oliveira, an academic and government official involved in
the development of the national defense plan which underlies the Brazilian "white paper,"
made several interesting observations about the development of the defense plan at an
address at a symposium in Santa Fe, New Mexico, in 1997 (Rizzo, 1997). In his address,
Rizzo noted that Brazil, rather than developing formal associations for hemispheric
defense, preferred the flexibility inherent in a rich, overlapping network of CSBMs with
neighboring states:
Instead of creating a small, regional NATO, the Brazilian army prefers the
development of mechanisms for crisis management and reciprocal confidence,
such as cooperation in the areas of intelligence; personnel exchanges in respective
training centers; inclusion of observers in military maneuvers; periodic
consultations between military staffs; meetings of military technical experts;
combined exercises and maneuvers; student exchanges for officers and non
commissioned officers; systematic bilateral meetings; visits and athletic
competitions; combined social events; combined activities between the services
(navigation of international rivers, search and rescue operations, air mail,
operations during natural disasters, protection of flights, etc.), (p. 14)
The fact that full military integration remains a thing of the future is not
prejudicial to the institutionalist explanation. The fact that it is considered so seriously
does undermine realist explanations.
The United States is present in some of this activity. The Fuerzas Unidas exercise
series, which had consisted of bilateral computer simulations with each country in the
region, became a multilateral exercise focused on peacekeeping through the initiative of
the United States in 1995. The 1997 exercise included, in addition to the United States,
Brazil, Argentina, Bolivia, Paraguay and Uruguay ("Army Chief of Staff," 1997). Also,
UNITAS, which has relied on U.S. sponsorship since its inception, has gained a
multilateral dimension. The Argentine phase includes Brazil and Paraguay, as well as the


340
Buchanan, Paul G. 1998. "Chameleon, Tortoise, or Toad." In Jorge I. Dominguez, ed.,
International Security and Democracy: Latin America and the Caribbean in the
Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh Press, 265-288.
Burr, Raymond. 1955. The Balance of Power in Nineteenth-Century South America:
An Exploratory Essay. Journal of Latin American History 35 (1): 37-60.
Buzan, Barry. 1991. People, States and Fear: An agenda for International Security
Studies in the Post-Cold War Era. 2nd ed. Boulder, CO: Lynne Rienner.
Buzan, Barry, Ole Waever, and Jaap de Wilde. 1998. Security: A New Framework for
Analysis. Boulder, CO: Lynne Rienner.
"Cardoso discusses regional issues." El Mercurio (Santiago), 28 September 1997,
Foreign Broadcast Information Service (on line), 7 October 1997.
"Cardoso, Menem Declare Support for Mercosur." Agenda Estado (Sao Paulo), 27 April
1997, Foreign Broadcast Information Service (on line), 28 April 1997.
Carr, Edward H. 1964. The Twenty Years Crisis. 2nd ed. New York: Harper
Torchbooks.
Castro, Jorge. 1998. "Perspectivas Estratgicas de Argentina a Fin de Siglo." Argentina
y la Seguridad Internacional, Rut Diamint, ed., Santiago: Facultad Latinoamericana
de Ciencias Sociales, 33-45.
Caviedes, Csar. 1988. "The Emergence and Development of Geopolitical Doctrines in
the Southern Cone Countries." In Philip Kelly and Jack Child, eds, Geopolitics of
the Southern Cone and Antartica. Boulder, CO: Lynne Rienner, 13-29.
Chan, Steve. 1984. Mirror, Mirror on the Wall ... Are the Freer Countries more
Pacific? Journal of Conflict Resolution 28 (4): 617-648.
Chayes, Antonia H. and Abram Chayes. 1994. "Regime Architecture: Elements and
Principles." In Janne Nolan, ed., Global Engagement: Cooperation and Security in
the 21st Century. Washington, DC: Brookings Institution Press, 65-130.
"Chilean-Argentine Joint Exercises, Mercosur Security Discussed." La Nacin (Buenos
Aires, Internet Version), 29 July 1997, Foreign Broadcast Information Service (on
line), 30 July 1997.
Chile. 1998. Libro de la Defensa Nacional de Chile. Santiago: Naval Press.


48
valuable to less advanced states. These considerations make burden-sharing
arrangements difficult to achieve under the assumptions of realism.
Taken together, the three dilemmas to security cooperation discussed here free
riding, cheating and equitable distribution of benefits combine to form a powerful
deterrent to cooperative security behavior. Table 2.1 summarizes the problems associated
with cooperation.


298
The mission of COMBIFROM is to supervise fulfillment of obligations, promote
mechanisms to solve security problems and strengthen relations between armed forces. It
is composed of the intelligence chiefs of each country's joint staff, representatives from
each of the armed forces and the national police. Its functions include the execution of a
program of inspections and permanent liaison with counterparts. Information is shared
through a system of border communications and future plans include the sharing of data
through a binational computer network. While it is too early to judge the impact of this
organization on actual operations, the many commitments that it represents offer
important evidence that, at least in the case of Venezuela and Colombia, the trend is
towards a cooperative approach to address transnational security problems.
Military relations between Ecuador and Peru were understandably strained during
most of the 1990s. But since the last conflict, both Peru and Ecuador have reported their
efforts to reinitiate CSBMs ruptured by the 1995 war. In March 1996, the two countries
established a series of detailed measures for implementation in the conflict area and
committed themselves to supporting the peace process of the guarantor nations. Among
the measures proposed were mutual humanitarian aid, bilateral intelligence conferences
down to the regional level and coordination for search and rescue operations. Some of
the measures reported bear more symbolic than actual value. Ecuador reported a measure
in 1994 that would have established an Ecuadorian-Peruvian Center for Peace Studies,
to promote shared patrimony between Ecuador and Peru to and develop a culture of peace
between the two countries. At this time, Ecuador was in the process of rearming for a
war it would initiate the following year.


14
that promises a definitive end to a conflict which has engendered two wars and a series of
militarized disputes.
In Central America, Honduras and Nicaragua are actively demining their common
border, under the auspices of the Organization of American States and the Inter-American
Defense Board. El Salvador is participating with the deployment of Salvadoran technical
experts to Honduras. El Salvador, Honduras and Nicaragua conduct joint patrols in the
Gulf of Fonseca. The armies of those countries, along with Guatemala, have formed a
subregional security organization to promote cooperation. And all countries of Central
America are signatories to the recently enacted Democratic Security Treaty, which
contains many aspects related to cooperative security and military transparency.
Throughout the hemisphere, we find hundreds of examples of close coordination
to address common threats, unprecedented transparency in military matters, arms control
agreements, and significant changes in force structure, that demonstrate a new spirit of
cooperation in the area of security. The incidence of militarized conflict has dropped to
near zero. 10
The hemisphere has seen periods of institutionalized security cooperation in the
past, generally promoted by the United States in the face of a systemic threat, but
cooperation in the 1990s is occurring in an environment that lacks a consensus threat.
Furthermore, patterns of cooperative behavior reach across historic rivalries. The United
10 Wallensteen and Sollenberg (1998) compare incidence of Latin American
conflict from 1989-97 with other major regions. They report eight militarized disputes in
1989 (out of a global total of 47), and only 2 (of 33) in 1997 (internal conflicts in
Colombia and Peru).


262
monitor the agreements reached, including the international military presence in the
region (Child, 1996).
By 1991, with the Contras already demobilized in Nicaragua, the peace process
had gained significant momentum across the region. At the 1991 Presidential Summit in
San Salvador, many of the specific measures mentioned above were incorporated into a
draft Central American Security Treaty. This initiative, proposed by Honduras, would
not be signed until 1995, but much of its content was put into practice much earlier. For
instance, a 1994 agreement on a Procedure for Establishing a Firm and Lasting Peace in
Central America, established commitments in the area of national reconciliation, and
supported cease-fire arrangements and free elections in order to eliminate the source of
internal tension within each country. This approach was linked to commitments that
attacked the source of tension between states commitments to cut off assistance and
safe havens to irregulars or rebel movements and to verify, control and limit armaments
(OAS, 1994a).
This impressive legacy of security cooperation did not occur in a vacuum. The
peace process was complemented by a broader subregional integration process initiated in
1991 with the Protocol of Tegucigalpa, which put into place the Central American
Integration System (SICA). SICA is a broad-based program of integration that goes well
beyond the Central American Common Market, established in the 1960s. In effect, SICA
provides an umbrella organization for many subregional initiatives leading to integration,
including extensive cooperation on defense and security matters. The aim is to achieve a
level of integration that will allow Central American nations to bargain more effectively
over the terms of their participation in the international system (Rojas, 1996b).


57
and responsive to U.S. security priorities. Krasner (1991), in his study of power and
institutions, provides some examples of how those priorities might be expressed in
institutional arrangements: (1) determination of who plays; (2) establishment of the rules
of the game; and (3) manipulation of payoffs through tactical linkages.
In sum, realism maintains that cooperation is especially difficult in security
matters, due to the security dilemma, cheating, shirking and free riding, and relative gains
considerations. Security cooperation, unless supported by a large power, is only likely as
a short-term instrument to balance threats. When the threat subsides, so will the
cooperation. While some cooperation is possible through unilateral bargaining strategies
of defensively positioned states (Grieco, 1988a, p. 507), the degree of cooperation
actually taking place will be limited to small, incremental, reversible steps, in issue areas
where survival is not at stake and early gains do not translate into large advantages.27
Institutions matter little in this conceptualization. The institutions governing
cooperative activities in the hemisphere reflect underlying power realities. They are more
likely to be arenas of conflict than instruments to enhance cooperation. States may be
involved in talks, but are more likely to view them as an instrument of power politics
27 See Stein (1990) and Jervis (1999) for a more expansive view of the
possibilities of cooperation under anarchy. In his attempt to bridge the gap between
realists and neoliberal institutionalists, Jervis includes many cooperative strategies within
what he terms defensive realism. Institutions, even effective ones, are also in his view
consistent with some forms of realism because they are created when states see them as
instruments to reap benefits of cooperation. Stein adopts a similar approach, although he
is not sanguine about the possibilities of cooperation in security matters. In any case, I do
not employ such a broad version of realism, partly because it makes development of
contending hypotheses impractical.


317
cooperative security behavior. These organizations were subordinated to the formal
integration structure of SICA and guided by a well-defined concept for subregional
security, dubbed democratic security. The organizations developed an ambitious plan
of action to forward that concept. In short, all the formal elements of a cooperative
security regime were present. Actual cooperative behavior often fell short of the
institutional standards, but where the United States or other countries provided resources
to support implementation of the plan of action, cooperative behavior generally emerged,
precisely in the areas that promised the greatest absolute gains. 1
In the Andean Ridge countries, the evidence appears more supportive of realism.
Countries there were generally unable to cooperate to address the most significant threat
to the subregion the nearly intractable security problems along the borders of Colombia.
The United States was unable to spur multilateral cooperation and no linkages to other
unifying regimes were operable. However, cooperation between Colombia and
Venezuela was significant in the areas of transparency, verification and burden sharing.
The activity was supported by a strong binational organization and a rich network of
cross-border contacts. This security integration parallels the high levels of economic
integration that characterize Colombia and Venezuela. Elsewhere in the subregion,
Ecuador and Peru ultimately agreed to binding measures with verification necessary to
implement the peace accord in the Cenapa region brokered by the guarantor states. The
1 The trade integration system in Central America shows some of these same
characteristics. It features a formal structure and an ambitious agenda, while actual
progress implementing the measures lags considerably (Jata and Weintraub, 1997).


140
A follow-up regional conference on CSBMs took place in 1998 in San Salvador.
The declaration from that conference also produced a list of CSBM categories, deepening
and expanding the original measures. The specific nature of both of these declarations
will be considered in the discussion of the evolution of security rules, following this
review of organizations evolution.
The OAS also took a heightened interest in the IADB, financially dependent on
the OAS although not included in the Charter. In fact, the relationship between the OAS
and the Board consumed the Hemispheric Security Commission during the early years of
its existence. In 1992, Resolution 1181, the Inter-American Defense Board, tasked the
Committee to define the legal-institutional relationship between the Board and the OAS.
After numerous studies, conducted both at the Board and the OAS, no consensus
could be reached. The United States supported the Argentine position that the Board be
designated as an entity of the OAS.3 But other delegations were leery of introducing such
a substantial military element into the OAS. Brazil argued for rechartering the Board as a
special organization.^ But special organizations only include countries willing to support
their operating costs and hence do not offer an appropriate institutional design for a
hemispheric organization. Mexico preferred the status quo, meaning the Board would
remain outside the OAS Charter. In the end, the status quo was essentially accepted, with
3 What entity status would mean for the IADB is unknown, since the General
Assembly would have to provide a charter for the entity in the form of a resolution. The
Argentine proposal would have the Board rewrite its regulations and submit them to the
Permanent Council for approval.
^ The Inter-American Commission of Women is an example.


276
place under international supervision, which may have restrained states from taking
advantage of temporary imbalances. However, the reductions do correlate well to the
rules addressing defense postures. Those rules, as we saw, go beyond reductions and
touch on modernization as well, limiting the acquisition of new weapons systems.
At the regional CSBM conference in El Salvador in February, 1998, the opening
address by foreign minister Gonzalez Giner, host of the proceedings, laid out the basic
parameters for defense postures under Central Americas new model of democratic
security discussed earlier. In his address, the foreign minister reminded his audience -
foreign ministers from throughout the Americas that the Central American security
model was based on the supremacy of civilian control of the military, a reasonable
balance of forces, security of civilians and their property, sustainable development and
the protection of the environment (El Salvador, 1998a).
Interestingly, this doctrine has led, not to a decline of involvement of the armed
forces in security, but rather to role expansion. Military participation in police work, civil
works, literacy campaigns, medical assistance to rural areas and disaster relief has
increased over the course of the decade (El Salvador, 1998c). This pattern is common to
all countries in the region. The increase in activity is not, in reality, a doctrinal shift,
since the armed forces of the region have long been used to address security threats that
were essentially internal in nature. However, there is general acknowledgement that
these security threats are more salient than the possibility of armed conflict between
countries of the region (Rojas, 1996b).
The evidence, however, does not entirely favor institutionalism. Troop
deployments and defense doctrines offer mixed support to both realism and


181
The regimes share a component of civil-military relations. Arguably, elements of
this the education for peace program and the DMAs lie closer to democracy than to
security. The effort to assert civilian control over militaries is fundamental to the
preservation of democracy. It may or may not be essential to the preservation of peace -
the hemisphere has seen more than one instance that probably violates the precept that
democracies never go to war with each other (Mares, 1997).
In any case, the well-worn theme of civilian control by military establishments is
broadly accepted. The message of appropriate roles and the limits on non-traditionaf
activities has been received and understood by all. There are pockets of resistance and
the potential for backsliding still exists, but the hemisphere survived recent scares in Peru
(1992), Guatemala (1993), Ecuador (1995) and Paraguay (1996). Venezuela under
President Hugo Chavez may provide an interesting test case.
This discussion will be taken up again in Chapter 5 and in the conclusion of this
study. It appears that the impact of institutions and rules promoting democracy can affect
the pattern of institution building, rule making and perhaps compliance in the
hemispheric security regime. But since the major focus of this study is guided by the
competing hypotheses of realism and institutionalism, we will leave the discussion
dormant for the time being.
Burden Sharing
Enshrined in the Rio Treaty and the resolutions and missions of the IADB is the
traditional burden sharing activity of the Americas: the provision of military forces to
contribute to the collective defense of the continent. But, the Rio Treaty has fallen into
disuse. It is not viewed as an appropriate instrument for the 1990s and has suffered


THE NEW POLITICS OF COOPERATIVE SECURITY
IN THE AMERICAS
BY
WILLIAM B. FULLERTON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENT OF THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1999


332
relations emerges there, characterized by a cautious approach to hemispheric initiatives.
This is perhaps because there is not an active domestic constituency for military
transparency. In Mexico, where civilian dominance was long ago decided (Wager, 1996),
enthusiasm for most of the rules promoted by new hemispheric security institutions has
also been muted. ^
Thus, where questions of civilian control are settled whether favoring the
military or the civilian sector domestic civil-military relations appear to have little
impact on external security behavior. In polities where these questions are still open, we
find that civilian and military sectors seek international allies. Security preferences are
not completely specified by the external environment because a two-level game is taking
place.
A Modest Proposal
The issue of civilian control occasionally emerges within the security institutions
at the hemispheric level. This is most evident in the criticisms of the IADB as a relic of
authoritarianism and the Cold War. The anti-democracy charge against the IADB bases
its argument in the lack of civilian oversight of the institution. We have already reviewed
the steps taken by the IADB to bring it more firmly under the control of the OAS
7 Consider this historical anecdote on the question of whether to include a
permanent defense council in the OAS Charter during the 1948 conference in Bogota:
"After others had spoken at the small meeting, it was the turn of the Mexican general.
Standing behind him in the crowded room was the Mexican Foreign Minister, who
headed his country's delegation. Quietly the Mexican general said that, while he
personally was in agreement with his military colleagues from the other countries, his
government had adopted a contrary position and that, as Mexico's representative, he must
oppose the creation of the permanent military body (CFR, 1962, p. 47).


353
Mitrany, David. 1966. A Working Peace System. Chicago: Quadrangle.
Morganthau, Hans J. 1973. Politics Among Nations: The Struggle for Peace and Power.
5th ed. New York: Knopf.
Morrow, James D. 1993. Arms versus Allies: Trade-offs in the Search for Security.
In International Organization 47 (2): 361-388.
Mueller, Harald. 1993. The Internalization of Principles, Norms, and Rules by
Government. In Volker Rittberger and Peter Mayer, eds., Regime Theory and
International Relations. Oxford: Clarendon Press, 361-388.
Muoz, Heraldo. 1993. A New OAS for the New Times. In Virn P. Vaky and
Heraldo Muoz, eds., The Future of the Organization ofAmerican States. New
York: Twentieth Century Fund.
Muoz, Heraldo. 1996a. Collective Action for Democracy in the Americas. In
Heraldo Muoz and Joseph S. Tulchin, eds., Latin American Nations in World
Politics. Boulder, CO: Westview, 17-34.
Muoz, Heraldo. 1996b. The Dominant Themes in Latin American Foreign Relations:
An Introduction. In Heraldo Muoz and Joseph S. Tulchin, eds., Latin American
Nations in World Politics. Boulder, CO: Westview, 1-16.
National Defense University. 1995. Strategic Assessment: U.S. Security Challenges in
Transition. Washington, DC: Institute for National Strategic Studies.
National Defense University. 1996. Strategic Assessment: Instruments of U.S. Power.
Washington, DC: Institute for National Strategic Studies.
Nolan, Jarme E., ed. 1994. Global Engagement: Cooperation and Security in the 21st
Century. Washington, DC: Brookings Institution Press.
Nolan, Jarme E., John D. Steinbruner, Kenneth Flamm, Steven E. Miller, David
Mussington, William J. Perry, and Ashton B. Carter. 1994. The Imperatives for
Cooperation. In Jarme E. Nolan, ed., Global Engagement: Cooperation and
Security in the 21st Century. Washington, DC: Brookings Institution Press.
Nordon, Deborah L. 1996. "The Transformation of Argentine Security." In Richard L.
Millett and Michael Gold-Bliss, eds., Beyond Praetorianism: The Latin American
Military in Transition. Miami: University of Miami North-South Center, 241-260.
Norden, Deborah L. 1998. "The Remains of Latin America's Past: Military Legacies
and U.S. Influences." Journal of Interamerican World Affairs. 40(3): 103-116.


105
Institution building did not match U.S. expectations and the United States entered
into a period of extensive use of power, beginning with the Spanish-American War
(1898).9 U.S. interventions were focused on its immediate neighbors: Mexico, Central
America and the Caribbean. That focus has not changed substantially over time.10
Use of power gave way to the second moment of hemispherism, with the advent
of Roosevelts good neighbor policy. The main focus of efforts related to security was
the institutionalization of a system of peaceful resolution of conflict between states of the
region. Although an informal intermediary system" had taken hold over time! 1 efforts
9 Mares (1997) organizes U.S. security relations in Latin America into three
periods: (1) 1898-1933, extensive use of power; (2) 1933-1959, reliance on nonviolent
means; (3) 1959 to present, renewed intervention.
Kelly (1997) notes that U.S. armed intervention has always been associated
with instability on the U.S. southern frontier (Cuba: 1898, 1904-1925, 1962; Haiti: 1904-
1934, 1994-1999; Mexico: 1846-1848, 1915-1917; Panama: 1904, 1990; Grenada:
1983; Nicaragua: 1904-1933; 1980s; and the Dominican Republic: 1920-1928).
1 1 Dominguez (1983) reports that "from 1925 to 1942, there were ten instances
when 35 countries acted as intermediaries to cool off or resolve border or territorial
disputes. The United States did so seven times, European countries five times, South
American countries 18 times, and Mexico and Central America and Caribbean countries
five times" (Dominguez, 1983, p. 139).


247
Burden Sharing
Security cooperation related to burden sharing has grown at a phenomenal rate in
the region during the 1990s. The prospect of an integrated defense strategy for the
Mercosur countries is now considered seriously, based on the ever-increasing numbers
and types of combined operations conducted by the member countries. Much of this
activity is centered on the three countries of Argentina, Brazil and Uruguay. There is talk
of a specialization of security functions to make more efficient use of resources (Villagra,
1998).
Transnational security problems, although not as acute in the Southern Cone as in
other subregions, have spurred security cooperation among concerned countries. In the
area of contraband smuggling in the triborder region of Argentina, Brazil and Paraguay,
for instance, new mechanisms are in place to improve coordination. The Tripartite
Border Command has now been in operation for over a year ("Justice Minister," 1997).
Operations are not inherently military, although Brazil did use its armed forces to react to
criminality and smuggling in the region in 1997. This action led to public protests in the
area and the closure of Friendship Bridge for several days ("Document Outlines
Demands," 1998).
In the area of counterdrug operations, military cooperation has been limited,
reflecting both uncertainty in the use of the armed forces in this sector and the relatively
low threat this kind of activity presents to the Southern Cone (Bolivia excepted), at least
when compared to other parts of Latin America. Military cooperation has been primarily
focused on air interdiction, with Argentina conducting detection and intercept exercises
with both Brazil and Uruguay. Argentina and Uruguay have conducted aircraft refueling


357
Roett, Riordon. 1992. Brazil: Politics in a Patrimonial Society. 4th ed. Westport, CT:
Praeger.
Rojas Aravena, Francisco. 1994. Security Regimes in the Western Hemisphere: A
View from Latin America. In Lars Schoultz, William C. Smith and Augusto Varas,
eds., Security, Democracy and Development in U.S.-Latin American Relations.
Boulder, CO: Lynne Rienner, 171-197.
Rojas Aravena, Francisco, ed. 1996a. Balance Estratgico y Medidas de Confianza
Mutua. Santiago, Chile: FLACSO-Chile.
Rojas Aravena, Francisco. 1996b. Centroamrica: Una Nueva Agenda de Seguridad.
Paz y Seguridad en las Amricas 9. Santiago: Facultad Latinoamericana de
Ciencias Sociales, 3-7.
Rojas Aravena, Francisco, ed. 1996c. Medidas de Confianza Mutua: Verificacin.
Santiago, Chile: Facultad Latinoamericana de Ciencias Sociales.
Rojas Aravena, Francisco. 1997. Latin America: Alternative and Mechanisms of
Prevention in Situations Related to Territorial Sovereignty. Peace and Security in
the Americas 13. Santiago, Chile: Facultad Latinoamericana de Ciencias Sociales.
Rojas Aravena, Francisco. 1998. Transition and Civil-Military Relations in Chile. In
Jorge I. Dominguez, ed., International Security and Democracy: Latin America and
the Caribbean in the Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh
Press, 80-100.
Rosenau, James. 1990. Turbulence in World Politics: A Theory of Change and
Continuity. Princeton, NJ: Princeton University Press.
Rowe, David M. "World War Economic Expansion and National Security."
International Organization 53 (2): 195-232.
Ruggie, John Gerard. 1983. International Regimes, Transactions, and Change:
Embedded Liberalism in the Postwar Economic Order. In Stephen D. Krasner, ed.,
International Regimes. Ithaca, NY: Cornell University Press, 195-231.
Romero, Carlos A. 1998. Exporting Peace by Other Means. In Jorge I. Dominguez,
ed., International Security and Democracy: Latin America and the Caribbean in the
Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh Press, 151-166.
Russell, Roberto. 1988. "Argentina: Ten Years of Foreign Policy Toward the Southern
Cone." In Philip Kelly and Jack Child, eds., Geopolitics of the Southern Cone and
Antartica. Boulder, CO: Lynne Rienner, 67-82.


78
Table 2-4: Contending Hypotheses at the Hemispheric Level for Behavior
Dependent
Variables
Realist
Hypotheses
Institutionalist
Hypotheses
Security
Given institutions with the administrative capacity to carry out the
Behavior:
functions of regime management (data collection, analysis and
reporting) and rules promoting transparency, defense reductions and
burden sharing ....
Transparency,
4. Transparency measures will
4. Transparency will be
not reveal information of
considerable, providing
Defense
value, reporting instead
relevant and reliable
information that is already
information.
Postures,
largely public.
Burden
5. Defense reductions are
5. Defense reductions will take
unlikely and in any case will
place where transparency
Sharing
reflect perceived threats.
6. Burden sharing is unlikely
provides assurances of
reciprocity.
unless sponsored by the
6. Burden sharing will take place
United States or resulting
where transparency provides
from balancing and/or
assurances of reciprocity. The
bandwagoning. The presence
presence of transnational
of transnational security
security problems makes
problems makes cooperation
cooperation more likely
even less likely because of
because of the added potential
the added potential for
relative gains.
for absolute gains.
Subregional Level
What modification of these hypotheses might be necessary at the subregional
level? There is good reason to believe that the basic tenets of realism and institutionalism
will operate there with little modification. Under realism, the dynamic of security


64
states discount the value of agreements on the basis of past performance (Keohane,
1984, p. 105), a reputation for compliance is valuable due to the prospect future gains.33
The costs of defection are raised because of the existence of institutions, just as the
transaction costs of cooperation are lowered.
Institutions that can be effective in this sense must have the administrative
capacity to support negotiations and collect, analyze and report data. These are
essentially performance criteria, which can be observed and measured in security
institutions.
Institutions rarely operate in a vacuum. They may overlap issue areas with other
institutions (Keohane, 1986), creating multiple links between states that connect, for
instance, rules on cooperative security with regional mechanisms to promote respect for
human rights, encourage economic development and strengthen democratic governance.
In the broadest sense, the effect of these issue linkages is to create a dense web of
constraints that fortifies rules and makes rule-based behavior more likely.
More specifically, linkage is important to institutionalist theory in several
concrete ways. First, linkage allows for more iteration, increasing the opportunities for
tit-for-taf reciprocation (Oye, 1986). Second, linkage may erode hierarchy, by allowing
small states to link security issues to other areas (Keohane and Nye, 1989). Third,
linkages increase the benefits of cooperation and the costs of defection. They allow for
33 Institutions can also reward cooperative behavior through provision of
leadership positions (president of the OAS Committee on Hemispheric Security, for
example) and sponsorship of security conferences.


175
This set of resolutions was matched by resolutions commending the progress
made in the Central American Demining Program. Since that program is related to
security cooperation, will be considered in the next section. However, the removal of
landmines on the borders of Nicaragua with Honduras and Costa Rica should also be
considered a substantial change in the defense postures of those countries.
Aside from a commitment to maintain reasonable balances and efforts to
remove and ban the use, production and stockpiling of landmines, there has been little
rule-making activity that impacts conventional arms control. For comparative purposes,
we will quickly review progress in a related area control of the illicit trade in small
arms, championed by Mexico. This effort culminated in the adoption in 1997 of the Inter-
American Convention Against the Illicit Manufacturing of and Trafficking in Firearms,
Ammunition, Explosives, and other Related Matters.
The purpose of the convention is to prevent, combat and eradicate the illegal small
arms, ammunition and explosives trade. It establishes a binding agreement with a
hemispheric consultative committee, structured with a secretary and a network of national
experts. Binding rules for domestic policy include: punishment for offenses and
recognition of the extraditable nature of the offenses; strengthening security of arms and
mechanisms for detecting and confiscating illegally obtained arms; and requirements for
marking legally manufactured weapons, establishing export/import licensing and
strengthening customs procedures. Cooperative measures between states include
exchange of information, experiences and appropriate technologies, cooperation with
trace requests, and training programs. Also countries are required to identify a central
national office for coordination purposes.


17
burden sharing in these areas. A pattern of security cooperation consistent with this
general optic would largely reflect U.S. interests.
Another possible explanation goes beyond a focus on state power and interest,
drawing on the proliferation of new international organizations and the growing
transnational linkages characteristic of a globalization of economic, cultural and political
issues. The globalization literature emphasizes the important role of new norms 14 and
institutions in organizing the "turbulence" of world politics (Rosenau, 1990; Vaky, 1993;
Muoz, 1996). Despite the recent bearish turn in the United Nations (after euphoric
predictions following the end of the Cold War), the general belief is that institutions can
help, especially if carefully designed with incentives that appeal to national interests.
These institutions are thought to be easier to create in regional organizations, such as the
OAS (Lepgold and Weiss, 1998; Lambom, 1998).
Corrales and Feinberg characterize the region in the 1990s as passing through a
high point of what they term "hemispherism": the active attempt by the nations of the
hemisphere to redirect their foreign policies in favor of closer and coordinated
cooperation with one another" (Corrales and Feinberg, 1999, p. 2). Unprecedented
cooperation in the 1990s is based on acceptance of new norms, creation of new
14 For example, some argue that a new norm on democracy and human rights
appears to be challenging the old, statist norm of sovereignty and non-intervention
(Cronin, 1998).
15 The importance of organizations as ways to institutionalize norms is
emphasized here. Ideas untethered from institutions are unlikely to have lasting impacts.
Institutionalist treatments of Kants perpetual peace are not vulnerable to assaults such
as Carrs (1964) attack on WWI idealism.


201
Pattern Across Sectors
Transparency/verification
As noted, transparency and verification measures are the most frequently reported.
In fact, almost 90% of the total measures reported were from this sector. Within the
sector, about 30% were strictly transparency measures, while 59% consisted of personnel
exchanges that can be taken as a substitute for hard verification. These are the
measures contained in Santiago category E. The charts below show the detailed
distribution.
1997 CSBM Distribution
1997 CSBM Distribution
by Santiago Category
by Geographical Scope
LEGEND
A.
Prior notification of exercises
Others
B.
Information on transfers/budgets
Prevention of transportation incidents
C.
Exchanges of defense doctrine
Cooperation during natural disasters
D.
Controls on conventional weapons
Training on CSBMs
E.
Personnel exchanges and
Meetings on security of island states
visits to military installations
Education programs for peace
Figure 4-1: Distribution of CSBMs (1997)


129
Conclusion
In this reading of balance of power and geopolitical analysis, the realist tradition
has produced the most enduring explanatory model for conflict in the hemisphere -
power restraining power.41 As Mares notes, "the system has essentially been
dominated by balance-of-power mechanisms, with the United States and some Latin
American states imposing their will on weaker members" (Mares, 1997, p. 197). Security
cooperation has traditionally worked to reinforce balance of power mechanisms, not
overcome them.
The politics of hemispheric security, during this epoch of general U.S. dominance,
reflect in large measure a pattern of institutional bandwagoning in response to U.S.
demonstrations of assurance support for multilateralism on the one hand and payment
of side benefits (offering outlets, preferential logistics agreements) to potential balancers.
It is evident that over the period reviewed U.S. sponsorship has been key to the success of
hemispheric security organizations, which the United States used to help multilateralize
and institutionalize its own initiatives.4^
But clear examples of institutional balancing are also present and provide a good
indication of the complicated security politics that would reemerge in the 1990s. That
41 Caviedes (1988) notes that geopolitical thinking can be consistent with
integration, with a focus on the economic value of cooperating to mutually exploit
resources and develop a strong economic nexus along critical lines of communication.
But the promise of integration is susceptible to relative gains problems.
42 Although subregional cooperation independent of the United States was also
possible: Argentina, Brazil and Uruguay organized tripartite patrols in the Atlantic in
1940 and Chile and Peru met to discuss joint defense (Mecham, 1962).


331
promotion of military transparency is at odds with Argentina and Brazil both those
countries participate in the UN registries, have produced defense white books and are
seeking ways to enhance the transparency requirements established at the hemispheric
level. Chile has taken all of these steps, but has been far less willing to move forward
with subregional burden sharing. It is this lag in the Chilean pattern that is puzzling. The
answer may relate to the internationalization of its democratization process. It is also
possible that civilian democratizers sought support in the hemispheric security regime to
help force its military establishment to become more transparent. They are far less
interested in other forms of cooperative security.
When more factors in the civil-military balance are considered, more extensions
seem plausible. Colombia, for instance, in the process of internationalizing its fight
against guerrillas and drug traffickers, has adopted the rules of the democracy regime,
which may limit the ways in which its military can be employed. Consequently, there is
less military involvement in the drug fight and less military burden sharing with
neighbors (Venezuela excepted) than might be expected. Argentina's pattern of
aggressive involvement in peacekeeping operations has been explained as a way for the
civilian sector to provide its military with a democracy-friendly, external role (Pala, 1998;
Norden, 1996). These linkages between domestic civil-military relations and the
international community merit further investigation.
In polities with closed, pacted transitions, such as Brazil, civil-military relations
appear to have less of an impact on interstate security behavior. The Brazilian military is
still relative autonomous, its opening having been implemented according to the
military's terms (Manwaring, 1996; Zaverucha, 1993). A more stable pattern of security


204
CSBM instruction one of the transparency measures is a basic part of the College
curriculum, presented to military and civilian personnel.
Some of this same sort of activity goes on in other hemispheric organizations,
particularly the specialized conferences of the armed forces. Subregionally that sort of
activity is more significant, because the information exchanged can be more closely
scrutinized. This is even truer at the bilateral level. In fact, most hemispheric and
subregional institutions offer opportunities for bilateral contact that results in information
and intelligence exchanges.
What is the direct impact of the United States in terms of the transparency
pattern? The United States promotes transparency measures in part through its own
example distribution of its defense while paper at the first DMA is evidence of this.
Five countries have followed suit since the United States disseminated its report.
Likewise, many of the hemispheric institutions mentioned above function with
considerable U.S. support.
But there is little reason to believe that the overall trend is due to the U.S. security
agenda. While it is true that the United States has exchange programs with many
countries and supports programs of visits to its installations and exercises, these programs
were significantly underreported by the United States. Other forms of support, such as
the open skies initiative" have been offered but have yet to gain wide acceptance in the
hemisphere.
Defense postures
Regime rules that affect defense postures include arms control agreements and
participation in education for peace. There is little that can be said about regime


295
corruption are transnational security issues that require some form of subregional
cooperation.
Colombia will not be able to solve these significant problems without
cooperation. It is true that Peru, acting virtually alone in the early 1990s dramatically
improved its security solution vis--vis guerrilla activity. However, the costs of
independent action were very high and Peruvian tactics provoked harsh criticism from the
inter-American community for disregard of democratic institutions and human rights. In
any case, it is doubtful that such a strategy could now succeed. Since the early 1990s, the
FARC, the Ejrcito de Liberacin Nacional (ELN) and paramilitary militias have become
increasingly involved in the drug trade, which now provides them with an important
source of financial support (Marcella and Schulz, 1999).
The conflict between Ecuador and Peru, at least for the time being, appears to be
resolved, although solution to the conflict required third party mediation from outside the
subregion. The guarantor countries of Argentina, Brazil, Chile and the United States
helped create the conditions for peace (an example of security cooperation among
Southern Cone countries). Ultimately, a comprehensive peace agreement emerged that
included side benefits intended to redress relative gains problems inherent in the border
settlement between Peru and Ecuador. But peace negotiations were accompanied by arms
build-ups in both countries and it is too early to confirm that a new relationship, based on
a shared set of security priorities, will take hold.
What role might institutions and rules play in the resolution of security problems
in the subregion? Let us now look at those institutions in more detail.


239
Submission). In addition, Chile is currently conducting research trips to the Antarctic
with Brazil (Brazil 1997 CSBM Submission).
Chilean contacts with the Uruguayan armed forces occur as well, but at a more
reduced level. Chile has, for instance, conducted flight simulator training with
Uruguayan pilots. Worth consideration is the fact that many countries of the subregion
take part in Chile's annual air show (Chile 1996 CSBM Submission).
Contacts between Paraguay and the other states in the region have been less
extensive, although all major countries of the region have institutionalized meetings
between their chiefs of the armed forces and those of Paraguay (1996 CSBM
Submission). In addition, Uruguay reports that since 1970, both countries have
participated in Operation Cimarron (Uruguay 1996 CSBM Submission). Nevertheless,
the level of security engagement between Paraguay and Bolivia and the rest of the
subregion is relatively low.
Surprisingly, given the density of intraregional exchanges, the United States is the
most important partner for both Argentina and Brazil in terms of numbers of contacts
reported under the rubric of transparency/verification. The United States is also an
important partner of Chile and Uruguay. It is difficult to see, however, how this might
affect the subregional security system since the contacts are not multilateral. The pattern
more clearly demonstrates U.S. attempts to maintain strong bilateral relations with
countries in the region.
Nevertheless, bilateral contacts can make a difference. U.S. contact with
Argentina is denser and deeper than with the other countries and includes a
confidentiality accord between the two countries which may actually erode regional


100
too isolated to be part of the system initially. However, the emergence of military might
in Paraguay under Francisco Solano Lopez affected the subregional balance of power and
led to the formation of the Triple Alliance in 1865, uniting Argentina, Brazil and
Uruguay to overthrow Lopez. Once defeated, Paraguay was not partitioned as some
envisioned, but rather allowed to remain as an independent state a buffer state between
the larger contending powers, which quickly resumed their rivalry.
On the Pacific Coast, conflict between then New Granada and Peru over
Guayaquil was partially resolved through the establishment of the state of Ecuador (1830)
- its independence guaranteed by its two larger neighbors. Farther south, the formation of
the Peru-Bolivian Confederation in 1835 led to war with Chile. Both Argentina and New
Granada also opposed the new confederation. In 1838 the Confederation was defeated
and a balance of power system on the Pacific Coast then began to take shape. Stable
power relations between Bolivia, Chile, Colombia, Ecuador and Peru were essential to
maintaining peace between those states. The issue of Ecuadorian sovereignty continued
to concern the larger powers, although increasingly rivalry between Chile and Peru
dominated the system. By the 1870s, the growing economic interests of Chile in the
Bolivian desert coast sparked Peruvian resistance to Chilean power.
As states developed more significant capability, the Rio de la Plata system and the
Pacific system began to interact more frequently. Some of this was evidenced by the
middle of the century; with the defeat of Juan Manual de Rozas in Argentina, growing
Brazilian influence in Uruguay and Chilean encroachment on Patagonia threatened the


49
Table 2-1: Dilemmas of Security Cooperation
Security
Behavior
Absolute
Gain
Collective Good
Problem and
Non-Cooperative
Behavior
Cheating
Problem and
Non-Cooperative
Behavior
Relative Gains
Problem and
Non-
Cooperative
Behavior
Military
Better
Transparency
Increased
Some states
Openness
security
approximates a
vulnerability if
may rely on
calculation -
collective good.
weaknesses are
secrecy more
avoids worst
exposed.
than others, to
case analysis
Strong incentive
hide
to free ride.
weaknesses or
disguise
States may not
aggressive
provide
intentions.
information or
may provide
Since
partial or
transparency
misleading
may not be
information.
reciprocated, it
is unlikely to
develop.
Reduced
More
No collective
Increased
Some types of
Defense
resources
goods problem;
vulnerability if
reduction may
Postures
available for
gains are private.
other states do not
benefit some
development
reduce defense
more than
and other
postures.
others.
goals
States may not
Reductions in
reduce postures or
some areas will
may offset
not be possible
vulnerabilities by
or will be offset
seeking alliances
in other areas.
(balancing).


310
for negotiation, and extended the shadow of the future as countries come to anticipate
more and broader cooperation.
There are outlyers. In the Southern Cone, the hyperactive peacekeeping program
of Argentina stands out, as does the deep involvement in the transparency regime by
Chile. That transparency is a puzzle, given that Chile is neither reducing its defense
posture or engaging in burden sharing activities at levels commensurate with the other
large states of the subregion. In Central America, El Salvador stands out as leader in the
areas of transparency and subregional burden sharing. These anomalies can perhaps best
be approached by connecting civil-military relations internal to the state with the
international context in which it operates. We will explore this possibility in the
conclusion of the study.


343
Desjardins, Marie-France. 1996. Rethinking Confidence-Building Measures (Adelphi
Paper 307). Oxford: Oxford University Press.
Deutsch, Karl W. et al. 1957. Political Community and the North Atlantic Area.
Princeton, NJ: Princeton University Press.
Deutsch, Karl W. 1963. The Nerves of Government: Models of Political Communication
and Control. London: Free Press of Glencoe.
Diamond, Larry, Juan J. Linz, and Seymour Martin Lipset. 1989. Democracy in
Developing Countries: Studies in International Political Economy and
Development. Volume 4. Boulder, CO: Lynne Rienner.
"Document Outlines Demands for Unblocking Bridge." RPC Telvision Network
(Asuncin), 3 June 1998, Foreign Broadcast Information Service (on line), 2 June
1998.
Dominguez, Jorge I. 1983. Political Relations in the Western Hemisphere." In Virn P.
Vaky, ed., Governance in the Western Hemisphere. New York: Aspen Institute for
Humanistic Studies, Praeger.
Dominguez, Jorge I, ed. 1998. International Security and Democracy: Latin America
and the Caribbean in the Post-Cold War Era. Pittsburgh, PA: University of
Pittsburgh Press.
Dos Santos, Theotonio. 1970. The Structure of Dependence. Reprinted in Robert J. Art
and Robert Jervis, eds., International Politics: Anarchy, Force, Political Economy,
and Decision Making, 2nd ed. Boston: Little, Brown, 1985, 303-311.
Downes, Richard. 1996. Emerging Patterns of Security Cooperation in the Western
Hemisphere. North-South Issues 5 (1). Coral Gables, FL: University of Miami
North-South Center.
Downes, Richard. 1999. Building New Security Relationships in the Americas." In
Donald E. Schulz., ed., Conference Report: The Role of the Armed Forces in the
Americas: Civil-Military Relations for the 21st Century. Carlisle, PA: Strategic
Studies Institute, 13-30.
Druetta, Gustavo, Luis Tibiletti, and Marcela Donadio. 1992. "Los Nuevos Conceptos
sobre Seguridad Estratgica Regional." Seguridad Estratgica Regional 1: 58-67.
"Economic Problems Seen Lessening Interest in Arms." Gazeta Mercantil (Sao Paulo),
6 August 1997, Foreign Broadcast Information System (on line), 8 August 1997.


241
while Argentina encouraged contacts which might confirm its relative weakness, the one
exception to this pattern may be significant the lack of contact between the Chilean and
Argentina armies. The absence of contact between Chile and Bolivia is also noteworthy
in this regard.
What is also quite evident is the presence of the United States in these
arrangements, especially with Argentina, but also with Brazil, Chile and Uruguay.
It is critical to note that more contact is present between Argentina and the United States
than any other dyad. Substantial U.S. security cooperation also is present with Brazil and
Uruguay, but the Argentine-U.S. connection is in a class by itself. Here is our first
indication that Argentina may be seeking a countervailing force to the power of Brazil
and perhaps Chile through an association with the United States.
In sum, then, we see significant evidence of cooperative security behavior in the
area of transparency/verification. Much of this behavior does seem to follow from
subregional security institutions. However, balancing does emerge from the evidence as
well. Let us now turn to an examination of defense postures.
Defense Posture
This section will first address military doctrine to try to gauge whether
subregional security is a priority for the major countries of the Southern Cone. Security
doctrine establishes the primary threats considered priority concerns for defense. Next,
we will look at defense orientation to see to what extent countries are willing to accept
risk in national security to address regional concerns. What do patterns of modernization,
reductions and redeployments show?


36
Other characteristics help define specific security behavior. First, capabilities of
member states must be transparent. Information must be reliable (verifiable), relevant
and timely, providing early warning of defection. Second, military postures should be
kept at low levels, with limited armaments capacity and constraints on exercises, ceilings
on forces, restrictions on deployments (Daadler, 1992). Third, a cooperative security
regime must have operational capacity to respond to crises at varying levels, through
good offices, formal mediation and ultimately through collective response and burden
sharing agreements. This operational capacity need not include a command element (as
is the case with NATO), but some mechanism should be available for organizing a
collective response to violations of the regime norms.
These three specific injunctions yield the behavior clusters that form the key
components of rule-based security behavior: transparency and verification; reduced
defense postures; and burden sharing (collective capacity to sanction offenders).
There are clear absolute gains from cooperation in these three behavior clusters.^
Military openness allows states to make a more informed choice about their security
needs. Their calculations can thus avoid a "worst case" analysis requiring them to expend
more resources than necessary. Reduced defense postures offers a similar absolute gain.
With a lower requirement for defense, states can spend resources on other priorities, such
as economic development and social programs. Regional burden sharing takes this one
5 The presence of absolute gains is an important prerequisite for cooperation
under the assumptions of institutionalism. See Keohane (1984), Keohane and Martin
(1995) and Hasenclever, et al. (1997). Military relations among states must be seen as a
problem of common security (Nye, 1991, p. 162).


41
Institutions simply reflect the underlying power and interest of participating states.
Institutions are arenas for acting out power relationships (Mearsheimer, 1994/5) or
representations of power aggregation in the form of alliances. 12
This does not mean that security cooperation never occurs, just that it is closely
circumscribed and not likely to result from institutions. To better understand why
cooperation under the assumptions of realism is so difficult to achieve, it is necessary to
examine more closely the theoretical literature on the standard dilemmas of interstate
cooperation.
Olson (1964), and others, laid out a logic of collective action to show why
"collective goods" are normally supplied at suboptimal levels. Because collective goods
are public or non-discriminatory in the sense that no state can be excluded from reaping
the benefits, states have powerful incentives to shirk and free-ride. 13 States will tend to
allow others to bear the costs for the provision of collective goods because the benefits
accrue more or less equally to all. Clearly, if each state follows this logic, collective
12 Classical realists such as Aron (1973) and Morganthau (1973) recommended
formation of alliances as stabilizing, balance-of-power mechanisms. Alliances are
institutions, based on Snyders (1990) definition: formal associations of states for the
use (or non-use) of military force, intended for security or aggrandizement of members"
(p. 104). But latter day realists have disagreed on the impact of alliances, as we will
discuss shortly.
12 Olson also listed as a key characteristic 'jointness of supply,' meaning an
individuals consumption does not preclude consumption by others" (Gowa, 1989, p.
1245). Public goods theorists also employ these concepts (Kindleberger, 1973).


125
closer to home, as the U.S. interventions in Grenada (1983), the Central American Civil
Wars (1981-1986) and Panama (1989) clearly demonstrate.34
Balancing at the Subregional Level
Even during periods of extensive U.S. influence, the distribution of power among
the other states of the hemisphere remains key to understanding regional security
patterns. U.S. military assistance during WWII and Cold War "had its major impact only
when it could be used to reinforce national interests" (Mares, 1997, p. 200).3 3 Patterns of
security competition and cooperation continue to match the "checkerboard of informal
ententes and traditional rivalries laid out in the previous century (Kelly, 1997).
As tensions during the centennial of the War of the Pacific demonstrated, the
desert north of Chile remains one of the most geopolitical sensitive regions on the
continent, despite the Treaty of Ancn (1929), which settled the territorial dispute
between Peru and Chile. According to the provisions of the treaty, Chile requires Peru's
approval to grant Bolivia an outlet to the sea through formerly Peruvian territory. Despite
Peru's endorsement of the Ayacucho Declaration (1974), which pledged to give "fullest
understanding to the land-locked position of Bolivia" (Rengger and Campbell, 1995, p.
315), Peru exercised its veto in 1976 when Chile offered Bolivia a corridor that included
34 Although even here, balancing was evident: Panama, Costa Rica, Mexico and
Venezuela provided support to the Sandinistas in Nicaragua in the 1970s and 1980s
(Mares 1997) support that was in part countered by assistance to the Contras from the
United States and Argentina.
33 Mares (1997) supports this assertion by noting that "when the United States
provided Brazil with equipment and supplies to defend its bulge on the Atlantic, the
Brazilians preferred to focus resources on their southwestern border with Argentina"
(Mares, 1997, p. 200).


93
Table 2-8 continued
Factor
Indicator
Source
Observation
Defense
Postures
Arms Production
Arms transfers
Arms control
agreements entered into
Authoritative
military analysis and
arms control
documents Janes,
SIPRI, IISS,
UNCDA, ACDA,
etc.
Burden
Sharing
Coordination to solve
technical problems -
protocols for border
situations
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
While coordination does not
necessarily represent a strong
commitment to cooperative
security, the conduct of
actual operations and training
exercises between countries
does, because of what those
activities reveal about
national doctrine, readiness
and capabilities.
Furthermore, such activity
can promote interoperability
- development of common
doctrine, common equipment,
training and logistics. In the
final analysis, it can also lead
to military specialization a
division of labor that makes
multilateral security
operations more effective, but
may leave countries much
more vulnerable; Uruguay
specializes in Peacekeeping
Operations, for instance.
Other examples include
developing force structures
based on light aircraft and
riverine craft for counter
drug operations or focusing
on humanitarian operations
such as demining.
Coordination to solve
technical problems -
protocols for safe
transport by ground, air
and sea
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
Bilateral and
multilateral exercises
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
Bilateral and
multilateral
humanitarian operations
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
Bilateral and
multilateral
peacekeeping operations
As reported in
IADB inventories
(95,96,97), OAS
inventories (96, 97,
98),
Utilization of common
training centers (e.g.:
School of the Americas,
Center for Hemispheric
Defense Studies, the
Argentine Peacekeeping
Operations Center, the
Central American Police
Academy)
Data from
institutions
involved.


294
the south Peru continued to compete with Chile. These rivalries still vibrate at the core of
security relations in the subregion. Tension remains high between Colombia and
Venezuela; the 1987 Caldas incident was categorized as a military dispute, but lower
level conflict in the 1990s has also been observed (Mares and Bernstein, 1998).
Significantly, while much of Latin America downsized its military forces in the 1990s,
the countries of this subregion did not (Dominguez, 1998).
However, it is difficult to cast the subregion as it entered the 1990s as one solely
based on power-constraining-power. Venezuela, despite border disputes with both
Colombia and Guyana and an exposed national territory (Kelly, 1990), has remained
remarkably pacific throughout this century. Venezuelan troops have not been involved in
outside armed conflicts and until their deployment to Central America under ONUCA
had not taken part in any military operations under the OAS or the United Nations, with
the exception of small observer missions (Romero, 1998). In addition, Venezuela has
taken pains to reduce conflict with Colombia. Torres (1998) argues that Venezuela
expanded its economic ties with Colombia precisely to overcome the effects of their
border dispute. And Colombia, despite a long history of internal violence, has not taken
territory by force from another state since the wars of independence.
The security dilemma does not appear to operate as the principal dynamic. More
reasonably, the insecurity dilemma centered on Colombia drives competition and
cooperation. The presence of powerful guerrilla movements based in the headwaters of
the Amazon and along all Colombia's international borders, as well as a shared drug
trafficking problem, cross-border criminality, displaced persons, ecological damage, and


95
than institutional politics of hemispheric security. Still, institutional politics had an
important impact that is well worth an investigation.
U.S. policy makers designing hemispheric defense plans during WWII were
aware of the difficulties inherent in a multilateral security policy. But the options of
purely bilateral and unilateral policies were expensive, and in a theater with few core
interests, the United States was not prepared to muster the resources necessary for such a
course of action. Making the case for multilateralism required convincing Latin
American states that Axis countries posed a concrete threat to the entire region. Until the
Japanese attack on Pearl Harbor consensus on that point was not overwhelming. This is
especially true of the Southern Cone countries, where German immigrants were an
important presence. Nevertheless, over time the United States was able to win the
interest of the other states of the region, working both bilaterally and multilaterally to
establish a system of defense for the hemisphere.
However, from the outset, there was a reluctance to organize hemispheric security
around a permanent military organization, such the IADB, which Latin American states
assumed would be powerfully influenced by the United States. But institutional
balancing lay dormant as long as mutual interests were served by the institution. That
changed once the United States redefined the threat from external to internal. U.S.
pressure on Latin American states in Caracas (1954) to make the presence of domestic
communist movements a trigger for invoking the Rio Treaty stirred resentment and
prompted the institutional balancing that ultimately undid the post-WWII multilateral
security system.


143
have elapsed since then.8 There is no lack of technical studies that address the question
of the Board and its position in the Inter-American System. This question is basically
one of organizational design: what is the best mechanism for providing military
advice/expertise on shared security concerns to OAS member states? Is some sort of a
permanent military body with links to the OAS warranted or are ad hoc defense
organizations similar to those envisioned by the OAS Charter and the Rio Treaty
sufficient?
But technical, institutional fixes have always come up short. This is because the
issue is primarily a political one that lies along the fault-line of U.S.-LA security policy:
how best to institutionalize security cooperation without degrading the principles of
sovereignty and non-intervention? This is an issue that invariably attracts institutional
balancing.
The end of the Cold War and the consolidation of democracy in the hemisphere
brought new considerations to bear on the question. Critics pointed out that the Board,
never intended as a permanent organization in any case, had lost its primary mission, the
planning and preparation for collective defense of the Continent. Furthermore, the Board
had only a weakly institutionalized legal relationship with the OAS and hence little
political oversight; it offered a poor model of democratic civil-military relations for a
region in the process of consolidating democracy. Finally, the Board, unable to keep
8 The 1995 Report of the Chairman of the Special Committee on Hemispheric
Security Concerning Cooperation for Hemispheric Security (OAS, 1995) explains that
the question of the Board was dropped in 1995, in order for the Committee to move on
and implement the provisions of General Assembly Resolution 1240 to access the Board's
technical competence.


2
Institutions and new rules regarding the use of military capabilities are emerging,
but at the same time interstate rivalries and power politics remain important features of
the foreign policies of all states in the region. Security cooperation itself is not new to the
hemisphere, but rather a normal tool of foreign policy that occurs within the framework
of the shifting informal ententes that characterize many security systems. A critical view
of the "new cooperation" might point out that much of the activity is restricted to the
conference rooms of international organizations that are far removed from military bases
where weapons modernization is ongoing and from borders that remained armed and
mined. The sources of conflict have not disappeared. 1 They have endured despite
previous attempts to engineer solutions that would replace power as the ultimate arbiter in
the international relations of the region.
What theoretical optic is available to sort through these conflicting tendencies?
The Latin American experience has been hard on theories. The field of political
economy, for instance, has seen the rise and fall and occasionally the rise again of
theoretical approaches as disparate as modernization theory, hegemonic stability theory,
Marxism, dependency theory, and bureaucratic authoritarianism. Explanations for regime
change have covered the theoretical waterfront as well. Consensus explanations of
macro-social change have been elusive.
1 For instance, border disputes, which have been the trigger for most militarized
disputes in the past two decades, are still an important source of conflict (Mares, 1997).
Downes (1998) cites the Secretary-General of the Organization of American States as
stating in 1995 that thirty-one border disputes were still active. For a listing of the more
than one dozen serious border disputes, see Rojas (1997).


68
an opportunity to voice preferences, build coalitions and reinforce negotiating positions
through linkages.37 Small states can leverage these advantages to engage in balancing
behavior that would not be possible in the absence of the institution. Furthermore, in
order to build support for its agenda, the United States may need to provide side benefits
and security outlets which ultimately strengthen the bargaining position of the smaller
states. While an equilibrium may be theoretically possible, a more likely outcome is an
eventual retreat from multilateralism, leaving the institution void of resources and
effectiveness. Mutuality of interests declines and the shadow of the future shortens as
smaller states come to expect a return to unilateralism on the part of the United States.
The determinants of U.S. security behavior are largely beyond the reach of
hemispheric institutions. The primacy of global interests and the impact of the domestic
agenda are clearly more important to the formulation of U.S. security policy than regional
concerns. When global concerns dominate, support for regional cooperative security and
respect for regional rules take a backseat to other strategic priorities. When domestic
concerns dominate, U.S. offers of assistance (arms, training, resources, information) can
be subject to intrusive end-use agreements which fall outside of the regime. The
certification process which oversees provision of counterdrug assistance and the Leahy
amendment which makes military assistance contingent on human rights performance are
two examples.
37 Jervis (1999) claims that institutions, even if created by the systems most
powerful states, can erode power by giving voice, legitimacy and forms of influence to
weak or new actors (p. 61). Institutions could restore some symmetry to U.S.-Latin
American relations. This finding can also be derived from Krasner (1985), although his
argument is a realist one.


207
to prevent transportation incidents. Of the remainder, most were humanitarian operations
- especially disaster relief and demining operations. Peacekeeping and counterdrug
operations and exercises were also reported. Small island security concerns, which often
involve security cooperation to address transnational threats such as illegal immigration
and drug trafficking, made up slightly less than 0.5% of the measures reported.
Very few measures reported contained significant hemispheric burden sharing.
However, there are some exercises that are hemispheric in nature, the best known of
which is perhaps UNITAS, a naval exercise sponsored by the United States that
circumnavigates South America in a period of five months. As the exercise fleet passes
participating countries, their vessels join the operation while off the coast of their national
territory. Operations include gunnery, maneuvers and search and rescue. In 1996,
reported participation included Argentina, Brazil, Chile, Colombia, Ecuador, Paraguay,
Peru, the United States and Venezuela.
UNITAS originated in 1959 under the auspices of the U.S. Navy and is now the
longest running exercise in that service. In its early form, the exercise was conducted
bilaterally, between the United States and each participating country. But in recent years
the exercise has expanded to become a joint/combined warfare exercise with participation
of the Marines, Coast Guards and Air Forces, as well as Navies of the hemisphere. There
is considerable overlap in participation by countries now and normally at least three
countries are involved at any given phase of the exercise. Here again, an old tool has
been revamped to build multinational coordination.
Most countries of the hemisphere with fully constituted defense forces (Cuba
excepted) participate in the various conferences of the armed forces, which are


130
balancing, combined with a lower level of interest from the United States, helped
contribute to the decay of the hemispheric security institutions. This dynamic is clearly
encapsulated by the struggle over the position of the IADB within the inter-American
system.
From the beginning the Board has never been fully embraced by the main political
organs of the IASS, principally the OAS. After the end of World War II, the Board lost
its focus and entered into a period of questioning. The Board doggedly sought permanent
status. In 1946, the Board submitted a proposal for a permanent military organ under the
OAS Charter to the conferences that led to the adoption of the Charter. In 1959 the Board
approved a resolution which would have replaced the Advisory Defense Committee of
the OAS with the IADB. In 1966, the Board submitted a resolution to member
governments that would have included the Board in the Charter revised in Buenos Aires
in 1967. None of these initiatives succeeded.43
At every historical juncture where the structure of the inter-American security
system has been debated, the United States has supported proposals for the establishment
of a permanent military organ within the OAS.44 Those proposals have never been
accepted. Several countries, especially Mexico, contended that a permanent military
43 For a different perspective, see the OAS (1978) study, which argues that the
Board is an entity created by competent organs of the OAS and is an integral part of the
OAS. According to this argument, the Board's prerogative to communicate directly with
the governments of the American states is simply an "operational modality" (OAS, 1978,
pp. 34-35).
44 The United States promoted the establishment of a permanent security organ in
1948 (creation of OAS Charter) and 1967 (revision of Charter and creation of the modem
OAS), as well as in 1994 (the most recent proposal).


75
autonomy. Through issue linkage and the legal principle of formal equality of members,
institutions work to reduce asymmetry and promote a more representational agenda that
may confound large power interests.
Sharper distinctions are possible with regard to the rules that these institutions
promote. Grieco (1988a) offers a set a contending hypotheses for international
agreements (rules), built around the problem of relative gains. These hypotheses can be
fairly easily adopted to our study of patterns of security cooperation in the Americas.
In the first instance, Grieco notes that under realism, states will in general eschew
cooperative arrangements with binding commitments that present high exit costs. This
should be particularly true in the area of security, where relative gains problems are most
acute. Thus realism would expect that hemispheric rules regarding transparency,
verification, defense postures and burden sharing would be voluntary and non-intrusive.
In the absence of reliable verification, the effectiveness of reciprocity will be limited.
This contrasts with agreements made between the United States and its security
partners, such as the network of mostly bilateral agreements for tracking and interdicting
illicit drug production and trafficking that exists between the United States and source
and transit countries. As rules become more inclusive, they will lose their specificity.
Under institutionalism, theory tells us that states, given the opportunity to
negotiate, will tend to insist on binding rules, with strong verification, whenever relative
gains may present an obstacle to cooperation, so that the potential absolute gains can be
locked-in. These steps are not coercive; member states are simply binding themselves
to reassure partners of their cooperative intent.


173
Treaty for the Prohibition of Nuclear Weapons in Latin America and the Caribbean
counted down the number of states still outside of the treaty.3^ As mentioned in Chapter
1, until 1994 the Treaty did not include the four countries in Latin America most likely to
proliferate in the region: Argentina, Brazil, Chile and Cuba. Membership has now
grown to include all of Latin American and the Caribbean, including Cuba, making this
treaty the broadest security treaty in the region.
Success in the area of conventional arms control has been far less successful,
despite the hemispheric consensus on the principle of limiting arms holdings. The vast
majority of resolutions relating to hemispheric security include reference to the desire to
reduce arms holdings and expenditures, based on the belief that reductions would
enhance the security of states thus reducing the risk of regional conflicts and hence
allow more funding of economic development. Transparency is almost invariably linked
conventional arms control and there is some evidence to suggest that is does create a
climate which makes reductions possible.3 1
Numerous resolutions address conventional arms control directly, beginning in
1992 with Resolution 1179, which contained several important pledges:
1. avoid excessive or destabilizing accumulations and transfers of conventional
arms
2. maintain only sufficient military capabilities for self-defense and fulfillment of
international commitments
30 OAS Resolutions 1239 (1995), 1355 (1996), 1414 (1998), and 1571 (1999).
31 Burden sharing, although it should lead to reduced defense expenditures by
using regional security resources more efficiently, has not yet had this effect. This is
because countries that use militaries to address transnational threats in cooperation with
neighbors and the United States have frequently had to first modernize their forces.


132
remediation and demining and peacekeeping operations. 47 But it became locked in a
"military ghetto," unable to infuse its recommendations and studies with political
relevance. It was reduced to providing a forum to military officers and its tradition of
closed proceedings and concern with protection of its documents limited its contribution
beyond the immediate circle of its delegates.
Latin American countries have accepted the benefits of cooperative security while
at the same time resisting attempts to centralize security relations and supporting the
principle of non-interference in domestic matters. Both Mexico and Argentina have
played the balancer and the spoiler to U.S. initiatives. In the 1970s, Brazil began to play
that role as well, although by that time the system had already begun a process of decay.
Brazil's role in the 1990s would be far more important in terms of the eventual make-up
of hemispheric security relations. But thwarting hemispherism is not the same as
preventing the United States from realizing its own objectives, which it often
accomplished through unilateral action or bilateral measures.
Reflecting on this period, both Dominguez (1983) and Mares (1997) note that
national interests, not institutions, have been the key determinant in security behavior in
the Western Hemisphere. They conclude that despite sporadic attempts to establish
47 The IADB (1996) lists the principal documents produced by the Staff: Basic
Strategic Plans; IADB Planning Manual; Civil Defense Against Nuclear, Biological, and
Chemical Warfare; Planning for Collective Air Defense of the Continent; Dictionary of
Military Terms; Coordination Plan for Defense of Inter-American Maritime Traffic;
Military Civic Action; Feasibility Study of Interoperability of Arms and Communications
Systems Among the Member Countries of the IADB; Periodic Intelligence Analyses.
This work contributed to the interoperability necessary for efficient cooperative security
activities.


80
out, states that can choose the context of cooperative arrangements will normally prefer to
limit membership. Realism, according to Grieco, predicts the opposite, since
arrangements with more members provide more partners and hence more opportunities to
offset relative gains advantages. For these reasons, realism would expect that cooperative
arrangements at the subregional level will be even more difficult to accomplish than at
the hemispheric level.41 Institutionalism predicts precisely the opposite pattern.
In sum, institutionalism expects more institution building and more binding rules
at the subregional level than the regional level. Realism predicts the opposite pattern. In
any case, both paradigms are operative at this level of analysis.
What then about the effectiveness of rules regarding security cooperation?
Institutionalism predicts that security behavior will be rule-based, following the logic laid
out in the previous section. Transparency will be critical because of its important role in
communication. Reductions in defense postures and burden sharing are likewise
important because of the absolute gains they promise. Realism, on the other hand,
expects that patterns of security behavior will follow the logic of balancing; cooperation
will be limited to informal ententes formed to counter subregional threats. We can
approximate the predicted pattern through balance of threat analysis for each subregion.
41 Defensive realists who, according to Jervis (1999), adopt many of the
standard cooperation under anarchy strategies, would disagree with Griecos
formulation. The security dilemma/collection action literature highlights the difficulties
of large alliances, since detecting cheaters and enforcing obligations is difficult with large
groups. But Grieco is not focusing here on cheating, but rather on the relative gains
problem, which is more intense in smaller groups.


309
Burden sharing activity was reported in all subregions. Burden sharing appears to
occur where problems require combined efforts and where resources are available for a
multilateral approach. We found a puzzle in Central America a disconnect between the
ambitious organizations and the actual burden sharing activity. But where the United
States was willing to sponsor such activity, it occurred. In the Andean Ridge, the lack of
reported cooperation between Colombia and her neighbors other than Venezuela was
significant and supportive of realism.
To compare the subregions more systematically, we need to recall the principal
differences between the cases: the impact of the United States, the nature of security
problems and the degree of nesting. U.S. behavior is generally as predicted. The United
States acts as a balancer in Southern Cone, approaches a security manager in Central
America, while the Andean subregion offers a mixed picture. In any case, the United
States was not the principal factor behind institution building in any of the subregions. In
fact, some efforts appeared in the Southern Cone to maintain the area free of U.S.
influence.
Realism predicted that transnational security problems would be the most
intractable to cooperative approaches. Yet we saw in Central America that these problems
were the most likely to be approached cooperatively. The cooperation between Colombia
and Venezuela shows a similar pattern.
Finally, the high incidence of nesting in Central America and the Southern Cone
contrasts to the low degree in the Andean Ridge. This correlates well to the general
pattern of subregional institution building, rule making and cooperative security behavior.
Linkages have raised expectations of larger absolute gains, provided more opportunities


72
American studies. Clearly the more general literature on the relationship of democracy
and war, the role of domestic institutions in other areas of international relations
(particularly trade) and studies linking military culture and strategic doctrine are also
valuable. These issues will be discussed further in the concluding chapter.
Returning to the neoliberal institutionalism, we can now readdress the issues
posed by the cooperation dilemmas outlined in Table 2-1. We will look to see if
institutions facilitate cooperative behavior in each of the behavior clusters without regard
to the institutional politics described for asymmetrical situations. Behavior there is
driven by the twin dynamics of reassurance and institutional balancing and not
necessarily a search for the absolute gains promised by cooperation, thus institutional
design may be only partially effective in addressing cooperative dilemmas.
Since military transparency can expose free riding, cheating and relative gains
problems, it is fundamental to institutional attempts to promote security cooperation. The
challenge is to verify military information through various mechanisms, designing
reporting formats that can be easily cross-checked and building an extensive network of
contacts that produces significant information flows. Institutions can promote reductions
in defense postures by addressing distributive gains through safeguards, and trade-offs in
types of reductions. But, as we shall see, there are no hemispheric or regional regimes
that include these measures as yet. Relative gains are normally not addressed. Will that
matter?40 With regard to burden sharing, again transparency exposes cheating and
40 These kinds of rules, which have been present in the international trade regime
for many years, include safeguards (related to balance of payment crises) and exclusions
(agriculture, textiles, etc.). See Hasenclever, et al. (1997).


144
pace with growing OAS membership in the 1990s, represented little more than half of the
34 states in the inter-American system (the number excludes Cuba).9 In an era of
constrained budgets, non-member states questioned why the Board was supported from
the OAS regular fund instead of through contributions from its members (as is the case
with specialized agencies of the OAS). In counterpoint, some delegations recognized that
the Board was responsible for important functions that would go unperformed in its
absence unless other hemispheric organizations could be redesigned. 10
In this context, the Board, prompted by the United States, took important steps
towards forging a new relationship with the OAS. In the first place, the Board
membership has been expanded. At the time of the Board's creation, its membership
included every independent American republic (with the exception of Canada). That
membership matched the list of signatories to the Rio Treaty, which was later adopted as
a membership requirement. As former dependencies became sovereign states, the Board
ceased to be representative of the Americas. In 1991, the Board dropped the Rio Treaty
9 At the time of these debates (1993-1994) the Board would have had 21
members, of which 17 were active.
19 Notes from the 3 May 1994 Committee Meeting lay out national positions in
some detail. Most of the discussion dealt with alternative institutional designs for the
Board that would bring it within the Charter of the OAS. Argentina proposed that the
Board be designated as an Entity, with support from Antigua and Barbuda, Chile, Peru
and the United States (although two days later the U.S. draft proposal mentioned above
was presented). Brazil proposed that the Board be a Special Organization, with support
from Ecuador and Mexico (although Mexico preferred to maintain the status quo while
awaiting studies of the other hemispheric security institutions). These positions roughly
correspond to those reported by Board delegations in 1993. A detailed analysis of these
alternatives and others was presented by the Board (IADB, 1993; IADB, 1994).


BIBLIOGRAPHY
Acua, Carlos H. and William C. Smith. 1994. "The Politics of Arms Production and the
Arms Race among the New Democracies of Argentina, Brazil and Chile." In Lars
Schoultz, William C. Smith, and Augusto Varas, eds., Security, Democracy and
Development in U.S.-Latin American Relations. Boulder, CO: Lynne Rienner, 199-
240.
Adler, Emanuel. 1988. The Power of Ideology: The quest for technological autonomy in
Argentina and Brazil. Berkeley: University of California Press.
Almond, G. A. 1988. "The Return to the State." American Political Science Review 82
(3): 853-874.
Anthony, Ian. 1993. "Assessing the U.N. Register of Conventional Arms." Survival 35
(4): 113-129.
Arias, Oscar Snchez. 1996. "El Mundo despus de la Guerra Fra: Los Principales
Retos." Paz y Seguridad en las Amricas 9. Santiago: Facultad Latinoamericana de
Ciencias Sociales, 8-9.
Argentina. 1999. Libro Blanco de la Defensa Nacional. Buenos Aires: Ministry of
Defense.
Argentina and Chile. 1998. Joint Presentation at the Third Defense Ministerial of the
Americas. Cartegena de Indias, Colombia.
"Argentine General Proposes Unified Mercosur Defense System." O Estado de Sad
Paulo (Sao Paulo), 15 April 1998, Foreign Broadcast Information Service (on line),
21 April 1998.
"Argentine, Uruguayan Armies Conclude Operation Ceibo." La Plata America, 16 July
1998, Foreign Broadcast Information Service (on line), 16 July 1998.
"Army Chief of Staff Leaves for South Cone Drill in Brazil." Buenos Aires Telam
(Buenos Aires), 26 August 1997, Foreign Broadcast Information Service (on line),
27 August 1997.
338


151
international solutions to multidimensional security problems (this was questioned by
both Brazil and Mexico). Finally, more civilians and more representatives of non
military security agencies are attending the College.
Taken together, these trends mark a significant transformation in the Board. But
despite an aggressive campaign to bring the Board in line with the OAS, 25 its legal basis
is still challenged, its limited budget questioned and its mandates restricted and it has
become a major battlefield for the institutional politics that mark U.S.-Latin American
relations in the security field.
New institutional structures
The 1990s have seen a number of new institutions that address aspects of
hemispheric security in ways that are consistent with the trend seen in the evolution of the
older organizations. These new institutions include the presidential summit and defense
ministerial processes, both put into motion by the United States.
While security has never been the focus of the presidential summits, security
issues are nonetheless discussed. The Miami Summit plan of action specifically
addressed the need to develop cooperative approaches to fight narcotrafficking and
terrorism. There was also explicit support in the plan of action for an evolving regime of
CSBMs and for the first CSBM conference held in Santiago. For all of these items, the
1994 summit referred action to the OAS.
25 The goal of maintaining a positive relationship with the OAS and with
individual OAS delegations was formally adopted by the Board as a key institutional
objective in 1996. For several years, the Board has been on a campaign to demonstrate
its relevance. It is worth noting that an organization that has to argue for its relevance has
probably already lost it.


240
transparency ("Confidentiality accord," 1998). The strong U.S.-Argentine connection is
an important part of the pattern of transparency/verification that must be considered.
In sum, there is clearly a dense network of contacts in the Southern Cone that
produce significant information flows, including some sharing of sensitive information
between Argentina and Brazil, Chile and Uruguay. Furthermore, the pattern of contacts
validates an extremely close relationship between Argentina, Brazil and Uruguay.
Contacts between those countries have developed into a trilateral relationship in many
areas (Brazil 1995 CSBM Submission). This follows the pattern of institutionalization
reviewed earlier.
Three important points follow. The network of transparency and verification is
real, involving useful information not publicly available. Second, the pattern runs across
axes that formed the traditional ententes of the subregion, not along them. Third,
cooperation with large powers and the buffer states is not a byproduct of competition for
influence cooperation in most instances is multilateral and inclusive of the large
countries which might be expected to be competitors.
Uruguay, the small state most involved in this activity, participates strongly with
both Brazil and Argentina there is no evidence that the larger states compete for
influence. An excellent example of the trilateral nature of this relationship was supplied
by Uruguay's 1997 CSBM report: the three countries now provide prior notification of
movements in border regions, as well as on equipment modernization programs for
border units (Uruguay 1997 CSBM Submission).
However, there are some indications that may point to the impact of underlying
rivalries. Compared to Argentina and Brazil, Chile participates less in this activity. And


323
manufacturing base. The relative success of defense reductions in Central America,
which continues even in the absence of international verification missions, is surely
related to the pressing need to prioritize social and economic development in that
subregion.
Third, the strong connection observed between transparency in military matters
and the preservation of democracy bolsters the democratic peace argument that
transparency inherent in democratic governments offers the best explanation for the
relative lack of conflict between democracies (Starr, 1999). Internal transparency
reinforces external transparency, which precludes worst case security analysis and
demonization of the leadership of potential rival states.
While in Latin American studies, the belief that democracies are more pacific than
military-authoritarian regimes is widespread (Varas, 1985), supporting evidence has not
been easy to muster. The claim that democracies rarely engage in conflict with each
other has not fared well. Democracies have fought each other; Ecuador and Peru have
done so several times since 1981. Democracies have also been attacked; Costa Rica, a
country without an army, was invaded twice by Nicaragua (Mares, 1997). Political
leaders of democracies have militarized relations with their neighbors in order to gamer
popular support in Colombia and Venezuela, as well as in Ecuador and Peru.2
2 See Mansfield and Snyder (1995) and Mares (1997).


225
condition of its armed forces. Furthermore, disclosure of military expenditures by the
central government was frequently a requirement of stabilization programs supervised by
entities of the World Bank this was certainly the case with Argentina. For all of these
reasons, it is clear that Argentina would have little to risk by endorsing a transparency
regime in the Southern Cone. Argentinas endorsement was important, because it
removed the most serious potential relative gains problem in the area of transparency.
But would the other countries of the subregion reciprocate? Would they work
together through institutions to enhance transparency? Would specific rules evolve to
structure available information in meaningful ways? Would strong verification
mechanisms be employed?
With regard to defense postures, conditions were also generally positive at the
outset of the decade. The need to downsize and reprioritize central government resources
affected all the major countries, meaning that the absolute gains to defense reductions
were considerable. The end to nuclear competition by the early 1990s removed that
negative dynamic from the subregion. But the twin problems of Brazilian strength and
Argentine weakness meant that any move to freeze levels could engender serious relative
gains problems. A regime that addressed defense postures successfully would have to
tackle these relative gains problems.
Burden sharing seems to offer relatively fewer absolute gains from cooperation.
Traditional security issues arms races, border control and sea patrols presented some
possibilities, but the lack of a common external threat complicates coordination in these
areas. In addition, few transnational security issues are present in the subregion, other
than contraband flows in the tri-border area of Argentina, Brazil and Uruguay and the


252
peacekeepers to Central America (Pala, 1998). There are clear absolute gains to be
gleaned from this activity combat-like experience for the armed forces; exposure to
advanced military systems, communications and organization; training in coalition
operations; and receipt of support equipment from the international community
(principally the United States). There is, however, an inherent dilemma. The deployment
of these forces overseas makes them unavailable in the event of a subregional dispute.
This may explain the relative restraint of Chilean peacekeeping policy. Brazil, for its
part, can afford the deployment without suffering an important relative decline in force
posture. But the willingness of Argentina and Uruguay to participate so extensively in
these out-of-area operations indicates a strong sense of assurance that subregional conflict
is nearly unthinkable.
Specialization for out-of-region operations does not imply surrender of critical
functions to other countries in the region; in other words, there is not necessarily a
relative gains problem with these activities. How can we relate the impressive activity in
burden sharing in peacekeeping operations to subregional defense integration? There are,
in fact, indications that the network of bilateral and regional agreements may coalesce
into a regional security arrangement although the exact parameters of such an
institution are hotly debated.
Will this security cooperation and integration lead to a NATO-like command
structure, with truly integrated forces? Unlikely as this seems, there is movement in that
direction. An important agreement on national and international defense was signed on
April 27, 1997, by Presidents Carlos Menem and Fernando Henrique Cardoso to
harmonize defense doctrines and enhance regional security. While the agreement did not


4
Contrary to Grieco, I argue that institutions do matter, although not always in
ways that promote cooperation. When interests are largely compatible and power evenly
distributed, institutions can be effective; institutional mechanisms facilitate cooperation
in a wide variety of ways. However, in the absence of clear consensus, especially under
conditions of asymmetry, institutions may not foster cooperative behavior. That is not to
say they are unimportant. Indeed, it is in these situations that institutional processes can
have their most significant impact over power and interests.
As we shall see, the institutional politics of hemispheric security institutions can
work against the interests of the most powerful states, particularly the United States.2
While military balancing against the United States may not be an option, smaller states
can develop a diplomatic line of resistance that rallies opposition and prevents U.S. co
optation of institutional tools. This "institutional balancing" is most likely when the
security agenda is complex and issue linkages are available conditions characterize the
1990s as a result of the processes of globalization. A divided security agenda emerges
that stunts the growth of security institutions at the hemispheric level. On the other hand,
in subregional areas, where mutual interests are more easily identified, a rough balance of
2 The power disparity between the United States and the rest of Latin America is
so great that comparisons are not always meaningful. At the beginning of the decade, the
United States had more men and women in uniform and spent about twenty times more
on defense than all of Latin America. Defense spending accounted for a larger
percentage of the U.S. economy (which was roughly three times that of Latin America)
than any other country in the hemisphere (with the possible exception of Cuba). Similar
comparisons of equipment, organization, technology, and training levels deepen the
divide. See Poitras (1990).


206
forward in many regions, principally the Southern Cone. Some analysts have suggested,
however, that the downward spike in the hemispheric trend line at the beginning of the
decade was principally due to the loss of Soviet sponsorship in Cuba. Thus, there may be
less variation in actual expenditures than indicated by hemispheric data. One point
remains unchanged in all of this however Latin American is still the least militarized
major region in the world.
What of the move towards civilian control empowerment contained within the
newly emerging OAS Education for Peace program? This factor is probably more
marked in Central American than in other regions. Representatives from Central
American militaries have participated in two subregional forums on the topic and the
conference themes are reflected in some restructuring the adoption of a volunteer force
and a civilian defense minister in Honduras, for example. But representatives from all
countries of the region have participated in similar training human rights, appropriate
roles for militaries under democratic control, etc. It is difficult, if not impossible, to link
those programs to arms reductions.
In sum, there is not much that can be made of low response to weak rules. But
comparison to burden sharing is interesting, because while that sector also has weak
rules, there appears to be more activity. It is possible that the presence of the United
States in burden sharing, versus its low profile in the promotion of arms control, may
explain the difference. Let us turn to that behavior cluster now.
Burden sharing
Seven percent of the measures reported were related to burden sharing. Most of
those are standard cooperation measures such as border contacts and coordinated efforts


34
size, weapon types, and operational practices, such as those contained in the
Strategic Arms Reduction Talks (START), Conventional Armed Forces in Europe
(CFE), and Intermediate-Range Nuclear Forces (INF) agreements; and the
nuclear, biological, and chemical weapons and ballistic missile nonproliferation
regimes ... cooperative verification and transparency measures, such as the data
exchanges and on-site inspections required by various arms control agreements,
the UN arms sale registry, and the 1990 Open Skies agreement, (p. 7)
All of these measures have been described as CSBMs.^
In this hemisphere, in the 1990s, there are rules already in place associated with
the composition of forces, such as prohibitions on weapons of mass destruction, arms
control/restraint, ceilings on troop strength and constraints on military exercises and
deployments, and restrictions on certain other military activities. There are also rules
relating to transparency, such as exchange of information about military activities (e.g.,
exercise notification), capabilities (e.g., participation in arms transfer registries), and
doctrine (e.g., publication of so-called white papers). There are mechanisms to promote
coordination in security matters, including meetings between high-level representatives of
the defense ministries, exchanges of personnel and military units, and participation in
conferences, symposiums, etc. Some measures are designed to develop cooperation in
certain joint endeavors. They include measures that promote interoperability, such as
3 It is easy to see that such an expansive catalogue of CSBMs may cause some to
decry the vagueness of the concept of CSBMs one author calls them the fastest
growing business of the post-Cold War era and points out the lack of accepted theory to
explain their growth or impact. Without a doubt, the term CSBM has become a
watchword for all sorts of new security relations throughout the world and includes
procedures that go well beyond the modest measures agreed to at the 1975 Helsinki
Conference, where the concept was first coined. These observations come from
Desjardins (1996), a critic of generalized acceptance of the value of CSBMs. Desjardins'
work is an invaluable corrective for tendencies prevalent in the literature to expand the
definition of CSBMs and exaggerate their value.


334
and provide the most effective mechanisms for peaceful resolution of conflicts, both
internal and external. Second, multilateralism the United States leverages multilateral
institutions to develop consensual approaches to common challenges. Third, cooperative
security the United States seeks to engage its partners in cooperative endeavors to
address shared security concerns before conflict escalates.
The second and third principles create a bias for some sort of multilateral forum to
promote security cooperation. The first requires that such a forum be consistent with
democratic norms in the sense that it is responsive and responsible to a hemispheric
political body. These broad parameters are consistent with many futures for the Board,
including outright closure.
Given what we have discovered about the institutional politics of hemispheric
security, insistence on maintaining the Board, either by incorporating it under the OAS
charter or transforming it into a military staff for the OAS Secretary-General, is likely to
be counterproductive.
An approach not yet tried is to draft a design that starts with the Inter-American
Defense College, not with the Board. This is viable for several reasons. First, the
College enjoys s broad legitimacy that goes beyond the Board. It has a dynamic
curriculum that is in demand. Canada sends students, for example, and more OAS
ambassadors and other dignitaries visit the College than the Board. Second, the College
has few of the handicaps suffered by the Board (uncertain mission, mandate and future).
Third, an expanded College can replace many of the functions performed by the Board.
With a properly resourced research department, the College could measurably
improve the quality of security studies produced by the Board. The College already has a


98
Realism has a relatively rich tradition that includes geopolitical and balance of
power analyses, 1 in addition to neorealist approaches that characterize much Latin
American conflict in the Cold War as a bi-product of U.S.-Soviet competition. Van
Klaverens (1996) description of power politics in Latin America encapsulates the basic
tendencies of the approach. Latin America is viewed in terms of regional competition in
which countries such as Brazil and Argentina attempt to maintain a certain equilibrium in
the Southern Atlantic area while Mexico and Venezuela compete for influence in Central
America and the Caribbean and Chile seeks influence in the Southern Pacific area (Van
Klaveren, 1996, p. 41). The smaller states Bolivia, Paraguay, Uruguay, Ecuador and
the Central American countries serve as buffers between their larger neighbors and
occasional recipients of benefits from competing influence attempts.2
The classic study of the balance of power mechanism in South America was
provided by Robert Burr (1955), in The Balance of Power in Nineteenth-Century South
1 Balance of power analysis is represented here primarily by Burr (1955), who
sees states as striving to maintain a balance of power within a group to prevent any
nation from becoming sufficiently strong to enforce its will upon the others or threaten
their independence" (Burr, 1955, p. 37). Geopolitical analysis is well represented by
Kelly and Child (1988) who define geopolitics as the relationship between power politics
and geographic features such as "locations between countries, distances between areas,
and terrain, climate and resources within states (p. 2). See also Kelly (1997).
Geopolitics shares some of the attributes of classical realism, but accepts more readily the
prospects of economic integration driven by resource exploitation and facilitated by
transportation networks (Kelly, 1997; Caviedes, 1988).
2 Buffer states have not fared well historically. "Shatterbelt," the term employed
by Kelly (1997), is indeed descriptive of their experience. They have lost territory, been
the site of wars and had their foreign policies manipulated (Kelly, 1997). Dominguez
(1998) does note however that some larger South American countries made territorial
concessions to smaller ones in exchange for economic benefits.


279
disagreement stemmed from attempts by Salvadoran nationals, living in territory ceded to
Honduras by international arbitration, to transport lumber taken from trees in the area to
El Salvador. Honduras claims that the areas in dispute have suffered irreversible
ecological damage and blames El Salvador for allowing a situation of insecurity to persist
on the border, citing involvement of the Salvadoran armed forces in the illegal border
activities (Bennett, 1996). El Salvador did not respond with its own border deployment,
but registered strong objectives to the Honduran troop presence ("Army Decides," 1997).
Conflict in other border areas has sparked less dramatic but nevertheless vocal
protests from both sides. As an example, Honduran suspicions of Salvadoran
reconnaissance activities to site a future hydroelectric dam in the border region of El
Tigre sparked a heated exchange in the media ("Minister Criticizes," 1997). Likewise,
reports of armed bands based in Honduras that allegedly threaten Salvadoran citizens
have raised concerns ("Salvadoran Defense Minister," 1998). Fishing disputes in the
Gulf of Fonseca also continued to cause friction between Honduras, El Salvador and
Nicaragua. Honduras also reports similar conflicts with Nicaragua in the Caribbean
(Bennett, 1996).
Honduran attempts to exercise its sovereignty in the Gulf have also generated
Salvadoran protests. A planned naval exercise between Honduras and the Republic of
Korea in the Gulf, ultimately cancelled, caused the Salvadoran defense ministry to
complain of destabilizing activity in that sensitive area. It is worth bearing in mind that
El Salvador has developed strong ties of its own to East Asian countries both with
Japan, through construction projects in the eastern region of the country, and Taiwan,
through negotiation of investment and access rights to the port of La Union on the Gulf of


178
responsibilities. Finally, the norm of democracy is shared by all member states of the
OAS.32
For the United States, support to democracy has been the cornerstone of its Latin
American policy under the Clinton administration. The Haiti intervention, while
dramatic, is not emblematic of that policy, even if it demonstrates one possible outcome.
The U.S. had redesigned a large number of its security instruments in an attempt to
inculcate respect for human rights in the region's militaries, most visibly in its bilateral
training programs (Cope, 1995), end-use monitoring provisions in Colombia and the
thematic focus of exercises sponsored by the U.S. Southern Command (with a change
from force-on-force war games to peacekeeping training). The United States also
promotes "new thinking" through multilateral institutions, particularly the Inter-American
Defense College. Buchanan (1998) offers an excellent review of these trends. Despite
some grumbling by the militaries of the region, this policy is broadly accepted by the
democratic governments of Latin America.
All in all, the democracy regime represents an excellent case of institutional
bandwagoning. Since the focus of this study is not on democracy, but rather on security
cooperation, it may appear to lack proper emphasis on this important aspect of U.S.
policy. If this study highlights U.S. efforts to forge cooperative action in other areas -
building capacity for drug trafficking and peacekeeping intervention for example that is
because these represent important interests that directly relate to the use of the military in
32 Cuba, of course, is the exception to this trend in the hemisphere.


304
condemn the human rights record of the military and its purported links vigilante groups.
There is a perception in the Colombian military that they are victims of an information
war that often targets their most successful battlefield commanders.
It is clear that Colombia, at the end of the 1990s, has chosen to maintain as open a
society and to respect human rights during the conflict. The alternative presented by the
Peruvian experience was not the road taken by Colombia. But the Colombian path, while
conforming to the hemispheric democracy regime, cannot succeed without massive
subregional and regional assistance. This will require a high level of burden sharing from
neighboring countries.
It may be, in fact, that the lack of a trend towards reduced defense postures may
be in anticipation of a subregional effort to deal with the transnational problems spilling
over the borders of Colombia. Venezuela, for one, has linked its defense posture in part
to the conflict in Colombia. But this would imply that burden-sharing activities are being
developed to deal with the Colombia problem. Let us now turn to that behavior cluster to
see what the evidence shows.
Burden Sharing
Along Colombias 6,000 kilometer border, there are 71 guerrilla fronts and 45
paramilitary militias (Regional Security, 1999). Many are involved in drug trafficking
and other forms of criminality that directly degrades security in neighboring states.
Security cooperation between states is urgently required. As an example, consider the
binational security agenda between Colombia and Venezuela, which includes:
fighting off the insurgents, denying them control of the border areas,
checking continuing actions of drug mafias associated with them, dealing
with the large number of undocumented immigrants, and mitigating


256
cooperation to address standard patrolling responsibilities (naval and air operations in the
Beagle Channel, refueling activity, aircraft carrier operations) as well as new
transnational threats (intelligence sharing and coordination between radar systems to
locate clandestine flights, cooperation in the tri-lateral border region).
The cooperation in the area of out-of-region peacekeeping operations is clearly
unprecedented. The training exercises for these operations are far and away the most
extensive multilateral operations in the hemisphere. The integrated nature of the
Argentine peacekeeping deployments to Croatia is equally impressive in this regard.
Likewise, the deeply institutionalized burden sharing in MOMEP between Argentina,
Brazil and Chile indicates a willingness to share responsibilities in areas contingent to
although outside of the Southern Cone. While the involvement of the United States is
important to some of these activities (especially MOMEP and some peacekeeping
training exercises), the majority of burden sharing activities does not feature U.S.
participation or sponsorship.
What do the geographic patterns reveal? As with contacts covered under
transparency/verification, burden sharing concentrates on Argentina, Brazil and Uruguay,
with the United States as an important partner to all three countries. There is little
evidence of balancing behavior along the old ententes of Argentina and Peru and Chile
and Peru.
Thus, the only pattern which falls most easily to a realist explanation is the strong
relationship between Argentina and the United States. That cooperation, fruit of the
dramatic change in Argentine foreign policy which led to collaboration with the United
States in the Gulf War and peacekeeping operations, has elicited criticism from both


55
choose to provide positive incentives for cooperation even if coercion can be cost
effective in the short run. It is possible that a great power such as the United States might
choose to underwrite significant security cooperation at this particular juncture in history.
Mastanduno argues that the United States, which enjoys privileges of the "unipolar
moment," faces strong incentives to reassure the other states in the system of its peaceful
intent.
In making this argument, Mastanduno relies on the hypotheses developed by
Stephen Walt to explain alliance behavior. In point of fact, Waltz had little to say about
security behavior in a unipolar world, which characterizes to a great extent conditions in
the Western Hemisphere during the 1990s. Waltz believes that any unipolar moment
will be short lived because of the propensity of other states to revolt against overweening
power at the center. In any case, the kind of security cooperation that Waltz envisioned
was simply a byproduct of superpower competition. Absent that competitive dynamic,
cooperation with prospective allies should diminish.
Mastanduno borrows from Walt to argue that the United States would rationally
take steps to appear less threatening to other states, in order to preclude their forming a
balancing alliance that would act as a counterweight to U.S. power. Thus the United
States could be expected to work through multilateral institutions, promote transparency
to signal peaceful intent and offer security outlets to potential rivals (burden sharing
opportunities such as leading roles in peacekeeping operations). While this hypothesis
may not square with current U.S. policy in Central Europe and Russia, it offers a
theoretically sound prediction of U.S. behavior to enhance security cooperation that is
consistent with realism.


152
The 1998 Santiago Summit addressed CSBMs more directly, singling out
transparency measures, particularly those that might lead to reduction and control of
military expenditures. Importantly, the action plan also gave support to several items that
fall within the burden-sharing category: peacekeeping operations, demining, search and
rescue and cooperative programs to deal with natural disasters. It also deepened the
degree of hemispheric cooperation to combat terrorism and narcotrafficking with a long
list of specific requirements.
The meetings of the Defense Ministers of the Americas (DMA) were initiated in
1995 by the United States to provide an opportunity to discuss defense issues at the
ministerial level. To date, there have been three DMAs, each reviewed briefly below.26
Mexico, virtually alone in the hemisphere this time, has not elected to join the DMA
process. A representative from the Mexican foreign ministry attends in observer status.
In general, there are four themes that consistently appear in the DMAs: military
transparency, civil-military affairs, burden sharing against transnational threats and
strengthening hemispheric security institutions. Every DMA has touched on these
themes. The lead theme and the precise topics vary, although the focus of any of the
DMAs is sometimes only obvious with hindsight. Each successive DMA, however, has
been less successful in developing consensus on hemispheric security.
The 1995 Declaration of Williamsburg presented a list of principles addressing
the role of the armed forces in democratic polities. The Williamsburg principles
establish the preservation of democracy as the basis for ensuring mutual security, support
26 The fourth DMA is slated for Brasilia in the year 2000.


18
institutions, and the inclusion of new items on the inter-American agenda. The U.S
decision to work through hemispheric institutions and willingness on the part of Latin
American states to engage in "institutionalized bandwagoning" has been key to the
success of cooperative initiatives (Corrales and Feinberg, 1999, p. 8).
A renewal process at the OAS 16 was initiated in 1991 through the adoption of
OAS Resolution 1080, The Santiago Commitment to Democracy and the renewal of the
Inter-American System. ^7 The resolution established a firm resolve to stimulate the
process of renewal of the Organization of American States, to make it more effective and
useful in the application of its guiding principles and for the attainment of its objectives.
While the renewal process is in large part an administrative reform to increase
organizational efficiency, responsiveness and transparency of resource allocation, the
movement includes rethinking and redefining several of the core functions of the OAS.
For instance, economic development has been broadened and redesignated under the
rubric of sustainable development and support for democracy has been strengthened by
the creation of institutions to monitor, verify and improve governance in newly
consolidating democracies.
Corrales and Feinberg claim that a new historical peak of hemispherism was
reached in 1994, at the Miami presidential summit. At that meeting, the presidents of the
16 See OAS (1995a) for a review of that process.
17 OAS Resolutions are listed by resolution number and title when that additional
information adds to the argument. OAS conferences are referred to by title or subject
matter and date. OAS publications are available through the Office of the Secretary
General, Organization of American States, Washington, DC.


82
other regions, states may be able to choose whether or not to cooperate with the U.S.
agenda.42
Regions more distant from U.S. power and security concerns and more developed
in terms of national power should retain some measure of autonomy, since relative
capabilities are less asymmetrical and intervention is less likely. In fact, such subregions
may enter into agreements that allow them to resist U.S. influence. Examples addressed
later include Brazilian resistance to U.S. attempts to influence peacekeeping training and
the reluctance of the Andean countries to enter into a multinational, operational
agreement with the United States to combat production and trafficking in illicit narcotics.
But even less autonomous regions can demonstrate some independence. For instance,
Nicaraguas participation in almost all aspects of Central American security cooperation
demonstrates independence from the United States, which until very recently followed a
policy of limited military-to-military contact with Nicaragua.
Despite this last example, it seems evident then that the subregional security order
in Central America will more closely approximate a situation of hegemonic control due to
the region's proximity to the United States and its relatively lower level of development.
Countries of the Southern Cone, on the other hand, operate with relative independence.
The Andean countries fall somewhere between these two extremes. In any case, if a
42 They will have two calculations to make: (1) does cooperation provide
absolute gains (given unreliability of the United States as a partner and potential
incompatibility of interests) and (2) will neighbors will enjoy relative gains through their
own cooperation with the United States.


193
campaign is best explained as an act of institutional balancing, not thoughtful promotion
of peaceful interstate relations. Otherwise, rules in this area have been remarkable for
their lack of specificity. There is explicit recognition that defensive postures must be
addressed at the subregional level, where the dynamics of security competition take place.
On the other hand, there is strong overlap with the democracy regime in terms
of appropriate roles for militaries under democratic regimes. The net effect of this
phenomenon is to bolster transparency while undercutting efforts to involve the armed
forces of the region in some burden sharing activities particularly to address
transnational threats which may involve illegal groups operating within countries. Strong
proscriptions on military involvement in policing and collecting intelligence on these
groups limits the armed forces to supporting roles to other domestic organizations.
Table 4-4 presents a summary of the key rules of the hemispheric security
institution, adopting the standard categories developed by regime theorists. This is
presented as a convenient way to compare the hemispheric security regime in the
Americas with other regimes in other regions and in other issue areas.


CSBM Conferences
CSBM Meeting of Experts
Santiago Conferences
San Salvador Conference
Commissions
Hemispheric
Security
Commission
Security Conferences
Declaration on Terrorism
Small Island Security
Security Institutions
Related Organizations
Unit for the Promotion of Democracy (UPD)
Inter-American Commission for Drug Abuse Control (CICAD)
College
Staff
Demining
Education
for
Peace
CSBMs
Natural Disasters
N>
SO
Figure 1-1: Inter-American Security System


231
the presence of mines along the border, emplaced by the Chilean military. Chile's
willingness to enter into a declaration that indirectly addresses the mine problem is
significant. Demining is a slow, arduous, expensive task, although in this case the mines
may be able to be removed by mechanical methods, due to good record keeping on the
part of the Chilean military and the open terrain. Although no concrete progress has been
made in this area, the apparent willingness of Chile and Bolivia to submit long-standing
border disputes to third-party arbitration, and to abide by those independent decisions,
demonstrates an important commitment to peaceful relations.
Worthy of special note is the gradual institutionalization of contacts between
Argentina and Chile, because of the long history of security competition between those
two countries. Military relations between Argentina and Chile have grown progressively
closer and now extend to limited burden sharing for some border activities, which will be
discussed shortly.
At the third DMA, Chile presented a joint report on Argentine-Chilean security
institutions that outlined considerable progress. The report began by claiming that the
gains to security cooperation were obvious in the bilateral relationship due to commercial
linkages, reciprocal investments, shared interests and common political values (Argentina
and Chile, 1998). The process to increase security cooperation began in 1993. A
Memorandum of Agreement for Security Cooperation between Argentina and Chile
entered into force that year to guide the following activities: exchanges between
academic institutions, staff exercises for mutual support in disasters, technological
developments, security consultations (Argentina 1996 CSBM Submission). Chile and
Argentina began a series of meetings at the commander-in-chief level of the armed forces


212
Geographical Patterns
Comparing the data according to individual countries is problematic, since
countries adopted their own threshold for reporting. Close inspection of the data reveals
that many measures reported by countries are not reflected in their partners submissions.
The incidence of overlap is substantially less than 100%, which would reflect perfect
standardization. Since reporting is voluntary, the number of submissions by country may
tell us more about a given countrys enthusiasm for the CSBM inventory project than
provide an indication of actual cooperative security behavior.
These caveats in mind, let us move ahead to see what emerges in terms of
geographic patterns. The table below includes measures from 1997 and some measures
from 1996, for member countries of the IADB only.
Table 4-5: CSBM Distribution by Country -- 1997
Declaring
Country
CSBM Reported by Country
Bilateral
Multilateral
Total
Antigua & Barbuda
0
10
10
Argentina
216
19
235
Bolivia
13
7
20
Brazil
88
69
157
Canada
23
21
44
Chile
57
14
72
Colombia
26
3
29
Ecuador
17
0
17
El Salvador
6
32
38
Guatemala
2
27
29
Honduras
1
28
29
Mexico
52
16
68
Peru
53
0
53
Trinidad &Tobago
0
7
7
United States
16
37
53
Uruguay
88
25
113
Venezuela
6
2
8
TOTALS
664
318
982


188
hemispheric rule simply urges that countries participate in small island security
conferences (a political bargaining chip played by Caribbean states), the conferences
themselves focus on a very wide range of security issues (including economic
development) and have developed subregional agreements to cooperate against
transnational threats in the Caribbean drug trafficking, organized crime, illegal
immigration and of course natural disasters.
Likewise, the measure urging improved border communications has wide
implications when stepped down a level to the regional context. Some of the border
contacts legitimized by this measure should be considered transparency and verification
measures. But many more are really border control measures to address transnational
phenomena, especially smuggling of contraband, which includes counterdrug operations.
This evidence points up the importance of the subregional level of analysis. In general it
is fair to say that the hemispheric security regime is much narrower for burden sharing
than for transparency and verification.
Analysis
So what is new in the hemispheric security institution? In the first place, the frank
institutionalist spirit that undergirds it. An evolutionary process has created a regime of
strong, overlapping institutions with expanding membership and resources. Some of
those institutions are new, others have been substantially remade, shedding their old skin.
In the second place, those institutions have produced mutually supportive rules.
The IADB and the OAS Committee on Hemispheric Security have worked hand-in-glove
to institutionalize CSBMs and conduct demining operations. The Summits and OAS
General Assemblies have held their own dialogue on hemispheric security. Both the


161
accomplished in the CSBM Meeting of Experts the previous year. The Santiago
Declaration marks the first effort to fully codify CSBMs and categorize them by type.
The Santiago measures are listed in Table 4-1. Although the CSBMs identified in
this document and developed over time include rules addressing burden sharing and
defense posture, the major impact at the hemispheric level has been to promote
transparency and search for means to verify that transparency.


63
If agreements are to be made on the basis of this information, however, its
relevance and reliability are critical. Information about the resources and formal
negotiating positions of potential partners is not sufficient. What is needed is information
about their internal evaluations of the situation, their intentions, the intensity of their
preferences (Keohane, 1989, p. 118). That information must be verifiable, through
formal inspections, visits, exchanges, data cross-checking, or other mechanisms (Nye,
1991; Oye, 1986; Anthony, 1993). Institutions can provide this service by levying
reporting requirements on members and establishing rules for verification. Institutions
also help verification through explicit codification of norms, making it easy to evaluate
the degree of compliance (Oye, 1986, pp. 16-17). Since these mechanisms taken together
greatly facilitate the operation of reciprocity, they are critical to agreements regarding
reductions in defense postures and burden sharing.
In addition to providing information, institutions facilitate the making of
agreements in a variety of ways. In the first place, they provide a forum, or constructed
focal point for negotiation (Keohane and Martin, 1995, p. 45), which increases
opportunities to negotiate (Nye, 1991). This allows bargaining to proceed in an iterative
fashion, promoting the tit-for-tat pattern of cooperation. Furthermore, institutions can
enhance communication by standardizing signals through the adoption of standard
formats and reporting calendars.
These steps make it easier for states to reach agreements regarding cooperative
security. They also create the expectation that states will comply with security rules.
This is because institutions formalize standards for behavior and create the means to
monitor compliance and evaluate state reputations (Hasenclever, et al., 1997). As long as


44
The theoretical literature attempts to address the problem of defecting as
illustrated by the Prisoners Dilemma through a variety of means. To the extent that
security cooperation is not a one-move game, states can signal their cooperative intent
by small incremental steps designed to elicit a positive response. If cooperation is not
forthcoming, a state can retaliate. Because of the shadow of the future gains from
future cooperation states are likely to cooperate if they do not discount future gains too
heavily and the last move of the game is unknown.
Axelrod (1984) allows for cooperative activity between two partners following
such a tit-for-tat strategy. If communication is enhanced through iteration, some
cooperative behavior could theoretically emerge. But in the security field this would
rarely amount to more than small incremental steps related to specific security problems,
such as protocols to reduce tension caused by unforecasted troop movements near
disputed border areas. These steps would have to be reciprocal and easily reversible (Oye,
1986).
In any case, even if the problem of lack of communication and uncertainty of
intent were somehow solved, the security dilemma contains the seeds of a second
dilemma to cooperation, the problem of relative gains. While inherent in the security
dilemma, the relative gains problem is not obvious in the Prisoners Dilemma. Even if
information and commitments are somehow made reliable, states will gain in different
measures from cooperation.
Grieco (1988a) clearly lays out the problem of relative gains from the realist
perspective, his argument rooted in the core assumptions of realism. He recognizes that
under some circumstances relative gains matter less and states may be able to cooperate


186
Some countries do employ the armed forces in the fight against drugs, most
clearly in the Andean and Caribbean subregions. The United States has extensive
bilateral programs to assist militaries involved in this area, although the threat of
decertification has sometimes undermined those programs. The annual certification
process has resonated loudly in international institutions. Many countries have
condemned the process and there is a move afoot now for CICAD to develop a
multilateral evaluation mechanism (MEM) one that would certainly closely examine
U.S. efforts to control demand for drugs within its borders. The MEM will be a key
agenda item at the next presidential summit, to be held in the year 2000 in Canada.
While the United States has agreed to participate in the development of a multilateral
process, it will doubtless retain some sort of unilateral mechanism.
In addition, the United States has promoted the establishment of a multi-national
operations center to coordinate drug interdiction activities. That center has not yet been
formed, due to objections by countries in the region. The objections raised by Panama,
where at one time the center was to be based, crystallize the debate. Panama does not
want military operations to fall under the purview of such a center, fearing that they
would lead to interventions in what the United States perceives to be narco-states.
Panama, of course, was a recent target of just such an intervention. But other countries -
Colombia, Mexico and Brazil are also reluctant to see a multinational security
organization take on an operational role. There is some doubt, however, that the United
States was ever seeking a truly operational center, concerned instead with maintained
access to a major airfield in the old Canal Zone Howard Air Force Base, the proposed
center for the new organization. At the DMA in Cartagena, the United States requested


31
area of security, tend to be conflictual. Cooperation is difficult to achieve, even if all
participating states would be better off in absolute terms.
The simple fact that a cooperative strategy exists that might reap absolute benefits
for participating states does not mean that states, acting rationally, will adopt such a
strategy. The dilemmas of cooperation the collective action (free rider) problem,
cheating and relative gains must be overcome for states to reap the absolute benefits of
cooperation. Institutionalism admits the possibility that multinational institutions
comprised of states and non-governmental organizations can foster cooperation. Realism
denies institutions any independent role.
Since the two theories differ strongly on the relative importance of institutions,
these offer solid analytical traction for examining cooperative security behavior and the
links between that behavior and norms regarding the use of force.
However, neither the realist nor institutionalist approaches in their elementary form
explain patterns of security competition and cooperation in the Western Hemisphere.
Both theories need to be modified to reflect the marked power asymmetry present in the
region. Power-based approaches must recognize that the United States, with its long time
horizon and generous resource base, will at times bear the costs and promote the agenda
of multilateral institutions, even in the face of short-term, tactical reversals. By the same
token, institutional approaches must recognize that no multilateral regime can effectively
constrain the United States from privileging its global or domestic interests in ways that
may offend institutional norms.
From a theoretical standpoint, the net affect of this great power asymmetry is to
mute the differences between the two theoretical approaches, making it more difficult to


354
Nye, Joseph S. 1993. Understanding International Conflict: An Introduction to Theory
and History. New York: HarperCollins.
Nye, Josephs. 1991. Arms Control and International Politics. Daedalus 120(1):
145-165.
O'Donnell, Guillermo. 1973. Modernization and Bureaucratic Authoritarianism:
Studies in South American Politics. Berkeley, CA: Institute of International Studies,
University of California, Berkeley.
O'Donnell, Guillermo and Philippe C. Schmitter. 1986. Transitions from Authoritarian
Rule: Tentative Conclusions about Uncertain Democracies. Baltimore, MD:
Johns Hopkins University Press.
Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of
Groups. Cambridge: Harvard University Press.
Oneal, John R., Frances H. Oneal, Zeev Maoz and Bruce Russett. 1996. The Liberal
Peace: Interdependence, Democracy and International Conflict, 1950-1985. Journal
of Peace Research 33 (1): 11-28.
Oneal, John R. and Bruce Russet. 1999. Assessing the Liberal Peace with Alternative
Specifications: Trade Still Reduces Conflict. Journal of Peace Research 36 (4):
423-442.
Oneal, John R. and Bruce Russett. 1997. The Classical Liberals Were Right:
Democracy, Interdependence, and Conflict, 1950-1985. International Studies
Quarterly 41 (2), 267-293.
Organization of American States. 1978. Status of the Inter-American Defense Board vis-
a-vis the Organization of American States. Washington, DC: Organization of
American States.
Organization of American States. 1992. Report of the Permanent Council on
Implementation of Resolution AGUCES. 1181 (XXII-O/92). Washington, DC:
Organization of American States.
Organization of American States. 1993a. Report of the Special Committee on
Hemispheric Security on Fulfillment of resolution AG/Res. 1237 (XXIlI-O/93)
Meeting of experts on CSBMs in the Region. Washington, DC: Organization of
American States.
Organization of American States. 1993b. Support for a New Concept of Hemispheric
Security. (CE/SH-12/93 rev. 1). Washington, DC: Organization of American States.


278
authorities also confirm that major issues remain, related to immigration and border
demarcation (Guzmn, 1996).
The Costa Rican position, as might be expected, is radically different. A recent
suggestion from the Costa Rican President to begin regional disarmament is instructive in
this regard. The Salvadoran Minister of Defense reacted strongly to this proposal, stating
that El Salvador cannot mortgage it is security to other countries ("Defense Minister
Rejects," 1998). This position by Costa Rica was not an isolated presidential remark.
The Arias Foundation, an NGO based in San Jose, regularly critiques the "militarized
triangle" of Guatemala, Honduras and El Salvador. 17
The pattern of troop deployments shows that lingering concerns of the possibility
of conflict are still present. Friction between Honduras and El Salvador along their
common border continues to be a problem, despite mutual acceptance of a 1992
arbitration ruling by the International Court of Justice.18 Both countries accepted the
ruling and created a joint commission to deal with ancillary matters (Guzmn, 1996).
Honduras claims that El Salvador has been slow to complete demarcation of the
border in accordance with the agreement. The willingness of Honduras to militarize the
dispute as recently as 1997 is particularly telling. The Nauhuaterique pocket, long a flash
point for these two countries, was the site of a serious disagreement between the two
countries, which culminated in the mobilization of Honduran troops on the border. The
17 See Torres (1998). Nobel Laureate Oscar Arias argues for drastic reductions,
in order to realize a peace dividend for the Central America.
18 Torres (1998, 226) provides an excellent summary of this dispute.


347
Inter-American Defense Board. 1966. Report to Accompany a Resolution
Recommending a Permanent Military Organization. Washington, DC: Inter-
American Defense Board.
Inter-American Defense Board. 1967. Report on the III Special Inter-American
Conference (Buenos Aires, 15-26 February). Washington, DC: Inter-American
Defense Board.
Inter-American Defense Board. 1979. Report of the Temporary Commission to Study the
Possibility of Changing the Legal Status of the IADB in Order to Insure Full
Privileges and Immunities. Washington, DC: Inter-American Defense Board.
Inter-American Defense Board. 1992. Estimate of the Strategic Situation. Washington,
DC: Inter-American Defense Board.
Inter-American Defense Board. 1993. Report of the Chairman of the Temporary
Commission on the IADB. OAS Relationship: OAS/IADB Juridical Relation.
Washington, DC: Inter-American Defense Board.
Inter-American Defense Board. 1994. Study of Possible Alternatives to Determine the
OAS/IADB Link. Washington, DC: Inter-American Defense Board.
Inter-American Defense Board. 1996. Informational Pamphlet. IADB. Washington,
DC: Inter-American Defense Board.
Inter-American Defense College. 1998. A Reference Book of Documents Pertaining to
Hemispheric Security. Washington, DC: Inter-American Defense College.
Inter-American Dialogue. 1997. The Inter-American Agenda and Multilateral
Governance: The Organization of American States: A Report of the Inter-American
Dialogue Study Group on Western Hemisphere Governance. Washington, DC:
Inter-American Dialogue.
International Institute for Strategic Studies. Issues for years 1984-1998. The Military
Balance. Oxford: Oxford University Press.
Jatar, Ana Julia and Sidney Weintraub, eds. 1997. Integrating the Hemisphere:
Perspectives from Latin America and the Caribbean. Washington: Inter-American
Dialogue.
Jentleson, Bruce W. and Dalia Dassa Kage. 1998. Explaining Regional Security
Cooperation and its Limits in the Middle East. Security Studies 8 (1): 204-238.
Jervis, Robert. 1978. Cooperation Under the Security Dilemma. World Politics 30 (2):
167-214.


299
Nevertheless, the comprehensive and binding agreement forged by the two
countries, with the help of the guarantor states, offers a dense network of specific rules
regarding not simply military aspects, but also political, cultural and economic factors.
The rules include measures to ensure the Ecuador has an outlet to the Amazon, while
respected Peruvian sovereignty as established by the 1942 peace accord.
There is far less institutionalization between Colombia and Peru and Colombia
and Ecuador on the transnational problems that addressed through COMBIFRON. Both
Peru and Ecuador have criticized the Colombian peace process, which has temporarily
ceded territory to the FARC in hopes of reviving stalled negotiations (Marcella and
Schulz, 1999). This action at least temporarily increases the strength of the guerrilla
movement in Colombia, now recognized as a threat to subregional security. However,
with the resolution of bilateral problems between Ecuador and Peru, more helpful
cooperation may emerge.
The impact of the United States on subregional institutions is mixed. The United
States has been directly involved in two aspects of regional institution building. First and
foremost has been its important contribution to the success of MOMEP. But the second
area of U.S. involvement has been far less successful. Since the mid-1990s, the United
States has promoted the creation of a multilateral center for coordinating counterdrug
activities, which have largely consisted of bilateral arrangements with the Andean and
Central American countries. As noted in Chapter 4, that proposal met with significant
resistance from the countries of the subregion.
Several important conclusions flow from this analysis. First, there is evidence of
institutionalization and binding rules regarding security behavior. Agreements between


59
In the 1980s, a particular form of institution international regimes became the
subject of most the conceptualizing and theory building associated with the
institutionalist framework. Hasenclever, et al. (1997) offer a detailed analysis of the
principal concepts employed in institutionalist theory, beginning their conceptual review
with the consensus definition of international regime proposed in 1982 in a special
issue of International Organization devoted to theoretical discussion of the concept. In
the introduction to that volume, Stephen Krasner defines regimes as set of implicit or
explicit principles, norms, rules and decision-making procedures around which actors
expectations converge in a given area of international relations. Principles are beliefs
of fact, causation, and rectitude. Essentially, they offer a coherent set of theoretical
statements about how the world works in a given area of international relations. Norms
are general standards of behavior. Rules are specific prescriptions or proscriptions for
action. Decision-making procedures are prevailing practices for making and
implementing collective choice.
Various regimes have been catalogued in terms of their constituent parts. The
interstate system itself can be considered as a regime that operates on the principle of
sovereignty, a norm of non-intervention and rules/procedures established in the UN
charter (Krasner, 1985). Cohen (1983) outlines an international monetary system that
operates on the principle of a stable international currency, the norm that states should
receive adequate but not unlimited funds for balance of payment problems and
rules/decision making procedures as outlined in the charter of the International Monetary
Fund. Keohane, in After Hegemony (1984), defines the international liberal trade regime
as one


234
The rules that emerge from these institutions are not generally binding, although
they are usually respected, as we shall see. They do much to increase transparency in
military matters through the extensive contacts the rules establish. They are generally
silent, however, in one important area defense postures. Since arms competition has
been a significant factor in the security dilemma of the region, the absence of formal
agreements here, even though encouraged by the hemispheric institutions reviewed in
Chapter 4, is an important observation.
These findings validate the institutional explanation for understanding patterns of
security cooperation and allow us to proceed with a test of our principal hypotheses
regarding security behaviors. We will discuss each of those in turn.
T ransparency/V erification
We have already seen that there is no significant difference between the major
countries in the Southern Cone in terms of their willingness to participate in the
transparency requirements established by the hemispheric regime. We have also seen,
however, that the data reported can be difficult to interpret. Furthermore, outside of the
nuclear regime, there are few strong verification mechanisms that would make declared
data on CSBMs and military capabilities more useful.
In the absence of formal verification schemes, military-to-military contacts must
substitute for more formal mechanisms. These contacts include exchanges of
information/intelligence, which may be accomplished through meetings, conferences, or
systems of communications and data exchange, as well as exchanges of personnel,
including visits to installations, reciprocal attendance at courses, and unit exchanges in
operational areas.


337
With a mixture of admiration and mistrust, Latin America generally looks to the
United States not hemispheric institutions for leadership in the area of hemispheric
security. Except for situations where a clear, consensus threat affects the key Latin
States, hemispheric institutions are not appropriate tools for action. Attempts to
strengthen, centralize and otherwise rationalize hemispheric security institutions are
likely to prove counterproductive. But it does not follow that security institutions cannot
make a contribution at the hemispheric level.
As the natural leader in the hemisphere, the United States should try to use
hemispheric institutions to better understand Latin American priorities, find areas of
common interest and gradually build towards consensus. Such an approach requires from
the United States patience, good listening, respect for dissenting positions and a sustained
focus. The historical trend shows that U.S. focus on Latin America is periodic rather than
sustained. Whether this will change in the future may depend ultimately on a factor that
is a staple of realism the distribution of power. As long as Latin America remains
marginal relative to the great powers, it will be difficult for the United States to remain
permanently engaged on hemispheric security issues. Institutions help by promoting
continuity and demanding a predictable level of attention. These are important
prerequisites to effective foreign policy. In asymmetric systems lacking a clear common
threat, it appears that weak institutions provide these better than strong ones.


19
Americas committed their countries to: "cooperate toward an impressive array of goals in
politics (advancement of democracy, good governance), in economics (free trade in the
hemisphere, financial coordination), and in security (the fight against the drug trade, the
collective defense of democracy)."
As Corrales and Feinberg indicate above, hemispheric security and conflict
resolution have been caught up in the reform movement. As a part of that process the
ministers of foreign affairs and heads of delegations, through OAS Resolution 1080,
expressed their decision to start a process of consultation on hemispheric security in
light of the new situation in the region and worldwide, from an up-to-date and
comprehensive perspective on security and disarmament.
The renewal process initiated change in the IASS in several areas: (1)
reorganization of the institutions that address security; (2) redefinition of the principles
that underlie hemispheric security; and (3) an ambitious plan of action that includes
concrete measures for adoption by member states that constrain use of force and promote
a vigorous defense of democracy.
In 1993, the President of the OAS Committee on Hemispheric Security, Dr.
Hernn Patio Mayer, presented a new model of hemispheric security cooperative
security to replace the flawed collective security arrangement established by the Rio
Treaty. Under "cooperative security," countries would join together to address potential
threats instead of waiting until open conflicts developed. Patio recognized that such a
model was based on the premise of shared interests, shared threats and a willingness to
cooperate to address these threats, using all elements of national power, to include the
armed forces. In his view, the hemisphere was ready (OAS, 1995b).


244
National Defense Plan breaks new ground, stating that the Argentine armed forces are
intended not only for "conventional operations for the defense of vital national interests,
but also for operations within the framework of the United Nations ... and operations in
support of the national community or that of friendly countries." The Argentine white
paper flatly states: the old view of a neighbor as an adversary and potential threat is
being replaced by a new equation: your risks are now also ours (Argentina, 1999, p. 2-
19). The mutual risk doctrine has become a cornerstone of Argentine security thinking
(Villagra, 1998, p. 21).
In conjunction with the important doctrinal change that now dismisses the
possibility of subregional conflict, the defense policy advocates new roles for the
military. Of these new roles which range from environmental action to various forms
of civic action peacekeeping missions is by far the most significant. Uruguay has
likewise embraced this new policy, again primarily in the area of peacekeeping exercises
and operations (Mermot, 1997). The policy shift to new missions is important to the
extent that it implies the diversion of resources away from conventional defense.
Argentina's high profile in this new area, which is addressed in some detail later in this
chapter, was the public justification given by the Clinton administration for conferring
major Non-NATO ally (MNNA) status on Argentina in 1997.
With regard to arms control, little has been accomplished in the area of
conventional arms, in contrast to the substantive progress in the control of nuclear,
chemical and biological weapons. Given the lack of substantive agreements on
conventional arms control and defense postures, it is not surprising to find little activity
in this area.


316
Unlike institutions at the hemispheric level, subregional institutions clearly
facilitated cooperation necessary to achieve absolute gains, whether in transparency,
defense postures or burden sharing. Subregional institutions added specificity to rules
established at the hemispheric level, making them more binding and more
comprehensive. Verification through an extensive network of contacts assured states that
neighbors were respecting those rules, although no formal mechanisms for reciprocity
were developed. Relative gains dilemmas did not, for the most part, prevent cooperation.
Where linkages were present, they generally reinforced cooperative behavior.
This was especially true in the Southern Cone. While no formal security
organization emerged there, numerous, overlapping organizations and structures some
with formal ties to MERCOSUR, the economic pact combined to guide security
behavior in ways that promoted cooperation. Within the Southern Cone, the
MERCOSUR countries particularly Argentina, Brazil and Uruguay achieved
unprecedented military transparency and burden sharing capability. Transparency
measures went well beyond those established at the hemispheric level, to include prior
notice of maneuvers and deployments in border regions, detailed sharing of budget data,
and visits to sensitive installations. In burden sharing, the scope of peacekeeping
exercises and actual deployments was astounding. Some balancing behavior was evident,
such as the strong U.S.-Argentine relationship, the relative reluctance of Chile to
participate in burden sharing, and the lack of contacts between Chile and Bolivia. But on
the whole, institutionalist expectations were confirmed by the evidence.
In Central America, more formal security organizations emerged, such as the
Security Commission and the CACAF, that promoted an aggressive agenda of


203
of the increase in measures reported since 1995 has been in this area. Furthermore, almost
two-thirds of the measures are bilateral most often between neighboring countries,
indicating the growth of transparency within the Americas at the subregional level.
There are some transparency measures reported that can be considered
hemispheric in nature. Participation in the IADB allows civilians and military officials to
conduct visits to installations arranged by the Board. This includes frequent visits to U.S.
installations. In addition, both the Board and College have a five-year plan of visits that
takes officials to each member country over the course of the entire cycle. Such visits
normally involved tours of military installations.
In recent years, the Boards program of visits has been disrupted. A programmed
visit in 1996 to Central America was cancelled because Honduras could not
accommodate the visit. A visit to some of the Southern Cone countries was cancelled in
1997 because of U.S. and Brazilian concerns that the trip might be wrongly interpreted in
Paraguay, during that countrys most recent political crisis concerning retired General
Lino Oviedo.
Exercises are occasionally announced at the Board in advance, but normally only
if the Chairman or staff officers are invited to participate. In those cases, a trip report is
presented to the full Council of Delegates. Many of these exercises are U.S. sponsored
events. Distinguished visitors to the Board and College, usually senior diplomatic or
military representatives, make presentations to the delegates and the students. The
College provides direct exposure to defense doctrines and policies of member countries
through its speakers program. Students present detailed analyses of the elements of
power of their countries and prepare monographs on specific security issues. Finally,


104
Even in times of relative peace, acquisition of advanced armaments continued
apace. According to one authoritative study, the Brazilian Navy was equipped with
Dreadnought battleships by 1910, before many major industrial powers including Japan
and Germany. Argentina responded by obtaining similar vessels. Argentina, in 1946,
had more sophisticated aircraft in its inventory than the Soviet Union (English, 1985).
Warships in that epoch were offensive weapons that would not be neutralized by
defensive technology until the introduction of submarines. Like the advanced aircraft of
today, deterrence (obtaining similar capability), not defense, was the rational response.
U.S. Power and Institutional Balancing
Power equations in the hemisphere changed dramatically at the end of the 19th
century with the emergence of the United States as a great power. At various times since
that point, the United States has participated in the development of hemispheric
institutions to help manage interstate relations. Corrales and Feinberg (1999) describe
three important periods of hemispherism": (1) 1889-1906, including the creation of the
Pan American Union; (2) 1933-1954, including the war years and the creation of most of
the current organizations of the IASS; and (3) 1989 to the present, the period under
examination in this study. As Corrales and Feinberg note, these periods correspond with
rising U.S. power in the region.
During the first period of hemispherism, the United States worked to create
formal institutions to "facilitate common political, economic, and security objectives the
establishment of a hemispheric peacekeeping system, including arbitration of the
settlement of disputes and the development of trade-enhancing rules" (Corrales and
Feinberg, 1999, p. 5).


182
frequent calls for revision or abandonment. The prevalent belief is that its collective
security provisions do not work and that in any case a concept of security that rests on
preventing interstate aggression is no longer sufficient.
Consistent with the institutional decay of the Rio Treaty, the IADB has dropped
the word collective (as in collective defense) from its mission statement and
abandoned plans developed in the 1950s to organize collective military response.
Consequently, there is no current planning effort to coordinate collective security
operations at the hemispheric level.
While there are still efforts to improve interoperability in the conferences of the
armed forces, there is no agreement on the purposes that such interoperability might
serve, other than to improve transparency. In short, in the absence of an agreed military
threat to the Americas, there is no operating burden sharing agreement at the hemispheric
level to defend the region against aggression.
Efforts to arrive at a consensus to identify new threats and develop rules to guide
military burden sharing have been problematic and largely unsuccessful. There are two
important aspects to consider. First, there is little hemispheric consensus on what
constitutes a threat requiring cooperative action. Second, there is even less consensus
regarding the use of the military in those areas. We have already seen these factors affect
the battle over a definition of hemispheric security, the formation of the OAS
Hemispheric Security Committee and the designation of tasks for the IADB.
These problems certainly arise from the reluctance of some Latin American
countries to provide a mandate to an international organization, outside of the United
Nations, to foster military cooperation to address security threats. The old fears of


46
reneges on an arms control agreement. That a distant partner did not may offer little
solace.
In the case of transparency, some states may rely on secrecy more than others. A
weak state may wish to present misleading displays of strength, while a strong,
aggressive state may wish to hide its strength. A strong, status-quo state, on the other
hand, would have no reason, based on relative gains, to fear transparency. In fact,
transparency could enhance deterrence in this case. Thus, even in the absence of
cooperation, transparent behavior on the part of these states might be observed.
A different argument, based on the nature in which military information is
handled in various states, exposes a second relative gains dilemma related to
transparency. In some states with open systems and mature civil-military relations, much
information about military activities is already public. That is not the case in other states
- in fact military information may not even be available to representatives of the
government. Thus, some states do not incur additional risk by endorsing military
transparency, while others do. This dilemma is well documented in studies of
transparency measures attempted between members of the North Atlantic Treaty
Organization (NATO) and the Warsaw Pact.20 In short, states that have much to lose
from transparency may refuse to provide information or may provide misleading
information intending to dissimulate weaknesses or intent.
With regard to reductions in defense postures, the most obvious relative gains
problem is the impact on arms producers, principally the United States but also including
20 See Desjardins (1996).


305
rapidly progressive environmental degradation. (Marcella and Schulz,
1999, p. 19)
Cooperation to address this agenda is present in the Andean subregion, but only the
cooperation between Colombia and Venezuela can be compared to the activity reported in
the Southern Cone.
In total, the Venezuelan Army has 20,000 soldiers along a 2,000 plus kilometer
border, supported by air and naval bases, as well as ground radar (Regional security,
1999). Activities along the border range from joint action against guerrillas and common
criminals, to joint civic action projects such as infrastructure repair and new construction
conducted by engineer troops. Colombia and Venezuela also conduct joint naval
operations in the Gulf of Venezuela a waterway with a disputed boundary.
But elsewhere, the pattern is different. Ecuador and Colombia do cooperate along
their common border, but to a much lesser degree. Cooperation with Brazil is ad hoc and
low level, despite the substantial Brazilian military presence near the border. With
Panama, there is even less cooperation, due to the limited control the government has
over its border province of Darin.
The U.S. presence is marked in all counterdrug programs in the region, although
this has not spurred independent subregional cooperation as in Central America. The
countries of this subregion, as yet, have not developed a coordinated mechanism for this
particularly crucial form of military cooperation.
Other types of cooperative activity are present. Colombia has entered into an
agreement with Brazil and Peru to offer medical services in hospitals to military and
civilians of all three countries (Colombia 1996 CSBM Submission). The U.S.-led


117
The Board recommended troop organization and defense priorities appropriate to agreed-
upon threats to hemispheric security. These plans and recommendations were remanded
to member governments in the form of Board Resolutions and periodically updated as the
strategic situation changed.23 The United States followed the general format of these
documents when crafting bilateral plans with partner states in Latin America, imbuing
even these documents with a sense of multilateralism (Grossi, 1997, 110-11).
Increase in U.S. military activity in response to the Cuban Revolution was
substantial and far reaching (Grossi, 1997). New institutions were created: the School of
the Americas (1961), the Inter American Defense College (1962) and the system of
annual or biennial conferences of the armed forces (1959-1961).24 The U.S. Southern
Command was established in Panama, along with a special forces group famous for
training the Bolivian rangers who tracked down Che Guevera. Joint military exercises
were conducted: Operations Solidarity, Fraternity, American, and Ayacucho. UNITAS,
a multilateral naval exercise that circumnavigates South American began during this
period. Finally, the United States assisted subregional organizations, such as
23 These strategic estimates were based on a coordinated system for intelligence
reporting established by the Board in the 1950s. Although it never functioned to the
satisfaction of the staff officers required to develop plans, several of the major countries
sent in periodic reports and many members made occasional presentations on specifics
security problems.
24 These conferences, explicitly established to build cooperation against
communist aggression, do not formally rely on U.S. assistance. Host countries incur
significant expenses. For this reason, host countries have sometimes been difficult to
find. In these instances, the United States has frequently stepped in to either host a
conference or provide additional funding. For the Army conferences, the United States
has hosted eight, including the first five consecutively. In addition, the United States
funds some of the communications architecture used by the conferences.


273
budget provisions to allow for more transparency in military expenditures (El Salvador
1997 CSBM Submission). At the other end of the spectrum, Honduras is far less
transparent perhaps partly due to the pattern of international observer missions in the
region.
Measuring transparency by participation in the hemispheric regime offers only a
partial picture of activity in the subregion. The states that underwent an internationally
managed peace process have, by nature of that process, an extremely high level of
transparency. The U.N.-mandated process in Nicaragua and El Salvador required that the
armed forces reveal troop strength and deployment and weapons holdings in great detail.
The reporting was physically verified by independent observers, who also managed the
transformation of the armed forces a process that required significant downsizing and
weapons reductions. The Guatemalan armed forces have begun a similar process since
reaching a peace agreement with the UNRG. 15
Consistent with its higher incidence of compliance with hemispheric transparency
rules, El Salvador has been the most active in terms of reporting actual contacts that could
be construed as supportive of verification. Its 1997 report includes periodic meetings of
border commanders. The 1996 report lists training courses conducted with Guatemalan
and Salvadoran participants. The 1995 report is more general, but does include frequent
student exchanges in the declared data.
El Salvador's 1998 report includes several more concrete measures: a "third level
course" with Guatemalan officials in El Salvador; Salvadoran participation in a human
See Torres (1998) for outstanding detail on this process.


13
pledging even greater security cooperation and military transparency. 8 That year, the
last of two-dozen border disputes between the two countries was resolved.
Argentina recently conducted joint military exercises with Brazil and Uruguay at
a 19th century battlefield of what was then Brazil's eastern province of Cisplatine.9
Brazilian, Uruguayan and Chilean officers participate in instruction on peacekeeping
activities at an Argentine training center at an old base near the same previously
conflicted sector. All countries of the region to include Bolivia and Chile, still with no
formal military ties have signed a declaration that establishes the Southern Cone as a
United Nations Zone of Peace. The declaration commits its signatories to the removal of
mines in the region significant given the recent publicity over the minefields that still
exist along the disputed international border of Chile and Bolivia (Foreign Ministers
Spar, 1997).
In the Andean Ridge, Colombia and Venezuela have established a border
commission that coordinates joint military operations. Cooperation between those two
countries extends to naval operations, despite an unresolved border dispute in the Gulf of
Venezuela. And, most significantly, Peru and Ecuador have ratified a peace agreement
8 Argentina, Brazil, and Chile signed a disarmament agreement in 1902 known as
the Pactos de Mayo. Under its terms, Argentina and Chile were compelled to resell ships
to foreign arms suppliers (Varas 1985). Somewhat ironically, the original agreement was
less an expression of trust and confidence than an act to solidify the balance of power
system that emerged in 19th century South America (Burr, 1955).
9 War began in 1825 between Argentina and Brazil over the Cisplatine province
(modem day Uruguay). It ended in 1828 under British mediation leading to the
independence of Uruguay.


246
Argentina is supported by border deployments in Brazil. In 1996, Brazil reported that it
was transferring the unit which had assumed frontier defense responsibilities on the
Argentine border after the Malvinas/Falklands war to duties in the Amazon.
In this context, the United States' growing relationship with Argentina strikes a
discordant note. As part of Argentina's foreign policy about face which led to
rapprochement with Brazil and Chile, Argentina also reached out to the United States,
replacing Brazil as the partner in that "special relationship" (Castro, 1998). Argentina
contributed two ships to the 1990 Gulf War, helped maintain the blockade against Haiti
and supported the United States in its 1994 intervention there (Pala, 1998). This behavior
on the part of Argentina, which lies outside of subregional institutions, seems consistent
with strategic balancing.
The United States declared Argentina a Major Non-NATO Ally (MNNA), a status
which facilitates transfer of defense items from U.S. inventories, in part as a reward for
Argentinas support to international peacekeeping operations.9 Under this arrangement,
the United States transferred 53 armored personnel carriers to Argentina at preferential
prices to support peacekeeping operations ("Confidentiality Accord," 1998). This
arrangement has elicited negative reactions from both Brazil and Chile ("Cardoso
Discusses," 1997), although neither country apparently sees it as a threat to the ongoing
integration of defense policy.
9 Prior to the MNNA designation, the United States has already loosened
restrictions on transfer of new military technology, previously blocked at the instance of
Great Britain. For instance, the United States provided advanced radar for Argentinas
A4M Skyhawk fighters (Pala, 1998).


346
Griffith, Ivelaw L. 1996. Caribbean Security on the Eve of the 21st Century (McNair
Paper 54). Washington, DC: Institute for National Strategic Studies.
Griffith, Ivelaw L. 1998. Security Collaboration and Confidence Building in the
Americas. In Jorge I. Dominguez, ed., International Security and Democracy:
Latin America and the Caribbean in the Post-Cold War Era. Pittsburgh, PA:
University of Pittsburgh Press, 170-208.
Grossi, Rafael Mariano. 1997. The Inter American Security System: Contribution and
Influence of the Argentine Republic in its Origins, Design and Implementation.
Disseration Thesis No. 571. University of Geneva, Geneva, Switzerland.
Gupta, Amit. 1997. Building an Arsenal: The Evolution of Regional Power Force
Structures. Westport, CT: Praeger.
Guzmn, Jaime. 1996. "Seguirdad y defensa: La Concepcin Salvadorea." Paz y
Seguridad en las Amricas 9. Santiago: Facultad Latinoamericana de Ciencias
Sociales.
Haas, Ernst B. 1964. Beyond the Nation-State: Functionalism and International
Organization. Stanford, CA: Stanford University Press.
Hartlyn, Jonathan, Lars Schoultz and Augusto Varas, eds. 1992. The United States and
Latin America in the 1990s: Beyond the Cold War. Chapel Hill, NC: University of
North Carolina Press.
Hasenclever, Andreas, Peter Mayer and Volker Rittberger. 1997. Theories of
International Regimes. Cambridge: Cambridge University Press.
Hirschman, Albert. O. 1945. National Power and the Structure of Foreign Trade.
Berkeley: University of California Press.
Hirst, Monica. 1998. Security Policies, Democratization, and Regional Integration in
the Southern Cone. In Jorge I. Dominguez, ed., International Security and
Democracy: Latin America and the Caribbean in the Post-Cold War Era.
Pittsburgh. PA: University of Pittsburgh Press, 102-118.
Ikenberry, G. J., D. A. Lake and M. Mastanduno, eds. 1988. The State and American
Foreign Economic Policy. Ithaca, NY: Cornell University Press.
Inter-American Defense Board. 1946. Inter-American Defense Council: Proposal for
the Creation of a Permanent Military Agency of the American Republics. Document
No. P-591. Washington, DC: Inter-American Defense Board.


165
Since, as we shall later see, most of the communications involve sharing intelligence
about border security activities, I have included it in the burden-sharing sector.
The only measure that directly addresses defense postures is a weak endorsement
to develop arms control measures. There is no specificity regarding types of weapons
systems or types of control measures. The other measures that affect defense postures
deal primarily with civilian control over the armed forces CSBM conferences involving
military and civilians and the education for peace program.
Education for Peace is a wide-ranging program of civic education to promote the
peaceful resolution of differences within and between groups in both government and
civil society. Approaches range from basic instruction in matters such as negotiation
techniques and crisis management, to longer-term programs aimed at fostering
cooperative attitudes through primary education and other media. Since the goal is to
replace military power as an instrument to resolve differences, I have included this
security cooperation rule under the rubric of defense postures.
In any case, there is still no consensus on what an OAS Education for Peace
program would look like. Predictably, much of the disagreement centers on the extent
that military organizations might be involved. Clearly, promotion of positive civil-
military relations falls under the umbrella of education for peace, but there is great
reluctance to ask any of the hemispheric military organizations to participate.
The Committee on Hemispheric Security was charged to develop a resolution on
Education for Peace for the 1998 General Assembly, leading to a Meeting of Experts on
the topic to be held in 1999. Several delegations including the United States,


236
pilots of transport aircraft have been ongoing since 1980 (Argentina 1996 CSBM
Submission), but since 1994 a robust series of pilot exchanges has developed which
includes exchanges of instructors and fighter pilots (Argentina 1996 CSBM Submission),
with simulation training taking place in Brazil since 1996 (Argentina 1996 CSBM
Submission). Other exchanges between Argentina and Brazil include search and rescue
crews and flight engineers (Argentina 1996 CSBM Submission).
A very significant program begun in 1996 arranges reciprocal visits to units
charged with maintaining radar systems for national coverage, as well as organizations
responsible for planning future distribution of these systems. The next logical step would
be to coordinate purchase of new equipment to ensure compatibility, so that the
overlapping national systems could complement on another. 5
Argentina and Uruguay have developed a similar network, which includes
personnel exchanges at all levels and across many functional areas, both in units and
courses of instruction. Reciprocal visits are also frequent, including visits to each
countries' intelligence centers and visits to air squadrons, which have been ongoing since
1988 (Argentina 1997 CSBM Submission). The contacts there have led to integration of
some sensitive technologies (Uruguay 1996 CSBM Submission) and extensive
cooperation in air defense, which has been exercised since 1989 (Argentina 1996 CSBM
Submission).
5 The organizations involved are the Centro de Instruccin y Perfeccin (a
training organization for non-commissioned officers), the Vigilancia y Control de
Aerospacial (a radar unit); and Cindacta-1 (a planning organization for the distribution of
radars).


259
is a tightly bound security community with security linkages that connect its members at
multiple levels, ranging from shared social structures and transnational elites to interstate
rivalries and competition for external patrons (Mares, 1997). Of major concern in this
study are the security linkages operating at the state level. Historical, those have
consisted of security competition between Honduras and El Salvador, between Nicaragua
and its neighbors to the north and south, and at times competition between Nicaragua and
El Salvador. Another important consideration is the role of Guatemala, the most
powerful state in the subregion, pulled both towards Mexico and its fellow Central
American republics.
In this subregion, far more than in the Southern Cone or the Andean Ridge, the
United States has played a direct role in security affairs, both to protect its economic
investments and to counteract actions by other states from outside the isthmus. But by
the beginning of the 1990s, with the demise of armed resistance movements and the
triumph of liberal economic models, the normal triggers for U.S. intervention had
disappeared. U.S. presence was substantially reduced, leading some to refer to a
security vacuum in the subregion (Sereseres, 1998, p. 214). With the balance of forces
essentially determined by national capabilities, subregional rivalries to returned to
prominence. Traditionally, competition has been sparked by numerous security concerns,
including the relative gains inherent in economic integration, the fate of extra-nationals
immigrating in search of land for agricultural purposes, access to the Gulf of Fonseca and
competition for hydroelectric resources. From a realist perspective, this foretells little
cooperation between El Salvador, Honduras and Nicaragua the triangle at the heart of
Central America that provides the focus of most of this investigation.


179
cooperative security, not its "non-use" in domestic affairs, which is one focus of the
democracy regime.
The organizations and rules that support democracy provide an undercurrent for
much of the effort in institutionalizing a regime of security cooperation, sometimes
reinforcing that regime, other times constraining it. This is because the two regimes
overlap in a number of important areas. Transparency, for instance, while principally
aimed at inter-American security relations, has a domestic component as well. A prior
step to empowering civilians to control military establishments is access to information.
Thus it is hardly surprising to see transparency measures adopted more easily than
measures which address burden sharing and defense postures.
Changes in defense postures are taken to be important indicators of the
effectiveness of the security regime. To the extent that democratic governments attempt
to reign in their militaries by reducing support, there is some evidence of overlap here.
But nothing in the democracy regime facilitates this activity. Some democratic
governments have reduced budgets and removed military privileges; others have not.
Some have ended conscription; some have not. Naturally, the process of
democratization in individual countries has led to reform of their military establishments.
It could hardly be otherwise. But the process is inherently a domestic one, linked to a
hemispheric regimes in ways that are not fully understood.
There is some overlap in the area of burden sharing, although the overlap
crosscuts the two regimes as much as it reinforces them. Peacekeeping operations, for
instance, have become an instrument for establishing democracy in regions tom apart by


139
they encompassed a broad view of security that goes well beyond traditional concerns of
external aggression. Because their small size and enclave economies, the island states of
the Caribbean see their security threatened by transnational phenomena such as drug
trafficking and the crime and corruption that activity brings with it, natural disasters such
as tropical storms which can destroy agricultural products and disrupt the tourist industry,
and a globalized economy that can drive rapid changes in relative prices that negatively
impact economic development. Such an ample view of security is not shared by the
majority of OAS member countries, including the United States, but nevertheless lays out
one vision of what security means in the post Cold War environment. Linkage to
economic development is also used as a negotiating asset by smaller states.
The most important specialized conferences specifically addressed CSBMs. The
Committee organized the First Regional Conference on CSBMs, held in Santiago, which
produced the 1995 Santiago Declaration. The Santiago Declaration, the founding
document for CSBMs in the hemisphere, contains a list of eleven specific CSBM
categories recommended for implementation by member states. Some of these measures
urged member state participation in international mechanisms already established.^
Other measures bear marked similarity to CSBMs employed in Europe. Still others drew
on experiences in the hemisphere, such as the Central American peace process and the
CSBMs employed between Argentina and Great Britain after the war in the
F alklands/Mal vinas.
2 For instance, the Declaration urges states to participate in the UN Standard
Instrument for Reporting on Military Expenditures, established in 1980, and the UN
Registry of Arms Transfers, established in 1993.


133
institutions that move beyond power politics, countries in the region are not ready to trust
national interests to such mechanisms. The institutions were built on the fault line that
separates the U.S. security agenda from Latin American concerns of sovereignty and non
intervention. All of the security institutions of this period reflected this tension, leaving
them largely incapable of ushering in a new era based on military transparency and
security cooperation. They would have to be transformed. Consideration of this
possibility is the subject of the next chapter.


148
was eliminated, to the delight of the Mexican delegation. The U.S. intent had been to
develop new goals and objectives that move the Board more aggressively into the areas of
peacekeeping, natural disaster relief, humanitarian operations, drug interdiction in short,
into areas of high priority to the U.S. government.
With the exception of humanitarian demining, all of these U.S. initiatives have
been resisted by balancers. U.S. led efforts at the 1999 OAS General Assembly to
include the Board on a new OAS committee on natural disasters were defeated by
Mexico, which threatened at one point to recommend the inclusion of a Caribbean natural
disaster organization chaired by a Cuban representative. 1 8 A U.S. initiative to update the
Boards peacekeeping manual, standardize training and establish the IADB as node for
sharing peacekeeping experiences was defeated without debate in the Council by the
Brazilian delegation, despite that countrys deep commitment to UN peacekeeping
operations and training programs and exercises related to those operations. 19 And a
18 The Board was also asked by the United Nations in 1994 to assume
responsibilities as regional coordinator for the use of military assets in support of disaster
relief operations. However, the OAS never clarified the Board's role in this area under
the terms of reference of Resolution 1240 and this request still lies dormant. Within the
Board, efforts to widen participation in an inventory of military assets for disaster relief
operations went unheeded (only five countries participate Argentina, Bolivia, Canada,
Mexico and Peru and the level of participation did not change as a result of the
committees work).
19 Debate also ensued over Board participation in U.S. sponsored peacekeeping
exercises in 1998 when the U.S. Southern Command changed its policy of inviting Board
officials to participate in the role of outside observers. The new policy required Board
observers to actually participate in the exercise, either as role players or members of the
exercise staff. Brazil, along with other delegations, objected vehemently to this change
which some believed to imply an operational role for the Board in peacekeeping.
Ultimately, however, the Board moved forward in both of these programs.


183
intervention by the United States prevent consensus on the matter. Some countries
suspect that such a regime would provide cover for U.S. intervention. Most assume that
the United States would not feel obligated to respect commitments established by such a
regime.
As indicated by the discussion on CSBMs above, some measures relating to
burden sharing do enjoy consensus. This is the case with natural disasters and operating
procedures for air, land and sea transport, to include accident avoidance and search and
rescue. In the case of natural disasters, specific resolutions included this threat to security
early on (Resolution 1179 in 1992 and Resolution 1236 in 1993). There are no operating
mechanisms that coordinate this activity at the hemispheric level, however, and none that
mention the role of the military. Cooperation on transportation procedures, on the other
hand, is well developed, but not new. As reviewed earlier, the technical aspects of this
kind of coordination have been shared for many years, through the conferences of the
armed forces, the IADB and other international forums.
Demining is an area of burden sharing that, although not listed as a specific
CSBM, has become deeply institutionalized. Numerous resolutions have addressed the
demining program in Central America, ongoing since 1993. Demining is an area of
substantive cooperation between the IADB and the OAS Unit for the Promotion of
Democracy. Specific roles and responsibilities for each organization have been
developed and approved by both the OAS and the IADB. The jointly administered
program has a detailed set of operating procedures. Although the program is now limited
to Central America, there have been suggestions that it could be expanded into other
areas, such as the border between Peru and Ecuador.


40
relative to other states in the system. The result is constant security competition, whereby
one states efforts to enhance its security degrades the security of others. 10
Waltz (1979) refined classical realism with a systemic approach, carefully laying
the baseline assumptions of neorealism: States are unitary, rational actors attempting to
maximize power in relation to other states in an anarchic, self-help system. His
assumption that states are unitary actors meant that detailed analysis of the source of
preferences was unnecessary. State interest could be reduced to the need to ensure
survival. Thus, as rational actors states will seek strategies to ensure survival in the
interstate system. That system is anarchic in the sense that no world government exists
to manage security relations between states. States operate in a self-help system -
meaning they can only rely on themselves to ensure survival. Consequently, states vie
inexorably to maintain a position of relative strength vis--vis their rivals.
Perhaps because of its privileged position, realism has been subjected to periodic
challenges that depart from its basic assumptions and attempt to construct an alternate
view of international relations. The principal challengers since the advent of the Cold
War have been variations of liberalism, ranging from the functionalism of the 1950s to
the regime theory of the 1970s, to what is now known as neoliberal institutionalism.^
Realism, as we shall see, is not sanguine about the possibility that institutions can
change the basic conditions in the international system in ways that promote cooperation.
10 Jervis (1978) found that some cooperation was possible in a defense-
dominated world. See Glaser (1997) for a recent answer to Jervis challengers.
11 Keohane (1989) adopts this term, from the liberal institutionalism coined by
Grieco (1988a).


86
Table 2-5: Contending Hypotheses at the Subregional Level for Institutions
Dependent
Variables
Realist
Hypotheses
Institutionalist
Hypotheses
Security
Institutions:
Given institutions as constructed focal points for negotiations and
absolute gains from military transparency, reduced defense postures
and burden sharing, in subregional security systems with general
power parity ...
Organizational
Features
1. Institutions will be under
resourced unless sponsored
by United States, in which
case their agendas will not
stray significantly from U.S.
interests. Issue linkage will
complicate cooperation.
1. Institutions will be resourced by
members. Agenda will be
multilateral, focused on issues
where absolute gains are evident.
Linkages will enhance iteration
and increase benefits of
cooperation and costs of defection.
U.S. influence varies by subregion.
Rule
Creation
2. Rules will be less binding at
the subregional level than the
hemispheric level.
3. Verification mechanisms will
be weaker at subregional
than hemispheric level.
2. Rules will be more binding at
the subregional level than the
hemispheric level.
3. Verification mechanisms will
be stronger at subregional than
hemispheric level to ensure
reciprocity.


99
America: An Exploratory Essay. Burr intended to use balance of power analysis to
provide an integrated pattern for the relations of Latin American nations (Burr, 1955, p.
37). As we shall see, the pattern in Latin America does not conform to the classic, fluid
balance of power mechanism, which requires that alliances form across any dyad. The
system that emerged in Latin America was more static, since historically alignment
choices have been constrained by border disputes and other alliance handicaps.3
Nevertheless, the balancing pattern that emerges is predictable, based on the elements of
threat laid out by Walt. Threats increase to the extent that they are: (1) proximate; (2)
exacerbated by border disputes (fueling perceptions of aggressive intent); (3) backed by
relatively significant military capability; and (4) exist in a military-technical environment
that favors offensive operations.
According to Burr (1955), once independence had matured to the point where
sovereign states could exercise minimum control over their territory and communications
were sufficient to develop awareness of the interrelated nature of security, a balance of
power system began to emerge. The South American system began as two separate
subsystems one based on the Rio de la Plata and the other on the Pacific.
The Rio de la Plata system was centered on the rivalry between Argentina and
Brazil. That rivalry produced a war over the Banda Oriental, which ended with British
mediation in 1828 and the establishment of the state of Uruguay. From then on, any
attempt by one power to dominate Uruguay would be resisted by the other. Paraguay was
3 See Jervis (1985) for his discussion on great power concerts and the conditions
behind a classic balance of power system.


362
Wendt, Alexander. 1995. "Constructing International Politics." International Security 20
(1): 71-81.
Whitaker, Arthur P. 1976. The United States and the Southern Cone: Argentina, Chile
and Uruguay. Cambridge, MA: Harvard University Press.
Wilhelm, Charles E. 1998. Statement before the Subcommittee on National Security,
International Affairs, and Criminal Justice. Committee on Government Reform and
Oversight. Washington, DC: Government Printing Office.
Zaverucha, Jorge. 1993. "The Degree of Military Political Autonomy during the
Spanish, Argentine and Brazilian Transitions. Journal of Latin American Studies 25
(2): 283-299.


97
fifty conventions among Latin American states pledged to resolve disputes amicably. Yet
this was the period of the bloodiest Latin American wars" (pp. 203-4).
Attempts to outlaw the use of force have not worked any better in Latin America
than in other regions. The Calvo Doctrine (principle of non-intervention) and the Drago
Doctrine (principle of equal treatment of foreigners) provided the basis for Brazilian and
Argentine arguments designed to limit the use of force to protect the property rights of
foreign citizens. But these principles had no measurable effect on the foreign policies of
those countries powerful enough to project military power (principally Great Britain and
the United States).
Even the no transfer norm, based on the Monroe Doctrine and backed by U.S.
power, was only indifferently enforced (Corrales and Feinberg 1999). Aside from the
French-Spanish-British invasion of Mexico and Spanish reannexation of the Dominican
Republic, which the United States may not have been able to prevent, the U.S. remained
passive when it could have acted: the British invasion of the Falkland Islands (1833), the
French blockade of Buenos Aires (in the 1840s) and the Spanish attack on Chile (1866).
But beginning at the end of the last century, the establishment of a U.S. sphere of
influence in the hemisphere ensured that no great-power concert of foreign powers would
emerge in Latin America. This development ushered in the asymmetrical power relations
that characterize the hemisphere. Analyzing this system requires use of both realist and
institutionalist tools.


73
relative gains problems. Free riders may enjoy benefits of enhanced security but are
denied benefits of interaction. These points are summarized in the table below.
Table 2-2: Cooperative Problems Regime Solution.
Security
Behavior
Absolute Gain
Regime Solution
Transparency
Better security
calculation avoids
worst case analysis
Reliance on reciprocal
verification.
importance of reciprocal
visits/exchanges.
designing formats that
can be easily cross checked.
Reduction of
Defense Postures
More resources
available for
economic
development and
other goals
Transparency exposes free
riding, cheating and relative
gains problems.
Institutions can address
distributive gains through
safeguards, and trade-offs in
types of reductions.
Regional Burden
Sharing for
Regional Security
More efficient
allocation of
security resources
Transparency exposes
cheating and relative gains
problems. Free riders may
enjoy benefits of enhanced
security but are denied
benefits of interaction.
Institutions can promote
equitable sharing of costs
and benefits, such as
rotation of leadership roles,
cost sharing for exercises,
negotiation of side benefits.


326
preferences in a given international context. Our previous examination established that
context by looking at: (1) the threats and opportunities faced by states (interstate and
transnational threats, absolute and relative gains of cooperation); (2) the capacity of
cooperative security institutions at both the regional and subregional level and (3) the
presence of nesting and linkages (both in terms of their impact on iteration and on relative
and absolute gains). State preferences were assumed to derive from an interest in
maximizing both absolute and relative gains while preserving core capability to ensure
survival in a competitive environment.
Once the state is disaggregated into its military and civilian components, we can
investigate the security politics between those sectors and determine preferences as an
outcome of that interaction as well as strategic interaction at the system level. Such an
investigation does help understand some of the major issues left unexplained by
institutionalism at the system level: the Argentine emphasis on peacekeeping operations
and Chilean and Salvadoran activism in the area of military transparency, for instance.
Jervis (1999) states that one of the most important ways that institutions can
change preferences is through the processes of domestic politics. And yet, theoreticians
have "underestimated the importance of these dynamic effects of institutions" (p. 61). By
adopting a "two-level game" (Hasenclever et al., 1997) we can look at civilian and
military sectors as players who negotiate with each other at the domestic level and with
their counterparts at the system level.
A focus on civil-military relations goes well beyond simple comparisons of
civilian and military governments, which are not generally useful for explaining interstate
behaviors. Acua and Smith (1994) found for instance that the relationship between


123
separate resolutions, remanding specific security recommendations to member
governments, none has even been considered since 1984.30
Continued cooperation after 1965 might have been possible for a time had the
United States offered bilateral benefits or significant outlets for potential balancers. But
U.S. leverage waned for a variety of reasons: a diversion of U.S. interests (Vietnam),
failure of the Alliance for Progress (1961) to deliver on economic assistance in return for
what amounted to a regional defense pact (Villagrn de Len, 1993), and the attraction of
the nonaligned movement. These elements, combined with other factors that made U.S.
assistance less desirable (shifts in assistance programs to counterinsurgency training, a
policy that favored sales over aid, and increasing Congressional controls), led Latin
American countries to search elsewhere for military assistance.31 The restrictive policies
imposed by of President Jimmy Carter accelerated this trend: Brazil broke off military-
to-military relations (as did Argentina and Chile), ending both bilateral arms and training
30 The last resolution considered at the Board Proposed Resolution CXV, dated
13 December 1984 recommended that the Board support the Contadora initiatives with
verification and assistance missions to Central America. The Council voted to postpone
consideration of the resolution. It was never reconsidered. The frequency of resolutions
had already fallen off sharply in the mid 1960s, for some of the same reasons.
31 U.S. refusal to sell F-5 fighters to Peru in 1968 led to its decision to purchase
French fighters and British bombers. This example was followed by Argentine, Brazil,
Colombia and Venezuela (Varas, 1985).


39
cooperation, especially in the context of the current theoretical debate over absolute and
relative gains, figure largely in this examination.
Contending Paradigms
Realism
Realism, tracing its origins back to Thucydides, has enjoyed a long run as the
principal paradigm for understanding interstate behavior, especially as relates to security,
classical realism, which took its modem form under Hans Morganthau (1973) after World
War II, adopts the state as its basic unit of analysis and the twin factors of power and
national interest as the best explanations for state behavior in the international system.^
The tradition, while rich and varied, coalesces around a set of mainstream
assumptions. Keohane (1989) lists three which together make up what he calls the hard
core of Classical realism: (1) states are the principal actors in world politics; (2) they
act as unitary, rational actors; and (3) seek power and define their interests in terms of
power.9 Because power is a relational concept, international politics becomes a "struggle
for power" (Morganthau, 1973). States tend not to trust one another, tend to rely on their
own capabilities to ensure their survival and strive to maintain their military capabilities
8 Keohane (1986) offers a useful review of the development of this school of
thought.
9 Mearsheimer (1994/5) lists five: (1) states operate in an anarchic system; (2)
the principal threat to a state is another state; (3) intentions can never be known with
certainty; (4) survival is the highest goal of states; and (5) states are rational actors. By
rational, we mean that states select strategies based on consistently ordered sets of
priorities (Keohane, 1986).


10
The second case study focuses on the main theme of the study hemispheric
security cooperation in the 1990s. While the dynamics of institutional balancing and
bandwagoning are similar, changes in the international context radically altered some of
the underlying conditions for cooperation present in the earlier epoch. Efforts to
consolidate a new hemispheric security regime have largely been frustrated, because its
construction falls along the fault line that separates U.S.-Latin American priorities for
regional security.
Despite this failure, however, cooperative security behavior is evident in the
region, going well beyond that prescribed by institutional rules at the hemispheric level.
Furthermore, significant geographical variations emerge, such as the degree of military
transparency and burden sharing in the Southern Cone; the depth of defense reductions in
Central America; and levels of security cooperation in border areas within the Andean
Pact. While some of the cooperative activity present conforms to the dynamics expected
by realism, subregional institutions play an important role.
Since the deviations vary by geographical area, we examine subregions in Chapter
5. Subregional comparisons allow a more thorough test of our contending hypotheses
and inform our conclusions with a greater degree of confidence. In the case of the
Southern Cone and Central America, hemispheric security institutions are replicated and
strengthened at the subregional level, while in the Andean Ridge countries, little security
institutionalization exists. Furthermore, in all three subregions, the relative importance of
the United States differs significantly. Thus the case studies allow a clearer look at the
role of both institutions and power and interests. Here, too, institutionalism offers the
better explanation, but here, as at the hemispheric level, it is still incomplete.


261
Cancn Declaration, which specifically addresses security, including steps for ending the
arms race and arms trafficking, eliminating foreign military advisors and creating
demilitarized zones. Perhaps most significantly, the Declaration prohibited states from
harboring armed bands intent on destabilizing a neighbor state. Each of these themes is
reinforced by later institution building in the region.
As Child (1996, p. 19) observes, the declaration also recommended that
verification be ensured through the development of "appropriate supervisory machinery."
Other relevant measures contained in the declaration include appeals to establish joint
border commissions and provide prior notice of troop movements in border regions. In
subsequent meetings, these measures were further developed and codified, but did not see
implementation under the Contadora process, which never overcame the differences
between Nicaragua's position and that of countries close to the United States. Instead,
many of the measures were incorporated into the "Arias Peace Plan" in 1987, as the
Esquipulas process took hold (Romero, 1998).
One important step taken by that process was the reactivation of the Central
American Security Commission, which was mandated by the April 1990 Montelimar
presidential summit. The key objectives of the Commission were to ensure that military
postures in the region were defensive in nature and that a reasonable balance of forces
was maintained, including foreign military presence. The overarching goal was to
develop a new security model based on cooperation, coordination, communication and
prevention. The Security Commission met frequently to discuss reasonable balance of
forces, inventories of weapons and troops, establishment of mechanisms to verify and


CHAPTER 4
HEMISPHERIC COOPERATIVE SECURITY INSTITUTIONS
Introduction
Comparison of cooperative security institutions at the hemispheric level in the
1990s with those of the Cold War reveals the similarities and differences expected by our
hypotheses. One the one hand, we see many of the same behaviors exhibited: the (now)
familiar strategies of the United States (assurance; provision of outlets; offer of side
payments); the practices of institutional balancing and institutional balancing (the
formation of a diplomatic line of resistance). On the other hand, U.S. influence attempts
are far less successful in the 1990s than in prior periods. Institutional balancing is easier
due to the processes of globalization discussed in Chapter 1. The results of this
complicated process of institutional politics is a hemispheric security agenda that
diverges from U.S. interests in significant ways.
To demonstrate this process, we ask three questions. How have the organizations
that make up the inter-American security system (IASS) changed during the 1990s?
Does the evolution of already established organizations and the creation of new ones
simply reflect the dynamics of assurance and institutional balancing? Is the fault line of
U.S. versus regional priorities still evident in institutional design?
Second, how have the rules for cooperative security evolved? In general terms,
are rules binding or voluntary? Are they produced through an inclusive decision-making
134


159
security issues and support for humanitarian operations such as demining. Rules on
defense postures include strong support for civilian control of the military,
encouragement to maintain defense structures at the lowest level consistent with national
security, elimination of weapons of mass destruction from Latin America, and removing
and banning the production, use and stockpiling of landmines. The rules, associated
norms and procedures are presented in Table 4-4, at the end of this section, as a reference
while we consider each of these areas in turn.
T ransparency/verification
Concern in the OAS over measures designed to promote military transparency
began in 1991. At the General Assembly in Santiago that year, member states approved
Resolution 1123, Cooperation for Security in the Hemisphere, which called for the
adoption of measures to ensure hemispheric security and advocated mechanisms of
mutual consultation and an exchange of regional information.
This commitment was reiterated a year later at the General Assembly in Managua.
The Declaration of Managua includes a commitment to continuing and expanding the
dialogue on hemispheric security among the member states. Resolution 1179,
Cooperation for Security and Development in the Hemisphere, called for timely
communication and exchange of information, prior notification of exercises, and support
of member nation participation in the UN arms transfer and defense expenditures
registries.
In a follow-up resolution in 1993, member states added a new transparency
measure: exchange of information on defense strategies (white papers). That years
General Assembly also included a specific resolution on the UN registries (Resolution


325
new wave of democratization has an effect on interstate security relations, but little
agreement on how.
This study of hemispheric and subregional security institutions shows the critical
importance of military transparency as an enabler for more cooperative security behavior
and the strong linkage that cooperative security and democracy share in this area.
Numerous security transparency measures endorsed at the hemispheric level directly
empower civilian control of the military. These include preparation of defense white
books, reporting on military expenditures and arms transfers, and training of civilians on
CSBMs. The end result of these measures is a strengthened democracy and a more
pacific security system. This important theme the links between external and internal
institutions is taken up again in the next section.
Limits of System-level Theories
The purpose of this section is to examine the limitations of system-level
institutionalism and to explore extensions that may offer an alternative explanation for
the patterns of security cooperation that we have observed at both the regional and
subregional levels. The focus here is on the linkage between the security institutions we
have examined and institutions at the domestic level, particularly the constellation of
civil-military relations that prevails in individual countries. Because of the wide scope of
such an undertaking, we will necessarily be selective. The intent here is not to offer
definitive explanations, but rather to suggest directions that might prove productive in
future research.
A major limitation of system-level theories is that unit-level preferences are
assumed, not derived. State behavior is simply the rational outcome of pursuing those


87
Table 2- 6: Contending Hypotheses at the Subregional Level for Behavior
Dependent
Variables
Realist
Hypotheses
Institutionalist
Hypotheses
Security
Given institutions with the administrative capacity to carry out the
Behavior:
functions of regime management (data collection, analysis and
reporting) and rules promoting transparency, defense reductions and
burden sharing ....
Transparency,
4. Generally, transparency
4. Transparency will be
measures will not reveal
considerable, providing
Defense
information of value, but
relevant and reliable
some information sharing is
information, so that security
Postures,
likely between states in a
cooperation can take place to
Burden
balancing entente.
5. Defense reductions are
reap absolute gains.
Sharing
unlikely and in any case will
5. Defense reductions will take
reflect perceived threats.
6. Burden sharing is unlikely
unless sponsored by the
place where transparency and
verification provide assurances
of reciprocity.
United States or resulting
6. Burden sharing will take place
from balancing and/or
where transparency and
bandwagoning. Security
verification provide assurances
cooperation should reinforce
of reciprocity. The presence of
ententes and sharpen rivalries.
transnational security problems
The presence of transnational
makes cooperation more likely
security problems makes
because of the added potential
cooperation even less likely
because of the added potential
for relative gains.
for absolute gains.
Operationalization
Careful consideration of the power and interests of the states in the system and
comparison of those factors with the design of security institutions and content of security
rules will illuminate the relative strengths and weaknesses of the two theories. National


350
Management and Changing World Politics. Albany, NY: State University of New
York Press, 31-54.
Laurance, Edward J. 1996. The Role of Arms Control in Coping with Conflict after the
Cold War. In Edward A. Kolodziej and Roger E. Kanet, eds., Coping with Conflict
after the Cold War. Baltimore, MD: Johns Hopkins University Press, 331-360.
Leadership Council for Inter-American Summitry. 1999. Mastering Summitry: An
Evaluation of the Santiago Summit of the Americas and Its Aftermath. Coral Gables,
FL: University of Miami North-South Center.
Lepgold, Joseph and Thomas G. Weiss. 1998. Collective Conflict Management and
Changing World Politics. Albany, NY: State University of New York Press.
Lijphart, Arend. 1971. Comparative Politics and Comparative Method. American
Political Science Review 65 (3): 683-693.
Liska, George. 1962. Nations in Alliance. Baltimore, MD: Johns Hopkins University
Press.
Linz, Juan J. and Alfred C. Stepan, eds. 1978. The Breakdown of Democratic Regimes:
Latin America. Baltimore, MD: Johns Hopkins University Press.
Lovell, John P. and David E. Albright. 1997. To Sheathe the Sword: Civil-Military
Relations in the Quest for Democracy. Westport CT: Greenwood Press.
Lowenthal, Abraham F. 1992. Changing U.S. Policies and Interests in a New World.
In Jonathan Hartlyn, Lars Schoultz and Augusto Varas, eds., The United States and
Latin America in the 1990s: Beyond the Cold War. Chapel Hill, NC: University of
North Carolina Press, 65-85.
Lowenthal, Abraham F. 1992/3. Latin America: Ready for Partnership? Foreign
Affairs 72 (1): 74-92.
Maingot, Anthony P. 1996. "Haiti: Sovereign Consent versus State-Centric
Sovereignty." In Tom Farer, ed., Beyond Sovereignty: Collectively Defending
Democracy in the Americas. Baltimore, MD: Johns Hopkins University Press, 189-
212.
Mansfield, Edward and Jack Snyder. 1995. Democratization and the Danger of War.
International Security 20 (1): 5-38.
Manwaring, Max G. 1996. "Brazilian Security in the New World Order." In Richard L.
Millett and Michael Gold-Bliss, eds., Beyond Praetorianism: The Latin American


71
Weak cognitivists hold that institutions contribute to this process both by
providing opportunities for contacts which may contribute to learning (Nye, 1991, p.
157) and through dissemination of new information which changes preference formation
(Keohane and Martin, 1994/5). Haas (1964) found that successful integration in Europe
was partly due to a learning process experienced by technocrats. Success in one area
spilled over into others as elites expanded the process of economic integration. Adler
(1988) looked at the power of ideas and learning by technocratic elites in the area of
nuclear power policy in Argentina and Brazil.
Hasenclever, et al. (1997) describe an interesting extension of this idea, based on
the work of Putnam and Zum. Introduction of domestic institutions creates a two-level
game actors within the state apparatus bargain with each other and with counterparts
outside the state simultaneously. This relates to the "second image" and "second image
reversed" logic that frequently appears in the security literature. For instance, Mares and
Bernstein (1999) argue that Argentina's high level of participation in peacekeeping
operations is promoted by the government as a way to provide the military with an
external mission. Laurance (1996) blames slow progress in arms control in Latin
America to the pressure that the military, anxious to protect its budget, exerts on the
process. Jervis (1999) notes that arms control agreements favor "doves," strengthening
their hand in internal negotiations.
There is a vast body of literature related to this theme in Latin America, especially
dealing with the two domestic institutions we are most concerned with the military and
civilian sectors of government. The literature on corporatism, authoritarianism and
transitions and consolidations to democracy are well developed in the field of Latin


282
region. The base is a staging area for deployments of special forces and engineer units
sent to work throughout the subregion with host nation militaries, to improve
infrastructure in rural areas, assist in demining, and provide assets for disaster relief and
support of counterdrug operations. This presence is consistent with the contention that
the United States acts as a security manager in the subregion.
While there is no evidence that the permanent presence in Honduras or temporary
deployments in other countries have a direct impact on defense postures, the U.S.
presence may help reassure Honduras in light of its relative weakness vis--vis El
Salvador and Guatemala. Military assistance from the United States is still flowing into
the region, but it is not providing new offensive military equipment that would markedly
affect the subregional balance of power.
On balance, it seems clear that most of the behavior in the area of defense
reductions is rule-based, not power-based. While there are indications in defense doctrine
and in troop deployments that demonstrate the enduring presence of old rivalries in some
sensitive areas, the weight of the evidence shows steady reductions and reorientations that
are consistent with the subregional model of security integration. Commitments to
maintain a "reasonable balance" appear to be holding, the absence of external monitoring
organizations notwithstanding.
Burden Sharing
The last behavior cluster to be considered here relates to subregional burden
sharing. From the outset, it is important to note that much burden sharing in Central
America in the 1990s has been underwritten by the United States. Two exercises series,


7
Geographically, the hemispheric case study covers the member states of the
Organization of American States, which includes all independent states of the hemisphere
minus Cuba. The subregional case studies focus on three subregions that share strong
common traits that permit identification as regional sub-systems: the Southern Cone,
Central America, and the Andean Ridge. Organizations, legal instruments and security
rules particular to these subregions are analyzed in turn. While a part of the hemispheric
case study, the English speaking Caribbean is not addressed as a separate subregion
because of its markedly different history, language, culture and political tradition, as well
as significant differences in its security environment. Only a handful of Caribbean states
are signatories to the Rio Treaty and other traditional security arrangements of the region.
Few are members of the Inter American Defense Board (IADB). Only recently have the
Caribbean states become members of OAS. The subregion is important and merits study,
especially in light of the significant security cooperation present there, but cannot be
easily compared to the other subregions.^
The time frame is post-1990, a start point that coincides roughly with the end of
the Cold War, the beginning of recent efforts to integrate trade and the end of
authoritarianism in the region. A shift in U.S. policy also became evident early in the
decade, from a security agenda focused on the Soviet and Cuban threats to a new set of
priorities that included controlling proliferation of weapons of mass destruction,
supporting new democracies and addressing transnational security threats along with a
4 Ivelaw Griffiths many contributions are important. In addition, see the recent
volume on Caribbean security edited by Michael Desch (1998) features pieces by
Rosenau, Dominguez, and Griffith, among others.


253
achieve the "common security system" some anticipated, it did fortify the framework for
further military cooperation between the two countries ("Government Denies," 1997).
Menem and Cardoso have agreed that their two countries represent a "strategic
alliance," which includes deep measures of confidence and security building, common
positions in international organizations and a willingness to work together to confront
regional threats ("Mercosur Defense," 1997). Menem did not speak in terms of a military
alliance, but rather a process of integration on all levels ("Menem," 1997). President
Cardoso emphasized the "common military doctrines" of the two countries ("Mercosur
Maneuvers," 1997).
High level officials in both governments have gone further, as thinking has
evolved ever closer to true military integration. The Brazilian presidents chief of the
military staff has spoken openly of a military alliance to pool resources with other
Mercosur members ("Military Household," 1997). The Chairman of the Argentine Joint
Chiefs of Staff has called for the development of NATO-like units in Mercosur to
address regional threats ("Argentine General Proposes," 1998).
The Uruguayan Defense Minister has spoken of deep integration and cooperation,
while denying that a new alliance is forming ("Argentine, Uruguayan Armies," 1998). Of
the major countries in the region, only Chile has categorically rejected the idea of a
security community (Chile, 1998, p. 66). ^
^ 1 See Escud and Fontana (1998) who argue that while Argentina wants to see a
cooperative security model for the subregion, although Brazil and Chile are more
restrained.


199
Added to the methodological difficulties already mentioned is the problem
working the data as a time series, since reporting procedures changed during the period in
question. The 1995 report was more of a modal nature reports on types of CSBMs that
had been employed since the early 1990s while the 1996 and 1997 reports were more
specific in nature. Still, close examination of the individual measures confirms the basic
trends submitted below.
General trends are relatively easy to spot. These general trends will be supported
with detailed analysis of the principal measures themselves.
There is a marked predominance of bilateral and subregional measures over
hemispheric measures; only 8 percent of the measures were hemispheric in nature,
compared to 64% bilateral and 28% subregional. Most of the hemispheric measures
relate to participation in institutions, treaties, meetings and conferences we have already
discussed, although there are some new items. Examples of subregional measures
include participation in subregional organizations and conferences; demining operations;
exercises, especially for peacekeeping and humanitarian operations; and disaster relief
training. Examples of bilateral measures include participation in bilateral institutions
such as intelligence sharing arrangements, security working groups and periodic staff
meetings; cooperation along borders; reciprocal notification of exercises; exchanges of
military units, officers, student and technical experts; visits to installations; observers at
exercises; and cultural exchanges, including athletic events.
As might be expected, judging from the varying strength of regime rules across
sectors, there are far more transparency and verification measures evident in the data than
measures for either burden sharing or defense postures. This finding is consistent with


335
small research nucleus, but it is woefully underfunded. The College library is also in
need of modernization. With a more direct relationship to the OAS, it could also serve as
a special collection for the OAS library. Finally, the College could expand enrollment.
The College is in the process of establishing a distance learning center to expand its reach
without having to increase its residency requirement. This is an important opportunity
that can have a positive impact on the next generation of defense leaders in the Americas.
Given funding and personnel support, the College could also promote a more
ambitious conference schedule. This is a proven capability; over the past five years, the
College has provided excellent support to international conferences on a number of
important security issues. But an even more aggressive schedule, synched with the
Committee on Hemispheric Security perhaps, would go a long way towards replacing the
IADB's function of providing a forum for discussion of hemispheric security issues.
This may be a fairly modest proposal, but fairly modest measures are precisely
what is called for when considering the design of hemispheric security institutions in the
1990s. At a recent working group of senior defense officials from throughout the
Americas, participants addressed the theme of how best to strengthen hemispheric
security institutions. Several suggested ways to rationalize, centralize and otherwise
improve the efficiency of existing institutional tools for cooperative security. One
offered a contrarian idea: perhaps we have exactly the hemispheric security system we
want inefficient, ineffective and decentralized.9
9 The occasion was a November 1997 conference in Santa Fe, New Mexico,
cosponsored by the IADB entitled The Role of the Armed Forces in the Americas: Civil-
Military Relations for the 21st Century.


CHAPTER 3
ASYMMETRY AND SECURITY
Introduction
This chapter deals with the first of our case studies, which provides an example of
the dynamics of institutional security politics at the hemispheric level. Security
cooperation in the Americas begins in earnest with the onset of WWII, tracing a trajectory
that rises during the war years to a highpoint in the 1950s. At its peak, U.S. military
influence reached astounding proportions. U.S. bilateral military relations boasted basing
rights, military assistance programs and logistics support agreements with key countries
throughout the hemisphere. U.S.-led multilateral initiatives resulted in the creation of
hemispheric security institutions, the adoption of common defense plans and the
employment of an Inter-American Peace Force. This level of influence declined during
the difficult 1960s to a turning point arguably at the Buenos Aires conference convened
to reform the OAS Charter in 1967. Attempts to bring the Inter-American Defense Board
under the Charter and create a permanent inter-American force were turned back.
Security multilateralism would not be revived until the 1990s.
This was not simply the inevitable result of the institutional politics of security in
an asymmetrical regional order. In the end, U.S. inattention, diversion of interest, and
aversion to the authoritarianism taking root on the continent, coupled with the impact of
economic development in Latin America, contributed more to the loss of U.S. influence
94


142
of the Inter-American Defense College.5 This scheme was basically consistent with a
recommendation prepared by the Board staff that same year.6
The United States presented its draft resolution on the Board in the context of
discussions that eventually led to permanent status for the Committee on Hemispheric
Security, thereby creating a permanent defense-related organ under the OAS Charter,
something long sought by the United States. But the future of the Board was left
unresolved; ultimately the General Assembly adopted language that simply urged
continued study on the matter. ?
Since most of the elements of this debate are over fifty years old, it is not
surprising that the Committee was unable to resolve the issue in 1994 or in the years that
5 The resolution did not abolish the Board outright, but laid the basis for such a
step. The resolution was drafted in the context of discussions about the future of the
Inter-American Defense Board, which took place in May 1994 in the Committee on
Hemispheric Security. Later, the draft was sent to the Permanent Council (OAS, 1994).
6 See the 1994 report by the IADB Staff, Study of Possible Alternatives to
Determine the OAS/IADB Link (IADB, 1994). The recommendations would have
established a military staff in the OAS. No provisions were made for the College. The
Council allowed distribution of the document to the College but apparently never
forwarded it to the Committee on Hemispheric Security. Nevertheless, dialogue between
the two organizations was considerable throughout the 1993-1995 period: Board
representatives attended Committee meetings convened on the subject of the Board and
the general topic of hemispheric security; Ambassador Patio Mayer addressed the
Board on 12 May 1994 on his Committee's work on the juridical link; the Board
submitted answers to a questionnaire developed by the Committee; and the IADB
Chairman spoke to the Committee in 1995. There was a generous exchange of ideas on
the issue.
7 General Assembly Resolution 1285 called for additional studies on the linkage.
They were never conducted.


308
To the extent that rules are comprehensive enough to adequately address shared
problems, the number of states willing to obligate themselves to those rules declines to
include only those intimately involved in the issue that a particular cooperative activity
addresses.
Cooperative security behavior likewise is more consistent with institutionalism
than with realism. Balancing is still present in all regions, but with the exceptions of
Argentina's new relationship with the United States, continued U.S. presence in
Honduras, and growing U.S. assistance to Colombia, balancing only shows up as a
secondary effect in behavior patterns. Although this may be due to the partial nature of
the evidence, it is an important finding nonetheless.
All the subregions studied strengthened transparency and verification mechanisms
to some measure. The absence of measures between Bolivia and Chile does show,
however, that underlying rivalries can still limit cooperation.
Defense postures, in general, attracted less attention than transparency, although
significantly more progress was possible at the subregional level than at the hemispheric
level. Clearly, movement was most marked in Central America. Countries there were
able to downsize to realize absolute gains even without immediate reciprocity. The
relative lack of progress in the Andean Ridge is surely related to the serious security
problems still affecting that subregion. In the Southern Cone, lagging progress on
conventional arms may point up a relative gains problem. Since these countries produce
arms and arms producers have more to lose than consumers do, the differences between
that subregion and Central America could relate to relative gains. Here is some
countervailing evidence for realism.


145
as a requirement^1 and in the next five years four Caribbean states joined: Trinidad and
Tobago (already a Rio Treaty signatory), Antigua and Barbuda, Barbados and Guyana.12
Surinams application for membership was accepted but the government never installed a
Chief of Delegation. Recently, the return of representatives from the Dominican
Republic and Nicaragua has given the Board even broader coverage.12 This is not a bad
11 The arguments for dropping the requirements, presented by the Peruvian
delegation, offer an early glimpse of the pressure that some delegations felt to make the
Board relevant in the post Cold War period. With seventeen countries in the OAS
outside of the Rio Treaty, the Peruvian delegation argued that the Board in reality only
represented half the Inter-American System. Peru favored opening up admission in order
to "transform the IADB, to give it new agility, so that no one can say we serve no
purpose. We have been the driving force behind the change to ensure that we do not lose
our present effect, and we said, on more than one occasion, that if we do not change, we
will be running the risk of dying." See IADB Minutes to Session 991 (5 July, 1991), doc.
C-2730 (pp. 15-17).
12 But the inclusion of Caribbean states continues to be resisted by some of the
original members of the Board, albeit indirectly. In 1998, the Council refused to modify
its mission statement by changing the word continent to hemisphere, a change which
had been requested by the senior Caribbean delegate present. Chile, and others, argued
that the term continent included the surrounding waters and islands of the American
mainland. The resistance to the Caribbean states goes beyond semantics it may stem
from an unwillingness on the part of some to identify the Board with the broad range of
security issues promoted by the Caribbean members.
12 Nicaraguan representation lapsed from 1989-1999. In 1989, its representative
left the Board in the aftermath of Council deliberations over whether or not to
characterize Nicaragua as a "threat" to the inter-American system. Dominican
representation lapsed from 1990 to 1997 for reasons that are not clear.


209
Most countries also reported participation in subregional organizations that help
coordinate burden sharing activities. Examples include the Regional Security System
(RSS), signed by the countries of Eastern Caribbean,'41 the Central American Integration
System, and the Southern Cone Strategy Symposiums. Most Central American countries
reported participation in the Framework Treaty for Democratic Security in Central
America, signed in 1995, which represents the culmination of years of effort in
confidence-building in that sub-region. Its many commitments to cooperation reflect the
large number of cooperative security measures already in place. The countries of that
subregion, at least those with militaries, appear to be adopting a common approach to
security.
Also significant are the many reports of military cooperation in areas that might
be considered police functions, including recovery and return of stolen vehicles, patrols in
inland waterways and along the coast, and prevention of smuggling. In addition,
agreements to combat illicit trafficking in drugs were reported. These observations
indicate that the military is involved in these areas, despite the lack of hemispheric
consensus on the use of the military to address these transnational threats.
The most vivid example of a humanitarian operation is the demining program in
Central America. Thousands of landmines have been removed from that region.
Unprecedented cooperation is taking place among civilian and military authorities, both
within the affected countries and the international community. Active operations include
Antigua and Barbuda, Barbados, Dominica, Grenada, St. Christopher-Nevis,
St. Lucia and St. Vincent and the Grenadines.


90
"white papers" or other official documents outlining strategy (intent), doctrine
(orientation), force structure (capability), and training/maintenance (readiness); advance
notification of exercises; and exchanges/visits at military schools, installations and
exercises; and weapons demonstrations.
None of these indicators is perfect. Measures reported to the OAS/IADB in the
CSBM inventories are often difficult to verify. Common methodology for comparing
defense expenditures has not been implemented in the region. The UN arms registries do
not address arms production and in any case do not include categories for many small
caliber weapons of importance to the region. Few defense white papers have been
released, perhaps because of the difficulties of coordinating inter-agency positions in
recently established democracies that are still institutionalizing new civil-military
relations. Exercise notification is still spotty. Finally, exchange programs are often not
designed to enhance transparency; they are frequently a tool for influence or deterrence.
Furthermore, exchange officers are sometimes excluded from participating in sensitive
defense matters.
Defense postures are somewhat easier to determine objectively. Good studies on
military capabilities and arms production and transfers are available. Trends in defense
postures during the 1990s can be established. But military readiness is more elusive; it
relies on factors as disparate as a state's recruitment model, morale in the armed forces
and national will. Data in these areas is incomplete or unavailable.
Indicators of burden sharing behavior suffer from some of these same problems.
Not all of the measures listed in Table 2.8 can be effectively verified. Furthermore,
subjective judgments must be made regarding the impact of some of the indicators on


116
The Board also reached its heyday in the 1950s. Planning activity was spurred by
the 4th Meeting of Consultation of Foreign Affairs on Inter-American Military
Cooperation (1951), which addressed the threat of communist aggression in light of the
Korean Conflict. Member states were urged to: (1) increase their resources and
strengthen their armed forces to contribute to the common defense; (2) cooperate with
each other in military matters in order to develop collective strength; and (3) support the
planning for the common defense to be conducted in the IADB.
The heavily qualified preamble to the resolution made clear that these steps were
voluntary, subject to the restrictions of the Rio Treaty, and not intended to prejudice the
security of individual states. These provisos indicate once again the tension between
collective security commitments and the priority of Latin American countries to maintain
sovereignty over defense decisions. Nevertheless, as a result of these directives, the
IADB initiated an ambitious program to develop defense plans to carry out its mission of
continental defense. Its International Staff (created in 1949) was organized into an
intelligence division, to carry out threat analysis and identify national capabilities that
could be used in cooperative security activities, and a plans division, to formulate
directives that would organize collective responses.
The Staff, under this organizational structure, produced plans for the defense of
the American Continent with annexes and coordination instructions to facilitate
communications, logistics, intelligence, transportation, and air and maritime defense.22
22 Separate plans (actually broad frameworks for cooperation) were designed to
address possible aggression from the Soviet Union, Cuba or international revolutionary
movements.


28
The logic of security politics at the hemispheric level is a special case that applies
to institutions that encompass asymmetrical power relations. As such, it can be derived
from the more general form of institutional theory. That theory, along with realism, is
developed in the next chapter. Both provide a set of hypotheses with which to examine
the successful security cooperation at the subregional level, as well as the limited
cooperation at the hemispheric level.


288
of Nicaragua in these operations shows independence from the United States, which until
very recently has had a policy of limited engagement with that country.
That policy changed dramatically in the wake of Hurricane Mitch in 1998. The
U.S. Southern Command deployed an engineer and medical task force to Nicaragua to
assist in reconstruction efforts after the hurricane.21 The country has now been included
in the regular humanitarian exercise program conducted by the United States in Central
America.
Institutionalism, as opposed to realism, would expect this pattern, which
combines both activity by the United States and independent cooperation between the
countries most likely to gain from cooperation. Furthermore, the cooperation is strongest
precisely in those areas that present transnational threats criminal activity on the border,
control of the Gulf of Fonseca and humanitarian operations.
Some of this cooperation is nonmilitary, especially that concerned with the
transnational threats of criminal activity. In counterdrug operations, an area where the
United States is involved, the regional partners are normally police forces. But this
behavior is consistent with the rules the institutions established for these areas. Thus, it is
fair to say that rule-based behavior dominates over power-based behavior.
21 The United States also temporarily established an aviation headquarters in El
Salvador, at the invitation of that government. Disaster relief efforts in Honduras were
supported from the permanent U.S. based in Honduras.


33
organized aggression cannot start or be prosecuted on any large scale. It seeks to
accomplish this purpose through institutionalized consent, rather than through threats of
material or physical coercion, an important element being the creation of collaborative
rather than confrontational relationships among national military establishments (pp. 4-
5).
A cooperative security regime supports these objectives through a set of
institutions, with norms that specify behaviors with regard to use of force.2 Sometimes
the norms simply proscribe force (with the caveat that use of force in self-defense is
always permitted); sometimes they encourage cooperative security arrangements to
address certain shared threats; sometimes they create the expectation of collective
responses to aggression.
An important point is that cooperative security requires a proactive attitude on the
part of participating states to deal with emerging threats before they become open acts of
aggression. Cooperative security is not collective security. Its relationship to collective
security is like that of preventive medicine to surgery (Nolan et al., 1994).
As Nolan et al. (1994) point out, a rich fabric of constraints has already
appeared in some parts of the world that offers instructive examples of elements that
might help construct a cooperative security arrangement:
limits on military operations, such as the various CSBMs in place in Europe and
the Middle East; agreements for averting wars, such as measures covering
accidents, hot lines, and crisis centers between the superpowers; limits on force
2 For Daadler (1992), cooperative security is based on a set of principles, norms
and rules that govern the military dimension of interstate relations created through an
intensive dialogue and habitual cooperation (p. 52).


291
principal competitor, along with it. Brazils presence in the Amazon and its shared
frontiers with Venezuela, Colombia and Peru acts on the subregion as another centrifugal
force.
Nevertheless, a system does emerge from the CSBM data, reviewed in Chapter 4.
That system is centered on Colombia, an Andean country with a large expanse of
Amazonian territory, coastlines on the Pacific and Atlantic and borders with Venezuela,
Ecuador and Peru, as well as Panama and Brazil. Colombias centrality is important in
another regard as well it has become the focal point of the emblematic transnational
security problem in the hemisphere guerrilla and paramilitary warfare fueled by illicit
trafficking in drugs.
Consider this assessment:
Colombia is the most troubled country in the hemisphere. Law and order
have broken down. Drug criminals, guerrillas, and paramilitary self
defense organizations are feeding a spiral of violence and corruption that
makes colombianization a metaphor for a failing state. Every day, about
10 Colombians are killed in politically related strife, while 85 percent of
the 30,000 annual homicides are caused by pervasive criminal violence.
More than 1.3 million people have been displaced by war. To some, the
country appears beyond redemption. (Marcella and Schulz, 1999, p. 1)
The inability of Colombia to contain this violence within its own borders spells security
problems for its neighboring states. Brazil, Venezuela and Peru have reacted by
deploying forces to the border region adding to the potential for conflict as well as
cooperation. The subregion suffers from an insecurity dilemma.
There are three practical reasons for including this subregion in the study. First, it
frames the transnational security problems of guerrilla movements and the drug trade in a
subregional context. Drug trafficking is widely recognized in the hemisphere and the


328
sectors that marks the transition back to democracy (O'Donnell and Schmitter, 1988;
Diamond, et al., 1989).
Despite a consensus that the internal make-up of the state "mattered" in security
decisions the "second image reversed" understanding of how and when remains
elusive in Latin America. The effort has been complicated by global trends in the 1990s.
Just as the state "came back in," it went out. The winds of globalization seemed to
overwhelm domestic institutions; there was a sameness in the adoption of open markets,
open political systems and open borders to new transnational interests.
How did "globalization" affect the military sector? The liberalization of
economic policy reduced the role of the state and forced a decline in central budgets that
directly affected the armed forces (Downes, 1998). In addition, the changing threat
environment made some military missions obsolete, while imposing others for which the
armed forces found itself unprepared. Finally, as the democratic norm became globally
accepted, military establishments bore the brunt of reactions against authoritarian
excesses. Civilian control was exerted over the military in varying degrees, gradually in
some case and more abruptly in others.
There are many ways to characterize the state of civil-military relations in a given
country. Possible indicators of military subordination include a declining resource base,
loss of policy-making autonomy, loss of coercive power, and loss of immunities. The
appointment of civilian defense ministers, increases in legislative oversight, removal of
the military from policing and intelligence roles and loss of conscription are also factors


250
Uruguay (1996 CSBM Submission) reports a series of naval exercises with Brazil:
Amigo (since 1983); Tenderex (since 1995); Salvex (since 1992); and Aguas Claras, a
naval demining operation (since 1996). In addition, Brazil and Uruguay conduct a one-
week naval countermeasures operation and a series of independent peacekeeping
operations (Brazil 1996 CSBM Submission).
Compared to other Mercosur countries, Paraguay has remained relatively isolated.
Argentina does conduct a bilateral exercise with Paraguay involving naval riverine forces
known at Sirena. Indications are that that exercise has been ongoing since 1987
(Paraguay 1996 CSBM Submission). Brazil also conducts a naval operation with
Paraguay known as Platina (Brazil 1996 CSBM Submission).
Chile's contacts and exercises with the original four Mercosur countries are
extensive, but not nearly as advanced in terms of regional integration as the other activity
already reviewed. One example of operational contact is the coordination to avoid
potential confusions during patrolling, taking place between the Argentina and Chilean
gendarmes along the border. A more important example is the Military Observer Mission
in Ecuador and Peru (MOMEP), which Chile participated in along with Argentina, Brazil
and the United States (Pamplona, 1996).
Exercises to promote coordination and develop multinational capability in the
area of peacekeeping operations offer strong evidence of activity related to burden
sharing. The combined forces of Argentina, Brazil and Uruguay annually participate in
operation Southern Cross, the largest exercise in South America, which focuses on
peacekeeping operations. This exercise began in 1994 as a game simulation for staff
officers (Argentina 1995 CSBM Submission) called Exercise Amigos, and quickly


Andean Region 290
Conclusions 307
6. CONCLUSION 311
Review of the Evidence 311
Contribution to Theory 320
Limits of System-level Theories 325
A Modest Proposal 332
BIBLIOGRAPHY 338
BIOGRAPHICAL SKETCH 363
iii


114
mission statement has evolved over the years, these basic functions have not changed.
They derive from the terms of reference established in formal resolutions of Meetings of
Consultation of Foreign Ministers and the General Assemblies of the American States.20
The Board became a useful vehicle for validating the system already put in place
by the United States (Mecham, 1962) and for disseminating U.S. doctrine, widely
accepted in the aftermath of World War II. Early resolutions adopted by the Board dealt
directly with the immediate threat, recommending steps to eliminate clandestine
telecommunications, organize civil defense forces, improve coordination for transit of
military aircraft and develop a convoy system.21 But the war turned shortly after the
Boards creation, and with the likelihood of Axis aggression on the continent low,
hemispheric planners turned to long range defense plans.
By 1945, the Board had recommended steps that if implemented would form a
broad basis for cooperation: a system of naval and air bases; standardization of
20 Resolution XXXIV of the 1942 Meeting of Consultation of Ministers of
Foreign Affairs (the third meeting of the security consultation system set up in 1936 as a
U.S. initiative at the end of the Chaco war); Resolution VII and XXXIV of the Ninth
International Conference of American States (which authored the OAS Charter) and
Resolution III of the Fourth Meeting of Consultation of Ministers of Foreign Affairs in
1951. The 1942 resolution established the Board as the organ for the planning and
preparation of the defense of the Continent. The 1948 resolutions established a weak
juridical link between the Board and the OAS and provided the Board with autonomy
over its internal organization. The 1951 resolution, which identifies the IADB as the
"only inter-American technical-military organ functioning," provided the Board a broad
mandate to plan and prepare against international communism and to establish a
coordinated system to exchange information on defense matters. Just as importantly, it
urged American states to support the Board and act on its recommendations.
21 Mecham (1962) claims that only the convoy system can be credited to the
Board, since the other measures were largely the result of U.S. sponsorship.


192
era. Their purpose was not verification, but rather promotion of the interoperability
necessary for coalition operations.
With regard to burden sharing arrangements, there is little consensus on which
threats are truly hemispheric security concerns, nor which of those should be met by
military means. Mexico has repeatedly criticized any effort to develop cooperative
security arrangements in these areas, as have other countries, fearing that cooperative
security may be a pretext for multilateral interventions. The only areas of military
burden sharing that do enjoy wide support are humanitarian in nature demining, disaster
relief, and search and rescue operations. In other areas, institutional balancing has
forestalled U.S. efforts.
The critical transnational threats trafficking in drugs, small arms and terrorist
activities enjoy varying levels of institutionalization. But military involvement has not
been endorsed at the hemispheric level. There is some consensus on these issues at the
subregional level, where countries share common approaches and have been cooperating
for years. This will become clear when we begin to sort through the actual activities that
countries are reporting.
Progress in formulating rules to reduce defense postures has been even less
successful. The two major accomplishments in this area were the products of regimes
outside of the IAS. Proscription on nuclear weapons and limitations on delivery systems
are major accomplishments in the hemisphere, but not directly related to the institutions
we have discussed. The anti-personnel landmine ban, another important measure, is also
not an IAS initiative. It is questionable in any case whether a ban on an inexpensive,
defensive weapon makes sense for a cooperative security regime. OAS support for this


56
A set of security institutions designed by the United States under these conditions
might look very much like a cooperative security regime. For instance, the United States
can facilitate transparency by providing examples (its own public reporting, for instance)
and training (how-to instruction on preparation of defense white papers, for instance).
The United States can facilitate verification by sponsoring specific mechanisms, such as
open skies fly-overs, and by participating in third party verification, such as the
Military Observer Mission in Ecuador and Peru (MOMEP). In addition, the U.S.
traditional guarantee of protection against extra-hemispheric threats may permit lower
levels of defense postures.
Finally the predominance of the U.S. military model in the hemisphere means that
simply by following U.S. standards (protocols, communications, arms, training, doctrine,
equipment) which are readily available and essentially cost-free coordination for
regional burden sharing is substantially facilitated. Largely, though not exclusively, it is
the exercise of U.S. influence over the militaries of the region and the strong presence of
the United States in multilateral institutions that is responsible for the emergence of a
common military language which promotes interoperability a prerequisite for
effective burden sharing and collective security action. In the functional language that
Keohane (1984) applied to regimes, the United States would be reducing the transaction
costs which serve as one barrier to cooperation.26 But such a regime would be based on
26 The existence of a common military framework helps solve some
coordination problems (e.g., coordination of ship passage), but is not enough by itself
to forge security cooperation. Coordination solves dilemmas of common aversions
(Stein, 1990). Cooperation, practiced to realize common interests, requires policy
adjustments to accommodate partners (Keohane, 1984).


69
The ever-present threat of unilateral U.S. action and the knowledge that
institutional constraints are insufficient to prevent such action undermines confidence that
other states in the hemisphere have in hemispheric institutions. Eventually they will
decay. But along the way, an agenda of security issues and related rules that is
substantially different from U.S. power interests is possible.
By contrast, realism rejects the possibility that institutions can have significant
effects and denies them any independent impact on state behavior. Even if institutions
facilitate agreements and make information available, that does not mean agreements will
actually be consummated or respected when interests clash with regime norms. In the
ultimate instance, regimes cannot guarantee survival. In a self-help system, what matters
is a states relative position. Created in the interests of their most powerful members,
institutions will not constrain behavior and will be abandoned when no longer
convenient.
What happens when we move from the system level and look at preference
formation within member states? This is the focus of the reflective version of
institutionalism, reviewed only briefly here, since we will not return to it until the
conclusion of the study when we consider the limitations of systemic approaches.
Keohane (1989) offers this summary of the reflective approach: "Institutions do not
merely reflect the preferences and power of the units constituting them; the institutions
themselves shape those preferences and that power. Institutions are therefore constitutive
of actors as well as vice versa" (Keohane, 1989, p. 162). Reflective approaches
encompass constructivist theory (Checkel, 1998), critical theory (Mearsheimer,
1994/5) and grotian theory (Krasner, 1983). These labels in turn lump together a wide


185
has emerged under the purview of the Inter-American Drug Abuse Control Commission
(CICAD).35 In addition, the topic has been a consistent theme in many other institutions,
especially the first Presidential Summit and the third DMA.
But there is absolutely no hemispheric consensus on the use of the military in this
area. For illustrative purposes, we will quickly review the Anti-Drug Strategy in the
Hemisphere, developed in 1996 and now adopted by all members of the OAS (I ADC,
1998).
The purpose of the Anti-Drug Strategy is to frame guidelines, of a
recommendatory nature, for cooperative to implement a set of measures and actions to
strengthen national efforts to combat the trade in illicit drugs. These measures are said to
be consistent with existing international conventions, at the UN and elsewhere.
The strategy and its plan of action aim to improve national measures of both
demand reduction (law enforcement, treatment, and education) and supply reduction
(dismantling criminal organizations; controlling precursors; detecting, tracking and
confiscating illegal shipments of drugs; strengthening border controls; and pursuing
alternative development). Cooperative measures between nations include sharing of
information and intelligence and harmonizing national laws. CICAD, as the executive
agent for the strategy, is tasked with developing mechanisms to compile and disseminate
statistics, model legislation and detect production.
Nowhere in the strategy or its plan of action is the military mentioned.
35 CICAD was established by the OAS in 1986 to promote and facilitate
multilateral cooperation among the member countries in the control of drug trafficking,
production and use.


339
"Army Decides to Maintain Presence at Salvadoran Border," El Peridico de Honduras
(Honduras), 19 March 1997, Foreign Broadcast Information Service (on line), 19
March 1997.
Axelrod, Robert. 1984. The Evolution of Cooperation. New York: Basic Books.
Baena Soares, Joao Soares. 1994. Profile of a Mandate: Ten Years at the OAS.
Washington, DC: Organization of American States.
Balconi, Julio Turcios. 1996. "La Seguridad Hemisfrica." Paz y Seguridad en las
Amricas 9. Santiago: Facultad Latinoamericana de Ciencias Sociales, 15-16.
Baldwin, David A., ed. 1993. Neorealism and Neoliberalism: The Contemporary
Debate. New York: Columbia University Press.
Balza, Martn. 1995. "La Seguridad entre los Pases de Mercosur." Ser en el 2000 8: 25-
27.
Barbieri, Katherine and Gerald Schneider. 1999. Globalization and Peace: Assessing
New Directions in the Study of Trade and Conflict. Journal of Peace Research 36
(4): 387-404.
Bennett, Jos Luis. 1996. "La Posguerra Fra en el Hemisferio." Paz y Seguridad en las
Amricas 9. Santiago: Facultad Latinoamericana de Ciencias Sociales, 17-19.
Binnendijk, Hans and L. Erik Kjonnerod. 1997. "Panama 2000." Strategic Forum. Ft.
McNair, Washington, DC: National Defense University Institutute for National
Strategic Studies.
Bloomfield, Richard J. 1998. Security in the Greater Caribbean: What Role for
Collective Security Mechanisms? In Michael Desch, Jorge Dominguez, and Andrs
Serbin, eds., From Pirates to Drug Lords: The Post-Cold War Caribbean Security
Environment. Albany, NY: State University of New York, 121-143.
Bossi, Ernesto Juan. 1998. Una Visin de la Poltica de Defensa de Argentina.
Military Review 78: 35-39.
Brazil. 1996. Poltica de Defesa Nacional. Brasilia: Governo Federal.
Brigago, Clvis. 1986. "The Brazilian Arms Industry." Journal of International
Affairs 40: 101-104.
Brzoska, Michael. 1992. Is the Nuclear Non-Proliferation System A Regime? A
Comment on Trevor Morris Tate. Journal of Peace Research 29 (2): 215-220.


249
aircraft in survival training in the Antarctic. (Brazil 1996 CSBM Submission).
Combined action to address natural disasters, while not common, is occurring. Examples
include the use of Argentine forces to help combat fires in the Brazilian Amazon
("Argentina General Proposes," 1998).
Burden sharing to address traditional threats to subregional security, i.e., joint
actions that pool defense resources to enhance Southern Cone security, have not moved
beyond training exercises. However, these exercises involve substantial resources and
commitments of time and troops. Combined exercises and operations have been the main
vehicle for developing regional capability. Sometimes these exercises are explicitly
linked to regional institutions. CAMAS uses several naval exercise help improve
coordination in this area: COAMAS (a maritime traffic control exercise) Transamerica
and Oceanic. In addition, a large, biannual naval exercise known as Altas Sur, began in
1993 between South Africa and Argentina, with Uruguayan observers. The addition of
Brazil and Uruguay in 1995, with Paraguayan observers, made the exercise truly
multinational. (Argentina 1996 CSBM Submission).
Argentina reports a series of combined exercises across all services with Brazil in
its 1995, 1996 and 1997 inventories. Exercise FRATERNO is a joint naval exercise that
occurs annually, taking place alternately in the territorial waters of each country.
Exercise ARAEX involved carrier operations on the Brazilian aircraft carrier Minas
Gerais, with touch and go landings by naval aviation assets on the carrier. This annual
exercise also alternates between Argentina and Brazilian territorial waters. Uruguay
began participating in 1995 (Brazil 1996 CSBM Submission).


312
But institutions also have a strong, independent impact on security rules and
behavior. At the hemispheric level, the security agenda escapes U.S. control.
Organizations facilitate balancing against the United States that would not be possible
otherwise. They restore symmetry, giving voice to smaller countries and allowing other
security priorities to emerge. Linkages with other issue areas enhance the bargaining
power of smaller states. A more comprehensive review of the evidence reminds us of
how that occurred.
In the post-WWII period, the United States used institutional approaches to build
a multilateral system of continental defense through the latel950s, but that system
ultimately floundered as Latin American states began to resist the United States on a
number of issues: the relationship of the IADB and the OAS, Cuba policy, anti
communism, and the formation of a permanent inter-American peace force. In the 1990s,
the United States once again embraced hemispherism and a hemispheric security
regime reemerged. That regime effectively developed security rules to promote
transparent security behavior a consensus goal in the Americas. Clear linkages to the
democracy regime bolstered transparency in military matters. Table 6-1 summarizes
some of the key historical events in the evolution of the IASS since WWII.
But U.S. efforts to develop military capacity to address transnational threats were
far less successful. And, more telling, smaller countries were able to use the hemispheric
regime to advance their own priorities; examples include a multilateral evaluation
mechanism for the counterdrug effort, a ban on anti-personnel mines and the gradual
marginalization of the IADB.


107
strategically important countries: Mexico, Cuba, Panama and Brazil. A "working basis
was established for wartime security cooperation" (Mecham, 1962, p. 208).
It is in the context of this pattern of unilateral action and bilateral cooperation
driven by U.S. power and interests that the hemispherism of WWII must be understood.
The United States opted to support the creation and promotion of regional organizations
(such as the Inter-American Defense Board), develop regional agreements on collective
security, 13 and support development of multilateral plans to ensure that the South
American continent would not become a base for Axis aggression.
These decisions were not taken lightly. In advocating the creation of the IADB,
for instance, the U.S. State Department overcame its own misgivings about
multilateralizing U.S. foreign policy and the active resistance of the U.S. defense
establishment, which preferred bilateral mechanisms for promoting hemispheric security
(Conn and Fairchild, 1989). Consider the negative assessment by the U.S. services
during WWII when the State Department initiative to create an IADB was first staffed:
It would be an excessively large body, too difficult to manage and direct
for effective action; Latin American military affairs required immediate
action and the establishment of the IADB would require too much time; it
would not be possible to discuss secret plans in a body of that size; the
members of the IADB would lack the authority to carry out the measures
which it might adopt; the IADB would use up the time available to high
level personnel who were urgently required for more pressing duties.
(IADB, 1979, p. 6)
13 Mecham (1962) describes two wings of the IASS: 1) institutions for
peaceful resolution of conflict and 2) institutions for collective security. Construction of
the first wing was complete with the Pact of Bogota (1948), discussed earlier. The focus
in this section is on collective security.


300
Colombia and Venezuela and between Peru and Ecuador stand out in that regard.
Second, institutions emerged in precisely the most conflicted issue areas. They include
comprehensive agreements and verification mechanisms that address both the
transnational security problem centered in Colombia and the interstate competition
between Peru and Ecuador. Third, there is no subregional security institution of
comparable strength to either Central America or the Southern Cone. What about the
pattern of actual security behavior? Evidence will show that concrete activity is limited
principally to Colombia and Venezuela.
T ransparency/Verification
By comparison with other subregions, contacts between Andean countries have
been far less significant. The most prevalent activity has taken place in the
institutionalization of bilateral security meetings, mostly contacts between intelligence
officers at the national level, with some activity down to local commanders in border
areas.
Peru's military diplomacy in this regard has been fairly impressive. Peru has
bilateral meetings on military intelligence three times a year with each of the armed
forces of Bolivia, Brazil, Chile and Colombia. These meetings are complemented by
other meetings between regional commanders. Much of the activity at this level began
during the early 1990s (Peru 1996 CSBM Submission). In addition, Peru reported in
1998 that it had signed conventions with Chile for technical logistical cooperation.
Colombia's activity is also worth reviewing (Colombia 1997 CSBM Submission).
Colombia has bilateral intelligence meetings with Argentine, Chile, and Paraguay, but
most of its activity is with neighboring countries. Colombian military and police meet


263
In addition, the United Nations under ONUCA and ONUSAL and the OAS
International Verification and Support Commission (CIAV), which included one short
mission by IADB observers, provided important external support that helped reassure
states that security agreements were being respected. Regionally, Argentina, Brazil,
Canada and Venezuela all provided military forces to help oversee demobilization (Pala,
1998; Romero, 1998). But that level of support ended by 1995.
Would security cooperation continue in the absence of external support for
verification and third party mediation or would enduring rivalries usher in a new epoch of
security competition? At present, with end of the internationalized civil wars, states in
the subregion are largely status quo. Nevertheless, the growing economic presence of El
Salvador and the looming return of Guatemala into the regional balance of power
equation have placed Honduras and Nicaragua at a relative disadvantage. El Salvadors
white paper frankly recognizes its favorable position (El Salvador, 1998c). Realists
would predict that this power dynamic would drive balancing behavior that would
preclude the growth of subregional security institutions and binding rules regarding the
use of force.
Aside from this potential interstate rivalry, the impact of transnational security
threats on calculations of absolute and relative gains is extremely important in this region.
Border flows of drugs and small arms, immigration concerns, and cross border criminal
activity head the list of concerns of Central American countries (Rojas Aravena, 1996,
p.6). Control of these transnational phenomena along the Honduran borders, in the Gulf
of Fonseca and on the San Juan River has become the top security concern of the


268
1. Notice of Exercises: 30 days prior notification of exercises that meet Security
Commission requirements with respect to number of troops, type and quantity of
equipment and location in relation to international borders (Article 28 of Treaty).
2. Participation in the UN defense expenditure registry and a commitment to
consult with each other on future defense budgets in the Security commission. The stated
goal is to develop defense budgets in concert, respecting the internal situation of each
country.
3. Commitment to a "reasonable balance." This balance explicitly includes the
presence of foreign military forces (Articles 32 and 33).
4. Commitment to eliminate terrorist and organized crime organizations that
threaten regional security (Article 18).
The inclusion of military provisions in the Treaty was not supported by Costa
Rica and Panama, who have registered disagreement with what they regarded as an
overemphasis on military matters (Rojas, 1996b). However, all countries of the
subregion were able to agree on rules proscribing the procurement of modem weapons
systems, adopting a declaration on the abstention from the acquisition of high-cost and
high-technology strategic weaponry and weapons of mass destruction (El Salvador 1998
CSBM Submission).
In November 1997, institution building took another important step forward, with
the creation of the Central American Conference of the Armed Forces (CACAF),
modeled after the hemispheric conferences and built on the same structure as the old
CODECA, which had been functioning from its seat in Guatemala as an organization
reduced to protocolary functions with little operational impact. Membership in CACAF
is currently limited to countries with formal defense establishments El Salvador,
Guatemala, Honduras and Nicaragua although there is no proscription that would
prevent Costa Rica or Panama from joining. The seat of the organization, now in


351
Military in Transition. Coral Gables, FL: University of Miami North-South Center,
223-240.
Marcella, Gabriel. 1997. Ecuador and Peru in the Amazon: The Search for Peace.
Carlisle, PA: Department of National Security, U.S. Army War College.
Marcella, Gabriel and Donald Schulz. 1999. Colombias Three Wars: U.S. Strategy at
the Crossroads. Carlisle, PA: Strategic Studies Institute.
March, James B. and Johan P. Olson. 1984. "The New Institutionalism: Organizational
Factors in Political Life." American Political Science Review 78 (3): 734-749.
March, James G. and Johan P. Olson. 1998. "The Institutional Dynamics of
International Political Orders." International Organization 52 (4): 943-969.
Mares, David R. 1996/7. Deterrence Bargaining in the Ecuador-Peru Enduring Rivalry:
Designing Strategies Around Military Weakness. Security Studies 6 (2): 91-123.
Mares, David R. 1997. Regional Conflict Management in Latin America: Power
complemented by Diplomacy. In David A. Lake and Patrick M. Morgan, eds.,
Regional Orders: Building Security in a New World. University Park, PA:
Pennsylvania State University Press, 195-218.
Mares, David R. and Steven A. Bernstein. 1998. "The Use of Force in Latin American
Interstate Relations." In Jorge I. Dominguez, ed., International Security and
Democracy: Latin America and the Caribbean in the Post-Cold War Era.
Pittsburgh, PA: University of Pittsburgh Press, 29-46.
Martz, John D. 1997. "Contrasting Military Roles in Democratization." In John P.
Lovell and David E. Albright, eds., To Sheathe the Sword: Civil-Military Relations
in the Quest for Democracy. Westport, CT: Greenwood, 15-31
Mastanduno, Michael. 1997. "Preserving the Unipolar Moment: Realist Theories and
U.S. Grand Strategy after the Cold War." International Security 21 (4): 49-88.
Matthews, John. 1996. Current Gains and Future Outcomes. International Security
21(1): 112-146.
Mayor, Patio Hernn. 1995. "The Future of Cooperative Hemispheric Security." In
Richard L. Millet and Michael Gold-Biss, eds., Beyond Praetorianism: The Latin
American Military in Transition. Coral Gables, FL: University of Miami North-
South Center, 1-10.
Mearsheimer, John H. 1994/95. "The False Promise of International Institutions."
International Security 19 (3): 5-49.


163
announcements. The details are left to countries to work out among themselves. As we
shall see in Chapter 5, some subregions have developed institutionalized mechanisms that
specify detail and reporting procedures.
The UN registries, long a favorite in CSBM discussions in the OAS, are singled
out for special mention. Those registries are well established, with specific reporting
procedures and calendars. Work is underway in the OAS to refine the registry formats to
make them more compatible to levels of spending and types of arms transfers common in
Latin America, but those refinements have yet to be completed. Since the adoption of the
Santiago measures, an additional transparency measure has evolved in the OAS that
enhances the UN arms registry. That new measure, endorsed at the 1999 General
Assembly, requires notification of arms transfers within 90 days of entry into inventory.
In some subregions, this effort has proceeded farther. Argentina and Chile, for
instance, have commissioned a study to harmonize their reporting mechanisms to enhance
mutual transparency. The Central American countries have moved on to adopt more
specific weapons registries. Once again, we see evidence emerging that demonstrates the
progress possible at the subregional level.
Exchange of defense policies and doctrines, the subject of "white" papers, is the
third measure. While efforts to standardize the format for this measure are ongoing, no
standard has yet emerged. As noted earlier, the United States and Chile pledged at the
third DMA to hold a workshop on this issue.
Leaving the fourth measure aside for a moment, which deals with defense
postures, the fifth measure promotes personnel exchanges. These programs encourage
transparency. Perhaps more critically, they provide a decentralized network of


269
Guatemala, will rotate every two years to the other member countries, in geographical
order from north to south (CACAF, 1998).
Several important meetings led to the establishment of this permanent
organization. The process began with the "Declaration for Peace, Democracy,
Development and Integration of Central America," subscribed to in June, 1995, in the
seat of the Central American Parliament by the defense ministers of Guatemala, El
Salvador and commander-in-chief of Honduras. In March, 1997, in Roatn, these
countries, with the addition of Nicaragua, met to create the Association of Military Chiefs
of Central America in support of the Secretary of Central American Integration (SICA).
The intent was to provide assistance to SICA in the areas of defense and security. This
association led to a meeting in Tegucigalpa the following month of the member country
armed forces chiefs, who proposed the creation of a Central American Defense
Commission to provide assistance in the area of regional security. In follow-on meetings,
these senior defense leaders agreed on the creation of the Conference of Central
American Armies, ultimately approved in November of 1997 (CACAF, 1998).
The organization's mission is to "encourage confidence and make a continuous
and systematic effort of cooperation, coordination and mutual support with the armed
forces of the isthmus, to contribute to security, development and military integration in
the region" (CACAF, 1998, p. 7). The objectives of the CACAF are to: (1) recommend
concrete actions against threats to democracy, peace and liberty, in order to achieve an
optimal level of defense; (2) exchange information and experiences in all areas of
cooperation and enhance mechanisms for confidence building; (3) participate in
discussions leading to regional integration of defense; (4) promote specialized, technical


352
"Menem on Upcoming Visit to Brazil, Mercosur Leaderhip." Buenos Aires TELAM
(Buenos Aires), 24 April 1997, Foreign Broadcast Information Service (on line), 24
April 1997.
"Mercosur Countries' Military Policies Viewed." El Pais (Montevideo), 26 January
1998, Foreign Broadcast Information Service (on line), 28 January 1998.
"Mercosur Defense Agreement Considered." La Nacin (Buenos Aires), 15 July 1997,
Foreign Broadcast Information Service (on line), 18 July 1997.
"Mercosur Member Countries Start Military Exercise" Buenos Aires TELAM (Buenos
Aires), 7 October 1997, Foreign Broadcast Information Service (on line), 7 October
1997.
"Mercosur Should Intervene in Paraguayan Situation." Folha de Sao Paulo (Sao Paulo),
15 February 1998, Foreign Broadcast Information Service (on line), 19 February
1998.
Mermot, Ral G. 1997. "The Role of the Uruguayan Army as Guarantor of the National
Defense from the point of view of the Future Integration of the American Continent
Looking Forward to the 21st Century." 22nd Conference of the Commander's of the
Armies of the Americas: Salinas, Ecuador.
Millett, Richard L. and Michael Gold-Biss, eds. 1996. Beyond Praetorianism: The
Latin American Military in Transition. Coral Gables, FL: University of Miami
North-South Center.
"Military Household Chief Advocates Regional Alliance." Jornal do Brasil (Rio de
Janeiro), 29 October 1997, Foreign Broadcast Information Service (on line), 30
October 1997.
"Military Leaders, Defense Minister Participate in Symposium." El Observador
(Montevideo), 26 June 1997, Foreign Broadcast Information Service (on line), 30
June 1997.
"Military Maneuvers Reflect of New Military Doctrine." Jornal do Brasil (Rio de
Janeiro), 19 October 1997, Foreign Broadcast Information Service (on line), 23
October 1997.
"Military Maneuvers with U.S. Air Force Scheduled." Buenos Aires TELAM (Buenos
Aires), 30 June 1997, Foreign Broadcast Information Service (on line), 30 June 1997.
"Minister Criticizes Salvadoran Border Demarcation Delays," Tiempo (San Pedro Sula), 9
April 1997, Foreign Broadcast Information Service (on line), 9 April 1997.


187
that other countries consider basing agreements in support of the counterdrug effort.
Recently, Ecuador has agreed to house U.S. aircraft involved in this effort at an airfield in
Manta, on the Pacific coast.
Attempts to organize military cooperation to combat terrorism have met a similar
fate. As in the case of illegal drugs, terrorism was recognized early in the 1990s as a
security threat. In 1996, OAS member nations adopted the Declaration of Lima to
Prevent, Combat, and Eliminate Terrorism and Plan of Action. The purpose of the
Declaration was to intensify regional cooperation and urge adoption of concerted and
effective measures to respond to threats of terrorism.
Cooperation promoted by the Declaration includes: establishing acts of terrorism
as serious crimes, providing mutual legal assistance in investigations and extradition,
refusing to make concessions to terrorists who hold hostages, exchanging information
and intelligence on terrorist groups, strengthening border security, and denying support to
terrorist groups. In 1999, OAS member states agreed to formulate a new convention
based on this declaration.
Again, there is no mention of the use of the military in this area, although several
countries of the regime most notably in Peru and Colombia have employed the armed
forces to attack terrorist organizations. The United States supports use of the military
when it can be effectively used against terrorism. At the third DMA, for instance, U.S.
Secretary of Defense Cohen made a major presentation, which suggested that the armed
forces could be an important tool to use against terrorist organizations.
Below the hemispheric level, however, agreements are easier to work. The
measure addressing small island security offers a glimpse of this dynamic. Although the


356
Perry, William. 1997. Current Trends in U.S. Policy Towards the Western
Hemisphere. Hearing before the Subcommittee on the Western Hemisphere, House
International Relations Committee. Washington, DC: Government Printing Office.
Pinheiro, Alvaro de Souza. 1995. Guerrilla en la Regin Amaznica: Una Experiencia
del Pasado; el Present y el Futuro. Military Review 75 (5): 26-45.
Pion-Berlin. 1992. "Military Autonomy and Emerging Democracies in South America."
Comparative Politics 25 (1): 83-102.
Pittman, Howard T. 1988. "Harmony or Discord: The Impact of Democratization on
Geopolitics and Conflict in the Southern Cone." In Philip Kelly and Jack Child, eds.,
Geopolitics of the Southern Cone and Antarctica. Boulder, CO: Lynne Rienner, 30-
51.
Poitras, Guy. 1990. The Ordeal of Hegemony: The United States and Latin America.
Boulder, CO: Westview.
Powell, Robert. 1994. Anarchy in International Relations Theory: The Neorealist-
Neoliberal Debate. International Organization 48 (2): 313-344.
Ramsey, Russel W. 1999. The Strategic Literature on Latin America in the Post-Cold
War Era. Ft. Benning, GA: U.S. Army School of the Americas.
Rennger, N. J. with John Campbell, eds. 1995. Treaties and Alliances of the World. 6th
ed. London: Cartermill International.
Resende-Santos, Joao. 1996. Anarchy and the Emulations of Military Systems:
Military Organization and Technology in South America, 1870-1930. Security
Studies 5 (3): 193-260.
"Regional Military Rapprochement Seen." Clarn (Buenos Aires internet version), 27
July 1998, Foreign Broadcast Information Service (on line), 27 July 1998.
Regional Security Efforts Along Colombian Border Viewed. El Comercio (Quito,
internet version), 20 September 1999, Foreign Broadcast Information System (on
line), 21 September 1999.
Ripsman, Norrin M. and Jean-Marc F. Blanchard. 1996/7. "Commercial Liberalism
under Fire: Evidence form 1914 and 1936. Security Studies 6(2): 4-41.
Rizzo de Oliveiro, Eliezer. 1997. Poltica de Defesa Nacional e Relacoes Civico-
militares no Governo do Presidente Fernando Henrique Cardoso. Ncleo de
Estudos Estratgicos. Campinas, Brazil: Universidad Estatual de Campinas.


302
that is still in process of demarcation, where demining operations are ongoing, is a
significant demonstration of mutual confidence.
But none of these transparency mechanisms reaches anywhere near the level of
scrutiny focused on the Colombian military as a result of the war and alleged human
rights abuses committed by its soldiers. Colombian strategy, tactics, equipment,
readiness and conduct are a matter of regional debate in the hemisphere. Frequent visits
by senior Colombian officials to Washington as part of bilateral discussions on the
problems heightens interest. Military assistance from the United States is subjected to
end-use monitoring, adopted under the Leahy amendment (Marcella and Schultz, 1999).
There is little about the Colombian armed forces that the neighboring states do not
already know.
Defense Posture
In contrast to the evidence presented for institutions and transparency/verification,
there is little indication of reductions in defense postures in the Andean Ridge. In fact,
the armed forces of Peru, Ecuador, Colombia and Venezuela have all undergone
modernization programs in the 1990s. This is not surprising in the case of Colombia,
given the continued presence of a powerful insurgent forces operating in its territory.23
But Colombia's neighbors have not reduced forces either, despite the successful peace
negotiations between Peru and Ecuador, the economic crises in Venezuela, or the
growing rapprochement between Peru and Chile.
23 On the contrary, what is surprising is the low level of mobilization given the
severity of the conflict (Marcelli and Schulz, 1999).


118
CONDECA, which developed a subregional communications system and conducted
combined operations with member countries.
The impact of the Cuban revolution on the Board was curious, since Cuba
remained a member until January 31, 1962, when it was suspended from the OAS.2^ The
United States called a special IADB session in 1961 and obtained a Council decision to
deny Cuba access to secret materials; nevertheless planning efforts to deal with
communism were undermined. Attempts to suspend Cuba from the Board were stoutly
resisted by Mexico and other delegations up to the OAS decision. But beyond this
episode, institutional balancing was not yet widespread, partly because of U.S. efforts to
assure Latin American states of its commitment to multilateralism.
Many of the organizations mentioned above have a significant multilateral hue;
the IADC provides a representative example. The United States pushed for establishment
of the IADC under the Board, taking great pains to emphasize its multilateral nature.
This step was not without controversy; Mexico objected to the creation of another
security institution at the hemispheric level. But the establishment of an inter-American
defense college had been a widely shared goal for many years and the U.S. plan
prevailed.
The IADC is the only multinational institution in the region dedicated to the study
and teaching of hemispheric security and it attracts distinguished guest speakers that have
included presidents, ambassadors and ministers of government. Still, in many respects it
25 As a result of the Final Act of the 8th Meeting of Consultation of Ministers of
Foreign Affairs in Uruguay.


321
than realism. The analysis conducted in this study of the IASS during WWII and the post
Cold War era shows how the dynamics of institutional balancing and bandwagoning are
supported by institutional features. Table 6-4 provides some examples of this institutional
balancing, taken from our analysis in Chapter 4.
Table 6-4: Balancing Acts Institutional Politics
Characteristics
US Reassurance
Institutional Balancing
Organiza
tional
Features
Forum
Support to multilateralism
Sponsorship of leadership
roles for other countries
Mexican resistance to
closer IADB-OAS
relationship
Agenda
Endorsement of
transparency
Mexican exclusion of
IADB in Education for
Peace role.
Brazil (et al.): resistance
to new focus on
peacekeeping and
counterdrug operations.
Canadian inclusion of
mine ban.
Linkages
Support to democratic
institutions and training for
improved civil-military
relations
Caribbean attempts to
broaden security agenda
to include economic
security.
Rule
Creation
Bindingness
Efforts to enhance response
mechanisms (democracy,
natural disasters).
Mexico (et al.):
resistance to legitimizing
democratic
intervention.
Verifiability
Offers use of open skies
aircraft.
Visits to US installations.
Development of
multilateral evaluation
mechanism.
At the same time that institutionalism is bolstered by these findings, hegemonic
realist theory is undermined. Multilateral institutions cannot be fully co-opted by
powerful states. While power may be fundamental to the creation and perhaps the


150
OAS Security Committee (Mexico submits its declared data on CSBMs directly to the
OAS). In fact, the work of the Staff has gradually begun to merge with the Committee's
work plan. 23 Clearly, much more activity would be possible if broader requests were
drafted and approved by the Permanent Council. But because of the consensual nature of
decision-making at the OAS, it is unlikely that the volume or scope of requests for
assistance will increase.
Fifth, the Board has taken measures to bring its operating costs within the
restricted budget provided by the OAS.24 Because the IADB must justify actual and
planned expenditures each year to the OAS Advisory Committee on Administrative and
Budget Matters, it is vulnerable to attacks on its budget. Mexico has led the charge, with
the United States organizing a defense for the Boards budget with other delegations at
the OAS.
Sixth, the College curriculum and class composition have evolved in ways
consistent with new thinking on security. More emphasis is placed on positive civil-
military relations. The College is collaborating with the OAS Education for Peace
program (Board involvement was blocked by Mexico). An important exercise simulation
has been included which trains students on the intricacies of developing inter-agency and
23 Although, the Board still occasionally acts on direct requests by its member
states. In 1999, the Staff developed a study on natural disaster remediation in response to
a motion by the Honduran delegation.
24 For instance, the civilian staff today consists of one-fourth the number of the
personnel employed at the Board ten years ago. Through a combination of outsourcing,
automation and use of military personnel (with salaries paid by member countries), the
Board has maintained minimum administrative and logistic support to the Staff and
College.


76
We can derive some realist predictions for specific behaviors in
transparency/verification, defense postures and burden sharing as well. Transparency in
military matters should be limited to information that is already public. Reductions in
defense postures should be based on threat perception, not regime rules. And burden
sharing, unless sponsored by the United States or another major power, should be
extremely tentative. In states that share the kind of cross-border problems that we have
dubbed transnational threats, burden sharing should be even more difficult because of the
added source of conflict.
Institutionalists, by contrast, would expect that relevant and reliable information
will emerge from institutional arrangements. Institutions with the administrative capacity
to handle the information will add considerably to the degree of transparency in the
system. The increase in transparency will allow reductions in defense postures to proceed
in a reciprocal manner, where levels might be higher than threats warrant. Under these
conditions, transnational security problems should attract cooperative behaviors, since
absolute gains to such cooperation are relatively high. Again, transparency will insure
that no state benefits unduly from such cooperation. Side negotiations facilitated by
institutions can redress distributional problems that might emerge.
Finally, with regard to institutional effectiveness, realism would contend that rules
will not be effective outside of the U.S. agenda. Patterns of cooperative security behavior
should conform to U.S. priorities. Participation will not be inclusive, but rather
concentrate on key U.S. security partners. This pattern may attain even in the absence of
security institutions, since the United States will pursue its own agenda bilaterally when
multilateral approaches do not advance its interests. Finally, from Mastanduno, realism


141
the important provisos that the status be considered temporary and that the Board not
have an operational military function.
This decision was expressed in the 1993 OAS General Assembly Resolution
1240, which stated that:
Provisionally, and until [a] definition is established, the General
Assembly, the Meeting of Consultations of Ministers of Foreign Affairs,
and the Permanent Council may, based on the criteria of and proper
follow-up by those political bodies of the Organization, require from the
Inter-American Defense Board advice and the delivery of consultancy
services of a technical-military character which in no case may have an
operational nature. Such requirements shall be made by the member states
directly concerned and be within the framework of the Charter.
This resolution represents yet another compromise that puts limits on military
cooperation within hemispheric organizations. Follow-up resolutions have continued to
call for a definition of the legal-institutional linkage, but no additional progress has
been made to date.
In 1994, the United States again attempted to bring the Inter-American Defense
Board under the OAS, in the form of an "Office of the Military Advisor" within the
Secretariat of the OAS. Under the provisions of the draft, the Office would provide
military-technical assistance to the OAS, with the Committee likely assuming oversight


5
power obtains between partners and the number of players is low, the same trends that
make security cooperation problematic at the hemispheric level will instead promote it.
These phenomena are best explained through institutionalist analysis. While
institutionalized cooperation based on mutual interest can sometimes be explained by
realism "defensive positionalists" argue that states may rationally adopt institutional
tools to facilitate cooperation under some circumstances (Jervis, 1986) realism cannot
account for institutional outcomes that contradict the balance of power and interests,
particularly in the area of security.
But as with all systemic models, there are gaps in both the realist and neoliberal
explanations of the security behavior examined in this study. In the concluding chapter, I
suggest that those gaps are best corrected by tools drawn from subsystemic
institutionalism, which link external security behavior to the character of civil-military
relations within individual countries. Interstate behavior may well be the outcome of a
two-level game, in which both states and domestic actors compete and cooperate to
realize their interests.
The scope of the study encompasses the Inter-American Security System (IASS),
which consists of hemispheric security organizations (the Organization of American
States, the Inter-American Defense Board and College, the presidential summits and
defense ministerials), security-related legal instruments (treaties, conventions and
declarations), and institutionalized rules pertaining to the defense forces of the region


126
a portion of Arica (Kelly, 1997). In fact, the seriousness of the negotiations between
Chile and Bolivia prompted Peru to increase its armament levels (Varas, 1985).
The increase in military capacity violated the spirit of the Ayacucho Declaration,
which included a general statement on arms limitations, although consensus on specific
weapons systems was never achieved. Country negotiating positions on arms limitations
reflected their national interests, with two camps emerging: Colombia, Chile and
Ecuador on the one hand, and Bolivia, Peru and Venezuela on the other (Varas,1985).
Once again, the balancing pattern of cooperation and conflict appears.
The conflict between Peru and Ecuador seems to have ended for the present. But
since Peru lost its war with Colombia in Leticia (1932), it has engaged in 34 militarized
disputes with Ecuador over the question of Ecuadors access to the Amazon.36 The most
dramatic incursion took place in 1941, when a Peruvian blitzkrieg forced Ecuador to cede
the disputed Maran Valley and thus its access to the Amazon River.3 7 But the Peru-
Ecuador War of 1995, the most recent conflict in this enduring rivalry, also provided
high drama, triggering a general mobilization in Peru and a reserve call-up in Ecuador.
During the war, traditional balancing patterns appeared again, with Chile and Colombia
backing Ecuador and Argentina backing Peru (Kelly, 1997). Mares claims that Chile
even provided Ecuador with intelligence and communications experts (Mares, 1996/7).
36 See (Mares, 1996/7) for an excellent summary of this enduring rivalry.
37 Kelly (1997) points at that Brazil is interested arranging a land corridor to the
Pacific in this region.


25
military traditions, and common security threats, significant security cooperation has
emerged. Often the trend towards more cooperative security was facilitated by specific
institutions that allowed defense policies to gradually adjust from confrontational to
cooperative strategies. In many cases, these subregional institutions have filled gaps in
the overarching hemispheric model, which has remained incomplete and rift by divergent
interests.
In Mercosur, for instance, links between economic and security institutions appear
to be mutually reinforcing. This is not to say that economic integration always promotes
cooperative security. The Andean Pact, created in 1969, initially triggered a dramatic
improvement in relations between Ecuador and Peru, whose leaders signed a number of
bilateral agreements for economic cooperation, but militarized clashes between the two
countries increased dramatically from 1977 onward (Mares, 1996/7). Colombia and
Venezuela engaged in an arms race during the height of the Andean Pacts success.22 El
Salvador and Honduras fought a war in 1969 at the peak of Central American integration,
over the status of Salvadoran nationals in Honduras who were in a sense the physical
embodiment of that economic integration. 23
22 Mares states that "Venezuela had previously reinforced border defenses after
Colombian guerrillas crossed. But the 1987 appearance of a Colombian navy vessel in
Venezuelan-claimed waters provoked mutual military mobilizations. Similar incidents in
the early 1970s also fueled arms acquisitions" (Mares, 1997, p. 213).
23 Perhaps these economic pacts leave behind institutional traces. Van Klaveren
(1992) asserts that CACM, almost moribund in the 1980s, nevertheless provided a
framework to begin the Esquipulas process that culminated in the Central American
peace accords.


216
indicate that subregional institutions are bolstering regime effectiveness in some areas.
But interesting intraregional patterns emerge as well that cannot be readily explained by
this factor. El Salvador, in particular, emerges as an outlyer when compared to other
countries in its region.
The most important finding in all of this analysis is the geographic clumping of
CSBMs. We have now seen evidence that CSBMs as a whole vary by region, as does
each sector transparency/verification, burden sharing and defense posture. The
principal subregions are the Southern Cone, the Andean region and Central America.
Each of those regions will be analyzed in depth in Chapter 5.
Conclusions
The cooperative security behavior present in the hemisphere does not completely
correspond with either regime rules or with U.S. interests. Neither liberal institutionalism
nor realism can fully explain the pattern, even within the narrowed range of cooperation
and conflict that is bounded by the constraints of large power asymmetry.
Hemispheric security institutions does appear to guide some behaviors.
Transparency CSBMs in the hemisphere have grown from occasional contact at
conferences and institutions to a complex network of personnel exchanges that reaches
into sensitive areas, making up almost 60% of the measures reported. Where other
measures of the transparency regime have been less prolific, the regime has looked for
ways to facilitate compliance; the upcoming workshop on white papers is an example.
Institutions can foster security cooperation in measurable ways. By contrast, in defense
postures we are faced with a null set; a low density of rules corresponds with marginal
reductions in military forces.


287
recommend the creation of a regional information net and enhanced border controls. But
again, as in the case of concrete measures in the fight against illegal drug trafficking,
there are no reports of actual activity in this regard.
Realism hypothesizes that cooperative burden sharing activity should either be
driven by the United States and consistent with its security agenda or emerge as a
byproduct of strategic balancing. The evidence does not support the balancing
hypothesis. The pattern is most concentrated on the three countries most likely to
compete with one another El Salvador, Honduras and Nicaragua.
Substantive U.S. support to cooperative security initiatives is a more plausible
explanation. This is certainly the case with the "Fuerzas Aliadas" exercise programs.
Peacekeeping operations in particular appear to be primarily driven by the United States,
since there is no obvious reason for the subregion to be concerned with this activity.20
Demining programs also count on U.S. assistance, principally training and logistics
support provided in coordination with the Inter-American Defense Board.
That said, independent security cooperation is evident in several important areas,
such as the Gulf of Fonseca and the Nicaraguan borders. In addition, independent
cooperation in the area of humanitarian operations is growing. Furthermore, the inclusion
20 The United States, grateful for the participation of Central American peace
forces in the Dominican Republic (1965) and Haiti (1994) interventions, sees utility in
helping to maintain that capacity. The militaries of Central America certainly benefit
from the combat-like experience and exposure to a large multinational operation. But it
is difficult to see how that capacity relates to security threats in Central American since
the conclusion of the peace process there.


277
institutionalism. As noted earlier, El Salvador is the only country in the region that has
presented a formal white paper on security strategy. That document contains
indications of a commitment to subregional defense. For instance, potential obstacles to
the ongoing integration process are cited as threats. The white paper also recognizes
that subregional conflict is unlikely, supports subregional agreements and cites a
commitment to collective security (El Salvador, 1998c). At the same time, however, the
document affirms the necessity to maintain a strong armed forces for deterrence,
traditional territorial defense and the realization of sovereign national objectives.
Public pronouncements from the other countries in the region that maintain a
formally constituted defense force do not put this ambivalence to rest. In 1996, the
Minister of Defense of Guatemala recognized that the changed regional context allowed
for reduction of arms and a focus on new transnational threats, such as criminality,
narcotrafficking and ethnic unrest in the region (Guatemala claims 23 separate
ethnicities). However, his statements indicated a continued concern over border disputes
and other traditional security threats. His doubts that the detailed provisions established
by the Security Commission under the Democratic Security Treaty could be fully
implemented in the short term strikes a strong note of realism (Balconi, 1996). 16
The Honduran high command, although in general supportive of the ongoing
security cooperation and a broad view of security, still registers concern over the
possibility of militarized disputes along its border with El Salvador. Salvadoran
16 Sereseres (1998) sees the problem of resourcing these measures as
characteristic of the subregion.


119
is a U.S. institution. It is based on a U.S. military installation (Ft. McNair), its budget is
supplemented by U.S. funding, and like the Board, its director is a U.S. general officer
(since the early 1990s, the IADB chairman has been dual hatted as the IADC director).
The original curriculum was created at the time of the Alliance for Progress, when
U.S. policy makers assumed that economic, political and social development would be
necessary to combat what they saw as the main threat to the region revolutionary
upheaval. The College thus began with a program of studies that included non-military
subjects, which would have seemed out of place in a Latin American war college at that
time. The ground was laid for a broad view of security and the threats that might
undermine it.26
By the mid-1960s, a U.S.-led multilateral security system was in place.
As Grossi (1997) notes:
For the first time in the century, it could be said that the United States had
succeeded in establishing a direct line of contact and cooperation between
its armed forces and those of the Caribbean and the South American
Republics. Maneuvers were being held regularly, exchanges of officers
were routinely organized, military field schools were opened indicating a
general consensus for the first time on security priorities of the entire
Western Hemisphere, (p. 185)
26 The first class, with 29 students representing 15 countries, received their
diplomas on 20 March 1963 from then Vice President of the United States, Lyndon B.
Johnson, who gave the graduation address. In his address, he commended the
multidisciplinary nature of the curriculum, which addressed all elements of national
power economic, military, political and cultural. See the 1967 IADB Report on the III
Special Inter-American Conference (Buenos Aires, 15-26 February, 1967), especially pp.
10-18 for insights on the multidimensional thinking evolving in the field of security.


137
Argentina and the Central American states as the principal proponents of a transparency
regime.
The Committee formed first as a working group, with instructions from the
General Assembly to the Permanent Council, in the form of Resolution 1123, to study
and make recommendations on the various aspects of cooperation for hemispheric
security. That report was approved by the Permanent Council during the course of the
year.
One of the reports recommendations was the creation of a Special Committee on
Hemispheric Security, a step the Permanent Council accomplished in 1992, thus further
institutionalizing the study of cooperative security. Although ultimately the Special
Committee would gain permanent status, 1 this was not an easy process. The same
reluctance to militarize the OAS, evident in discussions about the future of the Board in
the post-war years, divided the OAS between those anxious to see more military
cooperation and those who feared that such cooperation would hand the United States a
tool for intervention. Mexico, as in 1945, led opposition to increased hemispheric
security cooperation and remained committed to steering the Committee away from
developing forms of security cooperation that involve use of the armed forces.
Nowhere is this division more evident than in the discussion and reaction excited
by the first attempts to codify cooperative security measures. In 1992, the General
Un 1995, Permanent Council Resolution 653 approved article 24 to the rules and
Procedures of the Permanent Council, which established a Committee on Hemispheric
Security to study and make recommendations to the Permanent Council on any matters
relating to hemispheric security that may be entrusted to it... with a view to promoting
cooperation in the field.


37
step further. If states pool their capabilities or agree to specialize in roles where their
individual capabilities are already substantial and devote those capabilities to the region
as a whole, all states enjoy the benefits of more efficient allocation of security resources.
In addition, there are some regional security problems which may require cooperative
approaches for their solution. Problems such as trafficking in contraband, illegal
immigration, environmental degradation are either impervious to unilateral action or
extremely costly.
This last point brings us back to the beginning. What sorts of threats or acts
might trigger multilateral military action under a cooperative security regime?
Presumably, an aggressor state that openly attacked a member or members of the regime
would invite a military response. But what about the Haitian crisis? Does the disruption
of democracy, the violation of human rights or the threat of instability also merit military
intervention? What about uncontrolled trade in contrabandtrafficking in drugs or small
arms, for instance? If a state becomes a regional base for narco-terrorism, is military
intervention then justified? What about use of the armed forces to cooperate in
nontraditional roles, such as natural disaster relief, environmental protection,
peacekeeping and other humanitarian operations? The short answer to this is that an
urgent security threat is whatever the regime members say it is.
In this study, we will focus on two conceptual clusters that can be derived from
this discussion: institutions and behaviors. Institutions are related complexes of rules
and norms, which vary in terms of their level of institutionalization, ranging from


329
which could be compared across civil-military complexes in the region.^ But the
argument here is a relatively simple one. In states that underwent a process of
democratization during the 1990s that connected civilian and military domestic actors to
their hemispheric counterparts, cooperative security behavior emphasized elements
closely associated with the hemispheric democracy regime. That is, where reorganization
of the civil-military balance involved the participation of the international community, an
emphasis on military transparency emerged.
The clearest case of a country that democratized directly under the tutelage of the
international system was El Salvador. El Salvador's national reconciliation in the early
1990s was aided by massive international assistance and direct verification (Tenneriello,
1996). Most of its reforms have been implemented. This is in stark contrast to the other
states of the subregion, particularly Honduras and Guatemala, which have not yet
undergone a reform process open to outside verification. Nicaragua, although penetrated
by international organizations, has yet to consolidate civilian control over the military.
The civilian defense ministry remains weak.
This is not to say that the Salvadoran military has ceased to be an important player
in domestic politics. But the Salvadoran military as well as the civilian government
sector had substantial international contact during the transition and was directly
influenced by international organizations and other international actors.
^ Pion-Berlin (1992) and Stepan (1988) have fairly exhaustive comparative
templates.


CHAPTER 5
SUBREGIONAL COOPERATIVE SECURITY INSTITUTIONS
Introduction
This chapter takes our investigation of the explanatory power of realism and
institutionalism to the subregional level. The predictions of the two theories will be
compared against the evidence presented by the three subregions studied here: the
Southern Cone, Central America and the Andean Ridge. As laid out in Chapter 2,
realism and institutionalism offer sharply different predictions at the subregional level
regarding the evolution of security institutions, rule-making activity and patterns of
cooperative security behavior.
Realism expects to see a low presence of security institutions at subregional
levels, unless supported by the United States. Organizations and rules that do emerge,
regardless of origin, should be ineffective in overcoming relative gains problems,
especially under conditions of high threat exacerbated by a conflictive history and recent
declines in relative power by a member of the subregional system. Rules will not be
binding. Cooperation may ensue, but will not be rule-based. Rather it will conform to
the same balancing dynamics we observed in Chapter 3 when reviewing subregional
rivalries. Under only very limited situations will the member states of a subregion be
able to cooperate to realize the absolute gains inherent in transparency, reduced defense
postures and burden sharing. Links to other issue areas will not be helpful, but rather
introduce another element of competition to the subregion.
218


24
non-intervention, which also forms the basis of its Cuba policy (Gonzlvez Glvez,
1998). In 1999, when the United States pushed for an even stronger mechanism to
"defend" democracy, resistance from key Latin American states killed the initiative.2^
Sovereignty and non-intervention, as well as democracy, are norms of OAS Charter.21
Thus at the regional level, democracy has not been as significant an integrating
factor as some expected, in part because of the presence of marked power asymmetry.
But at the subregional and unit levels, the impact of democratization on security policies
has not been any easier to characterize. While most analysts agree that democracy has an
impact on security, few agree on precisely what that impact is (Hirst, 1996). The
consolidation of democracy in Latin America has taken a variety of paths and its foreign
policy expression is equally varied.
Perhaps more significant than the simple adoption of democratic forms of
government is the general rise in transparency between countries, based on enhanced
information flows, open markets and open political systems. This transparency has its
major impact in security relations between neighboring states.
For all of these reasons, conditions for subregional security cooperation were
enhanced in the 1990s. In subregions with substantial economic interaction, shared civil-
20 At the 1999 OAS General Assembly in Guatemala City, the United States
authored language to supplement Resolution 1080. An automatic OAS response would
be initiated upon detection of warning signs that democracy was failing, instead of
waiting for an actual disruption.
21 The "guiding principles" of the OAS are numerous. As stated in the Charter,
they include respect for international law, human rights, sovereignty, noninterference in
domestic matters, rejection of wars of aggression, and a recognition of the importance of
social development, justice and education, among others (OAS Charter, Articles 2 and 3)


164
verification between participating countries. In fact, this is the only measure that contains
a verification mechanism of any kind. Together, these four measures make up the heart
of rule making in the transparency and verification.
Although we will address burden sharing and defense postures in separate
sections, for the sake of continuity measures for those sectors listed in the Santiago
Declaration are considered now. The first burden sharing measure cited relates to
prevention of transportation incidents. In air and sea maneuvers, this measure has a long
history of interoperability and coordination. A body of international agreements and
standing operating procedures govern military cooperation in these areas. However,
there is little to no experience or established regulatory procedures for ground
transportation.
Another burden sharing measure listed in the Santiago Declaration includes the
only "transnational" threat that merits specific mention as a CSBM category: natural
disasters. There are no specific provisions mentioned for cooperative programs to
address this threat, although, as we shall see, some subregions have developed such
provisions. The CSBM urging consideration of the security concerns of small island
states rightly belongs to burden sharing, since almost all of those security concerns are
transnational in nature smuggling of drugs and armaments, organized crime, and natural
phenomena.
The last measure that can be considered belonging to the burden-sharing category
is development and establishment of communications systems along the border. This
measure is a mechanism that potentially supports both transparency and burden sharing.


27
However, as long as there is a mutual interest in the gains to be secured through
security cooperation (e.g., defense of the continent against the Axis countries in WWII),
institutional support for U.S. initiatives is likely to be forthcoming. Over time, a natural
decay may set in, as more issues are brought under the purview of the institutional
arrangement, leading to a more complicated agenda that is harder to support and easier to
subvert. At some point, a balancer may emerge at different times in U.S.-Latin
American relations, that role has been played by Argentina, Brazil and Mexico. Since in
an assurance game, cooperation quickly unravels once defection is probable, institutional
balancing can quickly develop into significant resistance.
The conditions wrought by globalism in the 1990s are particularly auspicious for
institutional balancing. The complicated security agenda makes it much more difficult to
define a set of security objectives that are equally important to all member states. This
same phenomenon makes it easier to link issues in ways that benefit smaller states.
Finally, the expansion of the inter-American system, with the incorporation of Canada
and the English speaking Caribbean, has added to the pool of potential balancers and
exacerbated the well known cooperation dilemmas facing large groups.
If all of this simply meant that multilateralism is difficult and likely to cause the
U.S. to resort to unilateral action or engage in states bilaterally ("divide and conquer"
tactics), then the institutional politics described above would be of little import. But in
fact, the outcomes that the process generates are important; they have an independent
impact on cooperative security behavior and the behavior that results is not always
consistent with U.S. priorities.


266
activities in the aftermath of Hurricane Mitch. Neither Nicaragua nor Honduras has had
students at the Inter-American Defense College in recent years.
Within the subregion, the impressive institution building in the area of security
cooperation that began with the peace process has continued unabated throughout the
1990s. The process is principally driven by subregional initiatives, although several
institutions have strong connections to hemispheric institutions mentioned above.
Four principal institutions are reviewed here: the Security Commission, which
was created as an explicit part of the peace process; the Democratic Security Treaty,
which includes detailed CSBMs and an institution charged with developing an annual
action plan; the Central American Conference of the Armed Forces, which was granted
observer status at the Conference of Americas Armies in 1997; and the armed forces
forums for a Culture of Peace and Non-violence.
The Security Commission, as reviewed earlier, was a direct outgrowth of the
Central American peace process. The Security Commission has met almost quarterly
since 1994, rotating its seat to each of the member countries (Honduras 1997 CSBM
Submission). It quickly approved a format for military inventories, identified types of
weapons for ceilings, and forged agreements on arms trafficking and areas of common
criminality.
In recent years, the Commission, which in 1997 was transformed as the
Subcommission on Defense and Public Security under the Central American Parliament
(El Salvador 1998 CSBM Submission), has developed a detailed action plan for the
implementation of specific measures to enhance military transparency and security
cooperation. The 1998 action plan includes among its provisions the reporting of military


190
not succeeded in winning endorsement of its concerns regarding transnational threats.
Second, the strong presence of the United States is evident in almost all hemispheric
institutions.
In sum, the indicators for organizational features developed in Chapter 2 show
more support for institutionalism than realism. There are ample organizational forums to
serve as focal points for negotiation of security rules. It is important to note, however,
that the structure is still rift by the failure to bring the IADB fully under the OAS. The
issue agenda is multilateral, reflecting opportunities for smaller countries to voice
preferences. Institutional balancing is evident in Mexican and Brazilian behavior.
Linkages and nesting tend to reinforce regime rules, with the exception of the area of
burden sharing.
What of the rules that those institutions establish? First, the rules are not
comprehensive in scope. There appears to be an inherent contradiction in attempts to
make rules both inclusive and comprehensive the divisive politics driven by power
asymmetries is one cause. Latin American countries are hesitant to include a wide
panoply of security mechanisms under the purview of multilateral organizations that
include a strong U.S. presence.
Second, the rules are not binding. Participation and reporting remain voluntary,
reporting formats are variable, thresholds for reporting are not established and no
mechanism exists for independent verification. For purposes of comparison, it is worth
noting that more rigor in all of these areas was necessary in Europe before states became
confident that CSBMs were providing true transparency and mutual understanding -


CHAPTER 6
CONCLUSION
Review of the Evidence
The purpose of this chapter is to review the evidence presented in the study and
consider in what ways the analysis may contribute to institutionalist and realist theory.
The chapter also includes a discussion of the limitations of system-level theory and a
brief excursion that suggests examination of the interactions between international and
domestic institutions may improve the explanatory power of institutionalism. Finally, I
conclude with a modest policy proposal.
On the whole, that evidence offers significantly more support to institutionalism
than realism. With the exception of the Andean Ridge, this is true for all of the cases
studied, both at the hemispheric and subregional levels. That is not to say that power and
interests do not play an important role. The U.S. security agenda holds as the single most
important factor affecting the evolution of both the post-WWII and post-Cold War
hemispheric security regimes. U.S. power and resources are usually necessary to
establish hemispheric organizations. Consistent with realism, Latin American states are
reluctant to entrust their security to those organizations. The rules they promote are for
the most part voluntary, partial and weakly supported by verification measures.
Furthermore, in each of the subregional cases studied, balancing patterns are visible in the
observed cooperative security behavior.
311


210
Costa Rica, Guatemala, Honduras and Nicaragua. Argentina, Brazil, Colombia,
Venezuela, Guatemala and El Salvador provide supervisors to the program, under the
direction of the IADB, and the United States provides considerable logistics support and
training.
The effects of Hurricane Mitch in 1998 delayed operations for a period, but
demining is now proceeding after the massive hemispheric support moved to the region
to help in disaster relief operations. Argentina, Canada, Panama, Mexico, Uruguay and
the United States all deployed military assets to assist in the effort.
Peacekeeping and natural disaster training exercises are also reported in the
CSBM submissions. These are generally, but not always, sponsored or cosponsored by
the United States. Separately from the United States, the countries of Argentina, Brazil
and Uruguay participate jointly in the largest exercise in South America Southern Cross
- which is a peacekeeping exercise involving thousands of troops. Likewise, Central
American countries now have their own disaster relief training exercise, modeled on but
independent of the U.S. sponsored exercises in the region.
Each of the guarantor states of the Rio Protocol reported participation in the
Military Observer Mission for Ecuador and Peru (MOMEP). Military personnel from
Argentina, Brazil, Chile and the United States worked together in a joint command (under
Brazilian leadership in 1998) to carry out observer missions on the border. MOMEP
required substantial U.S. logistics support initially, but with the passage of time other
countries assumed those missions. Brazil took on the aviation support requirements in
1997.


42
action will normally not be possible in the international system and as a consequence
collective goods will not be supplied. 14
Of the three clusters of behavior we are most concerned with military
transparency, defense postures and burden sharing transparency and the enhanced
regional security produced by burden sharing most closely approximate collective goods.
Information about military activities, once made public, is available to all and its utility
does not diminish as the number of states with access to the information increases.
Enhanced regional security is more problematic more difficult to organize and much
easier to privatize. But in its pure form, regional security also approximates a public
good.
Thus both of these goods are subject to free rider problems, which means, that
absent other factors, we should expect to see transparency and regional security provided
at suboptimal levels.
Another dilemma of cooperation is nicely illustrated in the context of the security
dilemma described earlier. There are two distinct problems captured by the security
dilemma: cheating and relative gains.
Cheating problems are often illustrated by game theoreticians through the
Prisoners Dilemma. 15 in brief, two prisoners, accused of collaborating in a criminal
activity carrying a stiff sentence, are held in separate cells. Their jailers have only
14 See Stein (1990) for a summary of this reasoning and a list of its major
contributors.
15 The numerous models used in game theory to describe the strategic behavior of
states are presented in Oye (1986).


101
balance of power in both systems.^ However, the fusion of the two systems was in
reality completed only as a result of the Pacific War (1879-1883).^
The complex foreign policy maneuvering which preceded the war shows how the
merger took effect. In 1873, Peru and Bolivia signed a secret alliance against Chile. Peru
sought aid from Argentina, which in turn was countered by growing dtente between
Chile and Brazil. Argentina was drawn to Peru and Bolivia, due to its developing rivalry
with Chile in the Patagonia region, however the threat of a countervailing Brazilian-
Chilean axis forestalled Argentine plans.6
Chiles victory in the Pacific War aroused resistance throughout the hemisphere.
Peru demanded the return of territory lost in Tacna and Arica and Bolivia pressed for an
4 Burr (1955) and Caviedes (1988). As Argentina exerted control over the Rio de
la Plata and began to populate its frontier, the country assumed the central position in the
Southern Cone. This engendered a rivalry with Chile over control of the Patagonia
region, the South Atlantic and Antarctica.
^ In addition to bringing about the merger of the two systems, the Pacific War
was "system-transforming" in the sense that it changed the major players, eliminating
Peru and Bolivia from the top ranks and solidifying Chile's position. It also had the effect
of elevating Argentina, which shortly thereafter began an intensive modernization
program to match Chile (Resende-Santos, 1996).
6 The entente between Brazil and Chile served as deterrent to Argentine
aggression but never materialized in military terms. As Resende-Santos notes (1996)
Brazil was Chiles natural ally during this period, but weak militarily, tending to free ride
on Chilean power. Ultimately an increased threat from pro-Germany Argentina during
the years leading up to World War I forced Brazil to launch its first major modernization
program.


235
By far the most significant network of exchanges in the Southern Cone is between
Argentina and Brazil, as shown by the distribution of CSBMs reported by countries in the
region. The network of reciprocal exchanges now includes personnel programs in
logistics, operations and support units and courses. It includes officers, non
commissioned officers and cadets. The flow of information that results from such a deep
network of interaction is significant. While some of these programs date back to 1980 -
such as cadet exchanges and student officer exchanges in the Command and General Staff
Colleges of each country most were initiated in the 1990s. For instance, exchanges of
intelligence officers began in 1996, exchanges of naval officers on ship deployments
began in 1994 and a program to send officers to each countries' Air Force medical course
began in 1996 (Argentina 1997 and 1996 CSBM Submission).
Student exchanges between the two countries have also become more frequent in
the 1990s, due to agreements that regulate the reciprocal nature of these kinds of contacts.
For instance, in 1994 Brazil and Argentina began a program of exchanges in their senior
defense courses, based on a protocol signed in 1988 (Brazil 1995 and Argentina 1995
CSBM Submission). Argentine and Brazil also participate in each country's
jungle/survival school (Argentina 1996 CSBM Submission).
Sensitive information is being shared. Reciprocal visits between the air forces
have involved programs that increasingly include sensitive installations in both Argentina
and Brazil. Since 1994, cadets from both countries have participated in reciprocal visits
to each other's aviation schools. A similar program with non-commissioned officers
began in 1996. In addition, Brazil (1997 CSBM Submission) reports reciprocal visits to
fighter bases, aircraft maintenance centers and the survival centers. Exchanges among


214
S ;
V '
)
I \
Dominican
public
~W. $*nt H*IU:S t0*"*** PMt!i ic< ti si
lflM Kinotion ^>rt' / AnfefWl *"d
, iswqnan *u-Print* St Kill* w-4
Honduras **
North
Atlantic
Ocean
North
Pacific
Ocean
feswg? Sf*nA -
Notes:
1. Lines show pattern of security
cooperation *,*,
2. Top 21 bilateral relations included p
3. Thickness of line indicates relativeL*
share of total
4. Size of end point indicates relative
number of measures by country
South
Pacific
Ocean
sytet hit'ri
(CWa)
Scale i 50 000.000
1300 KiiQireters
J ri L-1-
Attmrtbtt Eqvai-Aiaa Proj+ctw
Falkland {stands
...M M*ts Malvinas!
G* by U K
cUimed fey Argtfttiaa)
South Gr$.- and fha
South Sandwich islands
UdmaiaUtM fe ..
eJmw4 by Aigast>a{
Figure 4-2: Most Significant Bilateral CSBM Dyads by Number of Measures Reported


11
In the concluding chapter, we look at the limitations of the systemic approaches,
both at the hemispheric and subregional levels, and consider extensions and alternative
explanations. An institutional approach that connects domestic politics with hemispheric
security institutions looks promising, although its explanatory power may be limited to
special circumstances. We explore the approach and suggest a research design for further
investigation. The chapter concludes with a modest policy recommendation.
Recent Trends in Hemispheric Security
Is the emergence of security transparency and cooperation a significant break
from past interstate behavior in the hemisphere? The level and frequency of interstate
conflict in the hemisphere has historically been low when compared to other regions. By
most measures defense expenditures, size of armed forces, numbers of conflicts,
battlefield deaths Latin American is the least militarized major region in the world7
Security cooperation has not been uncommon and third party mediation has sometimes
helped limit conflict that does arise. In this environment, slight increases in security
transparency and cooperation may not be considered remarkable, especially in light of
integrative processes currently ongoing in the political and economic arenas.
This disclaimer notwithstanding, the use and threat of force in Latin American
foreign policies has been a frequent feature of the almost two centuries of independence
7 This characterization of Latin America has broad support. See Dominguez
(1998), Poitras (1990), Kelly (1997), Van Klaveren (1992), and Ramsey (1999). Desch
(1998, 256) notes that Singer and Small found that the Western Hemisphere had the
lowest number of wars per year (0.11) and the lowest battle deaths per year (3,100). But
it does not follow that the region is free of security competition. In the early 1980s, the
region accounted for almost 20% of the Third Worlds total arms expenditures (Varas,
1985, p. 38).


293
Third, the case study focuses on a subregion with the least effective experience in
regime nesting. The Andean Pact has been bedeviled by defection. Measures taken to
revitalize the economic accord, beginning in 1991, increased trade and investment
between members countries. But results are poor when viewed comparatively.
Furthermore, the Andean Pact has not included security items in its portfolio until very
recently. It is important to note, however, that economic relations between Colombia
and Venezuela (members of the Group of Three with Mexico) are characterized by a
relatively high level of interdependence.
The historical pattern of conflict and cooperation in the subregion, as developed in
Chapter 3, showed a major power dynamic to be competition between Colombia and Peru
over the status of Ecuador. The Leticia conflict between Colombia and Peru (1932-3)
nearly escalated into a major war (Kacowicz, 1997). But from the 1960s on, both Peru
and Colombia became involved in protracted internal struggles with powerful guerrilla
movements. These internal concerns sapped some of the salience from bilateral security
competition remoting to the 19th century.
In any case, that competition was overshadowed in the latter half of the century by
the Peru-Ecuador conflict in the Cenapa region, near the Colombian border. The
Peruvian-Ecuadorian conflict in 1995 shocked the inter-American policy community in
Washington. The assumption that democracies are pacific lay at the heart of initiatives to
strengthen democratic institutions and hastened their consolidation in the region.
In the 1970s, arms competition between Colombia and Venezuela also became an
important security concern. The pattern became one of bilateral competition between
Peru and Ecuador on the one hand, and Colombia and Venezuela on the other, while to


195
Table 4-4 continued
Principle
Norm
Rules
Procedures
Reductions in
defense postures
will make use of
force less likely
and thus promote
peace
No proliferation of
weapons of mass
destruction
(chemical,
biological, nuclear)
in Latin America
Treaties/conventions of
Tlatelolco, Mendoza, Cartagena
Established in
treaties/conven
tions
Develop
conventional arms
control programs
Hold talks; enter into conventions
Varies
according to
sponsoring
organization.
Continue the
dialogue on
CSBMs
Hold conferences to develop new
procedures.
Include legislators and other
members of civil society.
Hold in
organizations
of IASS.
Hold in
diplomatic
training
institutes,
military
academies,
research
centers,
universities.
No use, production
or stockpiling of
landmines
Enter into convention
Ottawa
Convention
Submit inventories on landmines.
CHS inventory
format.
Mines on border
should be
eliminated
CA demining program
Established by
OAS UPD and
IADB
Civilian control of
military
Establish education for peace
programs.
Still under
review. OAS
held meeting of
experts in
1999.


224
Argentina, Chile and Peru might work together to wean the buffer states away from
Brazil. Such a competitive dynamic would feature cooperative security arrangements
between larger powers and the smaller states. This level of cooperation is consistent with
realism. In fact, if realism is a valid guide to state behavior, some combination of these
patterns should be readily observable, although conceivably a policy of reassurance on
the part of Brazil might forestall such a reaction.
Argentina offers an interesting case within the larger context of the Southern
Cone. How would Argentina react to its condition of relative weakness? Under realist
interpretations, Argentina would have to either redress this situation with its own
resources (internal balancing) or through alliances (external balancing). In any case,
realists would not expect Argentina to embrace an institutional option and trust in a
network of cooperative security arrangements with Brazil to compensate for its relative
weakness. In this sense, Argentine behavior in the 1990s offers a good test for the
explanatory power of our contending theories.
Given the distribution of power and the historical alignments, what absolute gains
are operative in the subregion in the areas of transparency and verification, defense
postures and burden sharing? Do subregional institutions arise that facilitate the capture
of these absolute gains?
Excellent conditions for the success of transparency measures prevailed in the
Southern Cone in the early 1990s. All were democracies with civil-military relations
generally evolving in ways that pulled security issues into the public discourse. While
progress in this area varied widely, there is little doubt that it had proceeded farthest and
fastest in Argentina- the one large country which may have felt a need to conceal the


176
Nowhere in the convention is the armed forces mentioned, neither as a principal in
the effort to control arms production, nor as an agent to detect and confiscate illegal arms.
It is instructive to note the high level of cooperation possible when a security issue can be
considered as a technical problem, abstracted from the contentious debate over the
militarys proper role or the design of hemispheric security institutions.
In sum, military stockpiles have been left unscathed by hemispheric rules on
defense postures, with the high, strategic exception of nuclear weapons and the low,
tactical exception of land mines. Nothing in between is regulated at the hemispheric
level.
Where does the United States stand on this? The United States possesses nuclear
weapons, but those holdings are not subject to the Treaty of Tlatelolco. Limitations on
nuclear transport are a potential problem for the United States and its European allies, but
no specific hemispheric measures have yet been endorsed in this issue area. Elimination
of anti-personnel mines offers a clear challenge for the United States, but the hemispheric
security regime did not overcome U.S. reluctance that required the collective weight of
virtually the entire world.
At any rate, it is the great, uncharted middle between nuclear weapons and
landmines that deserves attention. Why have no agreements been possible here, given the
repeated exhortation to maintain reasonable balances? This could be because of a
relative gains problem between producer states and consumer states. Since some of the
largest countries export arms Argentina, Brazil, Chile and the United States they may
have been successful in restraining rule making in this area. But there is no evidence that
this was ever an issue.


146
record, considering the fairly recent nature of OAS expansion and the natural limits to
expansion by the Board. 14
Second, the Board approved a dramatic change in its mission statement an
explicit acknowledgement of its political subordination to the OAS. Resolution 1240
systematized the ways in which the OAS and its member states could request "advice and
consultancy services" from the Board and stipulated that such assistance not be of an
"operational" nature. 15 But it did not restrict the Board's activities directly. It was the
Board's 1995 mission revision that, by adopting Resolution 1240 as its terms of reference,
limited the Boards scope of action. 16 This step was strengthened further by the 1998
revision of Board regulations, which added a new Article 2 restricting the scope of the
Board's mission to resolutions of the General Assembly and the Meetings of
14 At this point, all countries with fully consisted defense forces are members of
the Board, with the exception of Canada, Belize and Jamaica. Canada sends students to
the College and has periodically considered joining the Board. Indications now are that
Canada will await the results of the current review of hemispheric security and
institutions.
15 Demining is a possible exception. Supervision is conducted under an
operating headquarters in Central America. But this function enjoys broad legitimacy
and has not been seriously challenged.
16 The current mission statement, approved in 1995, reads: "The IADB advises
the General Assembly, the Meeting of Consultation of Ministers of Foreign Affairs, the
Permanent Council of the OAS by means of its proposal and works in matters of a
military natures; acts as an organ of planning and preparation for the defense of the
American Continent; and performs advisory functions within its competence with a view
towards contributing to the maintenance of peace and security of the Continent." This
mission statement breaks significantly with previous versions in only one important
measure for the first time, the Board's mission statement links it directly to the OAS
instead of directly to its member states.


272
coordinated military operations to fight trafficking in contraband, specifically addressing
arms trafficking, and to exchange information on troop deployments (Child, 1996).
Security cooperation is also evident outside of the armed forces, especially in the
criminal justice systems of the subregion. Multinational rules have been adopted
directing the exchange of information on armed bands, the register of firearms, stronger
border controls, stiff prison sentences for arms traffickers and the adoption of an arms
importation registry (Guatemala 1997 and Honduras 1997 CSBM Submission).
Furthermore, El Salvador, with financial support from Japan, is establishing a training
institute for the police forces of Central America.
In conclusion, the subregion shows impressive evidence of institution building
and rule making around a new concept of security: "democratic security." There is little
evidence of balancing along the lines predicted by realism. Rules to support democratic
security are binding in the areas of transparency/verification and defense postures. There
are fewer rules in the area of burden sharing, but institutions are present that are charged
with the responsibility of developing cooperative behavior in this area. Let us now turn
to actual patterns of security behavior to see if they are rule-based.
T ransparency/V eriflcation
As noted in Chapter 4, the countries of Central America have been relatively less
active in complying with transparency provisions and reporting contacts than in the
Southern Cone or the Andean Region. The exception is El Salvador, which has published
a defense white paper, distributed numerous policy statements at hemispheric
conferences, disseminated its annual defense report including information on distribution
of personnel and major activities (El Salvador 1998 CSBM Submission), and rewritten its


243
transparency measures is to avoid accidental war by clearly demonstrating a viable
deterrent threat (Chile, 1998).
Brazil's National Defense Policy shows a deeper commitment to subregional
cooperation, but still prioritizes a commitment to national defense (Brazil, 1996). The
main goal of subregional cooperation appears to be the maintenance of a "ring of peace,"
based on cooperative relations with its neighbors. Brazil, at the center of the ring, can
thus project its influence in pursuit of hemispheric and global objectives (Brazil, 1996,
Article 2.10). At present, arrangements with neighboring states, while deeply
institutionalized, are not expressed by formal security treaties. Rather than formal
alliances, Brazil has sought to develop a fluid network of contacts that creates confidence
and channels of communication that facilitate coordination against common threats.
It is important to note, however, that the Brazilian defense plan recognizes that the
prospects of subregional conflict are much diminished; Brazilian policy no longer
defines Argentina as a threat (Escud and Fontana, 1998). Consequently, ancillary roles
for the armed forces are explicitly recognized, including a major change which cites a
military role in support in the fight against illicit trafficking in narcotics (International
Institute for Strategic Studies, 1997/8).
Argentine doctrine has only recently been reformulated within the government
and publication of its defense white book was delayed until 1999. The defense
establishment has been consumed by the ongoing examination of civil-military relations,
which led to significant restructuring and the adoption of a volunteer force (Bossi, 1998).
Earlier national defense plans established the mission of the armed forces as one of
territorial defense against external aggression. However, the recently approved 1998


166
Nicaragua, and Argentina wanted to include a call to the IADB to make contributions to
the effort. The refusal of Mexico to accept such a statement led the Venezuelan
delegation to suggest a compromise, which was ultimately accepted, to omit mention of
the Board and limit the request to the Inter-American Defense College.
In the course of the debate, the Mexican delegation reaffirmed its position that the
Inter-American Defense Board not be considered part of the hemispheric security system.
Mexico termed the inclusion of a reference to the College, after all a subordinate
organization of the Board, a "constructive ambiguity" which allowed member countries to
read what they wanted into the phrase and thereby support the declaration by consensus.
The Second Regional Conference on CSBMs took place in 1998 in San Salvador.
The declaration from that conference also produced a list of CSBM categories. Table 4-2
lists those measures. Most of the measures deepened CSBMs already in place, adding
rigor and specificity. There are some measures that are basically new, although they do
not introduce new themes into the regime; involvement of legislators in CSBMs and
cooperation to deal with problems of sea transport of nuclear waste are two examples.
Other items of interest in the San Salvador Declaration include recognition of the
value of the conferences of the armed forces and the meetings of the defense ministers
(thus formally linking these sets of institutions with the OAS), a recommendation that the
Hemispheric Security Committee complete a program on Education for Peace, and a
request that the Santiago Presidential Summit task the OAS to hold a meeting on
hemispheric security and the architecture of the IASS (which it did, as we saw earlier).
The 1998 CSBM resolution reiterated the call for member states to adopt
measures in both the Santiago and San Salvador Declarations and continued to add


359
Snidal, Duncan, Robert Powell, and Joseph Grieco. 1993. The Relative Gains Problem
for International Cooperation. American Political Science Quarterly 87 (3): 729-
742.
Snyder, Jack. 1996. "Military Force and Regional Order." In Edward A. Kolodziej and
William Zartman, ed., Coping with Conflict after the Cold War. Baltimore, MD:
Johns Hopkins University Press, 291-308.
Solingen, Eric. 1997. Economic Liberalization, Political Coalitions, and Emerging
Regional Orders. In David A. Lake and Patrick M. Morgan, eds., Regional Orders:
Building Security in a New World. University Park, PA: Pennsylvania State
University Press, 68-100.
Spector, Leonard S. and Jonathan Dean. 1994. Cooperative Security: Assessing the
Tools of the Trade. In Janne E. Nolan, ed, Global Engagement: Cooperation and
Security in the 21st Century. Washington, DC: Brookings Institution Press, 131-
174.
Spracher, William C. and Daniel O. Mason. 1994. "The Inter-American Defense Board."
Revista del Colegio Interamericano de Defensa 20: Washington, DC: Inter-
American Defense College.
Stanely, William. 1999. Rappateur's Report on Subregional Cooperation." In Donald
E. Schulz, ed., Conference Report: The Role of the Armed Forces in the Americas:
Civil-Military Relations for the 21st Century. Carlisle, PA: Strategic Studies
Institute, 141-150.
Strange, Susan. 1983. Cave! Hie Dragones: A Critique of Regime Analysis. In
Stephen D. Krasner, ed., International Regimes. Ithaca, NY: Cornell University
Press, 337-354.
Starr, Harvey. 1997. Democracy and Integration: Why Democracies Dont Fight Each
Other. Journal of Peace Research 34(2): 153-162.
Stein, Arthur A. 1983. Coordination and Collaboration: Regimes in an Anarchic
World. In Stephen D. Krasner, ed., International Regimes. Ithaca, NY: Cornell
University Press, 115-140.
Stein, Arthur A. 1990. Why Nations Cooperate: Circumstance and Choice in
International Relations. Ithaca, NY: Cornell University Press.
Stepan, Alfred. 1978. The State and Society: Peru in Comparative Perspective.
Princeton, NJ: Princeton University Press.


258
That said, there has been little reduction in defense postures, despite a high level
of transparency developed through participation in the hemispheric institutions relating to
transparency, the security institutions of the subregion and the dramatic increases in
economic and governmental interrelations. But it is unlikely that the countries of
Argentina, Brazil and Chile fear that a reduction in defense postures would put their
security at risk. A more plausible explanation is that these countries see for themselves
an important role outside of their subregion and indeed outside of the region.
Transparency has created a level of confidence that allows these countries to project
power sometimes in concert with one another well beyond their once conflicted
borders. This may point up one of the limitations of the subregional approach we have
adopted here.
Central America
"Democratic Security"
Introduction
In December 1996, Central American closed a long chapter of instability, conflict
and war with the signing of a peace agreement between the government of Guatemala and
the Unidad Nacional Revolucionaria Guatemalteca (URNG). The end of civil war in
Guatemala followed peace agreements in Nicaragua in 1990 and El Salvador in 1992.
These civil wars were linked to interstate competition in a variety of ways. Central
American states provides safe havens and other assistance to guerrilla movements
fighting in neighboring states, while external patrons, principally but not exclusively the
United States and Cuba, overlaid their own competition on these internal conflicts.
The coincidence of these conflicts and their resolution demonstrates what was
obvious early on in the nation-building phase of Central American states this subregion


229
delved deeply into the prospects of cooperation across a wide variety of areas (Argentina
1996 CSBM Submission). Contacts of this nature will be addressed in the following
pages in more detail.
More importantly, perhaps, is the emergence of a multilateral forum that
developed from this network the Southern Cone Strategy Symposium, which has been
ongoing since 1987 between Argentina, Brazil, Chile, Paraguay, and Uruguay (Argentina
1996 CSBM Submission). With the emergence of Mercosur, the symposium adopted a
new namethe Mercosur Joint Staffs and Defense Strategic Studies Symposium.
Bolivia and Ecuador joined the group as observers in 1998, for full integration in 1999
(Creation of regional defense, 1998).
This annual meeting, which derives from meetings between the military
commands of Argentina and Brazil following the 1986 Declaration of Iguazu, has
increasingly served as a forum to discuss the prospects for military integration in the
region. For instance, the 10th Symposium, held in Montevideo in 1997, featured analysis
on Mercosur and "subcontinental security" (Military leaders, 1997). The 11th meeting,
held in 1998 in Buenos Aires, focused on the possibility of creating a South American
military bloc. Delegates from each country exchanged views on a proposal to establish a
permanent mechanism for the planning and coordination of security and defense issues.
The oldest multilateral security institution in the region is CAMAS (Control of the
Maritime Area in the Southern Atlantic), a regional organization which remotes from the
1960s and is charged with the responsibility to maintain communications between its


CHAPTER 2
THEORY
Introduction
Following a brief introduction, this chapter develops a theory-based description of
the components of a cooperative security regime, made up of institutions that facilitate
cooperation by providing organizations for negotiation and rules that prescribe specific
behaviors. 1 Once that important mark is established, the chapter investigates the realist
and institutionalist traditions in some depth, particularly with regard to the obstacles and
opportunities these paradigms posit for security cooperation between states. From that
examination, contending hypotheses emerge to explain the patterns of security
cooperation present in the hemisphere. Finally, the chapter concludes with a discussion
of the operationalization of key explanatory factors.
Realism and neoliberal institutionalism both address the problem of security
cooperation and conflict in the international system. The theoretical frameworks share
many of the same assumptions of how the world works. Both assume that states are the
principal actors in an anarchic global system. In the absence of world government,
states must rely primarily on their own resources to ensure survival. Since the main
threat to survival is aggression from other states, international relations, especially in the
^ Mere membership in international organizations does not constitute
participation in a regime because independent decision-making is in no way constrained
(Stein, 1990).
30


167
specificity to mechanisms for reporting. The calendar for submissions of declared data,
first established by the General Assembly in 1996, has now grown crowded, but wherever
possible submission dates have been standardized with other reporting requirements that
countries may have, both at the OAS and the UN.


290
resources shows that underlying rivalries exist between Honduras, El Salvador and
Nicaragua. Salvadoran dynamism shows up throughout the subregion particularly in
the areas of transparency and burden sharing activities where it is a subregional leader in
many respects. El Salvador's excellent relations with the United States may be a cause
for envy. However, despite these factors the trend is firmly in the direction of more
institutions and more binding rules.
Actual security behavior is less solidly supportive of institutionalism.
Transparency and verification are incomplete. There are many mechanisms that have yet
to be implemented. But much was accomplished under third parties and the trend since
those elements have departed is still positive. Downsizing of defense postures is ongoing
and directly related to institutions. Countries of the subregion have been willing to
realize absolute gains implied in defense reductions even in the absence of reciprocity.
Relative gains were not addressed by institutions, but do not appear to have presented a
serious obstacle. Burden sharing is less common among militaries, but does occur. It is
difficult to find much balancing behavior in the pattern of security cooperation.
Andean Region
"Insecurity Dilemma"
Introduction
Of the three case studies considered in this work, this may be the least developed
in terms of its identity as a separate subregional security order. At the northern extreme,
Venezuela is pulled towards the Caribbean, where it competes with Mexico and the
United States for influence (Griffith, 1998; Kelly, 1997). At its southern extreme, Chile
is more closely involved in Southern Cone security issues, at times pulling Peru, a


355
Organization of American States. 1994a. Central America's Remarks at the Preparatory
Meeting for the Meeting of Expert on Confidence -Building Measures and Secuirty
Mechanisms in the Region. (SEGRE/doc.l 1/94). Washington, DC: Organization of
American States.
Organization of American States. 1994b. Report of the Permanent Council on
Implementation of Resolution AG/RES. 1181 (XXlI-O/92) AND 1240 (XXXIIl-O/93):
Inter-American Defense Board (AG/doc.3095/94). Washington, DC: Organization of
American States.
Organization of American States. 1995a. A New Vision of the OAS. Washington, DC:
Organization of American States.
Organization of American States. 1995b. Report of the Chairman of the Special
Committee on Hemispheric Security Concerning Cooperation for Hemispheric
Security. (CP/doc2609/95 rev. 1). Washington, DC: Organization of American
States.
Oye, Kenneth A. 1985. Explaining Cooperation under Anarchy: Hypotheses and
Strategies. World Politics 38: 1-24.
Oye, Kenneth A., ed. 1986. Cooperation under Anarchy. Princeton: Princeton
University Press.
Pala, Antonio L. 1998. "Peacekeeping and its Effects on Civil-Military Relations." In
Jorge I. Dominguez, ed., International Security and Democracy: Latin America and
the Caribbean in the Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh
Press, 130-150.
Pamplona, Francisco Stuart. 1996. Lecciones Aprendida en la MOMEP. Military
Review 75 (5): 64-66.
Patio Mayer, Hernn. 1996. The Future of Cooperative Hemispheric Security in the
Americas. In Richard L. Millett and Michael Gold-Bliss, eds., Beyond
Praetorianism: The Latin American Military in Transition. Miami: University of
Miami North-South Center, 1-10.
Peirce, Holly and Tom Kelly. 1996. Rapporteurs' Report of a Conference on Security
Cooperation in the Western Hemisphere: Lessons from the 1995 Ecuador-Peru
Conflict. December 4-6, 1996. Coral Gables, FL: University of Miami North-South
Center.
Pentland, Charles. 1973. International Theory and European Integration. London:
Faber and Faber.


83
particular security issue is not high on the U.S. agenda, then resources are not likely to be
available for influence attempts.
With the end of a threat of nuclear proliferation in the Southern Cone, the main
U.S. strategic concerns in Latin America are threats posed by illicit trafficking in drugs
and general social instability in nearby regions (causing illegal immigration or providing
a haven for terrorist/guerrilla groups). These threats intensify as they become more
proximate.
Aside from the sharpened distinctions between contending hypotheses, moving to
the subregional level offers a key methodological advantage. It creates more case studies
and allows use of the comparative method. The comparative method allows us to gain
analytical traction and arrive at a more robust case for our contending paradigms. I plan
to deal with the small-A problem by focusing on the comparable cases cases that are
similar enough to allow for good comparison, but that differ in important ways that affect
the relative gains problem. 43
The subregional cases considered in this study the Southern Cone, Central
America and the Andean Ridge certainly do not exhaust the possibilities for scholarly
research. There are other regions that merit examination: the Rio de la Plata subsystem;
the Amazon region, the Caribbean, North America; the Pacific Rim countries, etc.
43 Colliers (1991) update of Lipjharts (1971) influential essay on the
comparative method is useful. On the "small N, many variables" problem, the literature
offers two approaches: (1) Most Different Cases: gaining analytical traction by selecting
disparate cases to isolate explanatory variables and (2) Most Similar Cases: gaining
analytical traction by selecting similar cases so that other variables are "canceled" out,
leaving the explanatory variable to explain differing outcomes.


67
characterized as rule-based). But the important point is that while institutions may reflect
the distribution of power and the interests of their members, they have an independent
impact on behavior and mute the impact of power over outcomes (Keohane and Nye,
1989). Cooperation between states, acting as unitary, rational actors attempting to
maximize interest in an anarchic, self-help system, is possible: (1) because states are
interested in attaining benefits of reduced tension and enhanced security cooperation and
(2) because institutions help them overcome the dilemmas of defection free riding and
cheating and help them negotiate solutions to the distributive problems that relative
gains present.
How might this work in situations of asymmetry? The kind of cooperation
described under realism in which a large power establishes institutions as tools to achieve
its own ends is a possibility admitted by neoliberalism. In fact, where mutuality of
interests is high, it may be difficult to find significant differences in the manner in which
these kinds of institutions evolve. But the politics of security cooperation under
asymmetry are likely to be more complex. Keohane and Nye (1989) provide a set of
potential outcomes that differ sharply from realism. While these were constructed for a
specific context what the authors called "complex interdependence" -1 argue that the
effects they predict are evident in cases where a large state practices a policy of
reassurance (as described by Mastanduno) that opens negotiating room for smaller states.
The United States may opt for a multilateral approach as a cost-effective solution
that simultaneously reassures its security partners and engenders the cooperation it seeks.
However, such a strategy is likely to produce perverse outcomes over time. This is
because the institutional forum it promotes restores symmetry by offering smaller states


155
and cooperation have taken root. Countries in these regions made joint presentations -
the Guatemalan representative covered security institution building in Central America;
the Venezuelan representative presented on the growing border cooperation between
Colombia and Venezuela; and the Chilean representative made a similar presentation on
Chilean-Argentine cooperation. The fact that these were joint presentations and that they
offered concrete examples of security cooperation is significant. This offered a hint of
the degree to which security cooperation was now present in the hemisphere, as well as
an indication that cooperation clumps in specific geographic subregions. These trends
will be further developed when we discuss regime effectiveness.
In spite of these important presentations, the DMA made only incremental
progress towards advancing burden sharing in the hemisphere, a theme that was the
logical next step for the DMA process, given the advances at Williamsburg and Bariloche
in transparency and civil-military relations. The United States made a presentation on
terrorism and Peru presented on narcotrafficking and both of these threats were included
in the final declaration, along with the problem of small arms, natural disasters and
landmines. However, the declaration and working group papers endorse no concrete
steps to enhance security. Ironically, given the setting of the DMA, mention of the threat
of drugs was absent in the final draft of the declaration considered at the closing session.
The Colombian defense ministers, presiding over the proceeding, added a general
reference to the declaration at the closing bell.
U.S. support for the Cartagena DMA was considerable and included heavy
involvement in preparation of the agenda. The U.S. delegation and support team in
Cartagena, numbering over one hundred, dwarfed representation from other countries.


255
United States (Argentina 1996 CSBM Submission). In a recent report, sources indicate
that the United States will begin sharing the exercise planning process with Argentina. In
addition, the United States is providing support to the Argentine peacekeeping school
("Confidentiality Accord," 1998). This evidence of U.S.-Argentine axis is consistent
with the findings under transparency/verification.
What reaction has this elicited from Brazil? Since the late 1970s, Brazil has
consistently opposed most initiatives that include the United States in some sort of
operational security arrangement. Unfounded rumors that the United States would
support the formation of a Mercosur army elicited a negative response from Brazil in the
media ("Argentine General," 1998). In the broader, hemispheric context, we have
reviewed Brazil's opposition to involvement by the Board in peacekeeping operations,
precisely because of the presence of the United States in that organization. The Brazilian
army rejected the idea of a formal, operational arrangement with countries in the
hemisphere due to problems associated with the "asymmetry of power and disequilibrium
in the American continent" (Rizzo, 1997, p. 14), a clear reference to the United States.
Furthermore, the Brazilian military rejects what it perceives as an attempt on the part of
the United States to replace anti-communism with the threat of narcotics as the new
organizing principle for hemispheric defense (Rizzo, 1997). Put in simple terms, Brazil
wants the United States out of Southern Cone security affairs (Escud and Fontana,
1998).
Analysis
There is clear evidence of security cooperation to share the burdens of regional
security. While no NATO-like structure has yet emerged, there is considerable


91
actual security postures and their effectiveness in addressing common security risks.
Such judgments can rarely be wholly objective.
Even where data is available, rigorous quantitative analysis may not be possible.
In the case of the CSBM inventories, for example, while geographical and thematic
patterns are readily discemable, quantitative comparisons between states may not be
possible due to problems with the data as it has been compiled to date. The 1995
inventory of CSBMs, for instance, prepared by the IADB, suffers from a series of
limitations that preclude any accurate quantitative analysis. In the first place, little more
than half of the members of the IADB participated. Second, those that participated did
not follow any standard format for categorization of CSBMs. Third, CSBMs were not
standardized in terms of size or duration, leading to crippling aggregation problems.
Fourth, the data reported was not verified, and bilateral events reported by one country
were often not reflected in the partner countrys submission. Due to lack of
standardization, countries made their own decisions about what measures to report.
These issues are slowly being address in the reporting formats, but the data submitted
since 1995 do not reflect a marked improvement.
Despite these methodological problems, the cluster of indicators for transparency,
burden sharing and defense postures combine to give weight to the analysis of security
cooperation. A picture of security cooperation emerges which can evaluated against
power and interests, on the one hand, and institutional rules on the other.


168
Table 4-2: CSBMs in the Declaration of San Salvador
a) Encourage contact and cooperation among legislators on confidence building measures
and on matters of peace and hemispheric security, including conferences, the exchange
of visits, and a meeting of parliamentarians, in order to strengthen its process;
b) Extend to diplomatic training institutes, military academies, research centers, and
universities the seminars, courses, and studies envisioned in the Declarations of Santiago
and San Salvador on confidence and security building measures, disarmament, and other
issues related to peace and hemispheric security, with participation in those activities by
government, civilian and military officials and by civil society;
c) Identify and carry out activities promoting cooperation among neighboring countries
along their border regions;
d) Promote the exchange of information, inter alia, through the publication of books on
defense or official documents, as appropriate, permitting greater transparency with
respect to the defense policies of each country, and on the organization, structure, size,
and composition of the armed forces;
e) In order to promote transparency, and with technical support from the appropriate
international economic agencies, encourage the carrying out of studies for establishing a
common methodology in order to facilitate the comparison of military expenditures in
the region, taking into account, inter alia, the United Nations Standardized International
Reporting of Military Expenditures;
f) Develop a cooperation program to address the concerns raised by maritime transport of
nuclear and other waste, and to cooperate and coordinate in the relevant international
fora to strengthen standards governing such transport and its safety;
g) Continue supporting the efforts of the small island states to address their special security
concerns, including those of an economic, financial and environmental nature, taking
into consideration their vulnerability and level of development;
h) Improve and broaden the information submitted by the member states to the United
Nations Register of Conventional Arms, so as to enhance the hemispheres contribution
to pursuing the aims of that register, in compliance with the relevant resolutions of the
UN General Assembly; and
i) Continue consultations and the exchange of ideas within the Hemisphere to advance the
limitation and control of conventional weapons in the region.
It is important to note what was left out of the San Salvador Declaration. Early
drafts of the Declaration included calls to evaluate the effectiveness of CSBMs in place;


147
Consultation. For all intents and purposes, the Board is now under the "civilian" control
of the OAS.17
Third, the Board has adopted a post Cold War security agenda. This focus
emerged naturally and not as a result of the OAS reexamination of hemispheric security.
The long anticipated end of the Cold War sparked Board studies dating back to the 1980s
on subjects as disparate as terrorism, drug trafficking, peacekeeping and corruption. But
the formal acknowledgement of the end of the Board's Cold War mission came in 1992.
That year's strategic estimate rejected as implausible any external threat to the hemisphere
and instead focused on new, "transnational" threats. In early 1993, in light of the absence
of a clear external threat, the standing defense plans of the Board were abrogated by the
Council of Delegates. In the revised mission statement, the phrase collective defense
has been dropped.
The Council's focus then shifted to "non-traditional threats" ranging from
terrorism, to drug trafficking, to natural disasters topics that had already been under
study in the Staff for some time. But formal inclusion of non-traditional threats on the
Boards agenda came only after much dissension, primarily but not exclusively centered
on the Mexican delegation, and was later removed. The United States, in an effort to
move forward collectively to address these areas, initiated a broad review of the Boards
mission and activities. In the context of the review, the phrase non-traditional threat
17 There are those who would trap the Board in a paradox any move by the
Board to subordinate itself to the OAS is in fact evidence of autonomy. Theoretically, the
Board could reverse field by changing its mission and its regulations again but such
action would require a two-thirds approval of all member states, who would certainly be
acting on instructions from their governments on these important issues.


22
revitalized subregional trading blocks: Mercosur, the Andean Pact, the Caribbean
Common Market (CARICOM) and the Central American Common Market (CACM).
In the area of security, subregionalization was driven by a number of related
factors. In the first place, security issues were thought to be less salient with the end of
the Cold War. The unifying theme of East-West confrontation had disappeared, given
way to a new focus on economic relations. In particular, Latin American security issues
were thought to be of marginal importance, given the lack of U.S. strategic concerns in
the region. This new "marginalization" opened space for subregional initiatitives
(Druetta, Tibiletti, and Donado, 1991).
In the second place, the "dark side" of globalism, with the emergence of
transnational threats, complicated the regional security agenda significantly, leading to a
natural breakout of specific subregions based on particular security challenges (Stanely,
1999). Hence, while the Southern Cone focused on overcoming a rather standard security
dilemma in order to move economic integration forward, Central America and the
Caribbean were confronted with the transnational challenges of contraband and
criminality. In the Andean Ridge, countries faced a witch's brew of tense interstate
relations and deeply rooted transnational security threats.
Transnational threats are often considered likely triggers for cooperative security,
since solutions require multilateral approaches. But U.S. attempts to include
transnational threats on the hemispheric security agenda have been met with resistance.
Latin American countries are nervous about a broadly defined security agenda that
includes many threats of a non-military nature and are unwilling to allow the United


245
In fact, the three major countries are all undergoing arms modernization programs
of various degrees at this time. General trends as reported in World Military
Expenditures, other independent sources, as well as declared date in the UN registries
bears this out. While Argentina did undergo a major downsizing during the 1980s
("Economic Problems," 1997), other countries in the region did not follow suit. The
trend since the mid-1990s has been solidly towards new procurements to support defense
modernization plans ("Senate Approves," 1998; and "Economic Problems," 1997).
Nevertheless, there are signs that this modernization will not escalate into a major
arms race. The United States recently lifted a ban on access to high-technology arms -
specifically advanced fighters, removing what little supplier restraint had existed on these
acquisitions. But no sales of advanced aircraft have been consummated since the end of
the ban. Chile, which had indicated its interest in procuring new fighters, put its decision
on hold, choosing instead to place priority on other needs7
J
Furthermore, officials in both Argentina and Brazil have supported each other's
modernization programs. For example, the Argentine Chairman of the Joint Chiefs of
Staff, LTG Zabala, also claimed that each country was buying material that could be used
for the "regional defense of the Southern Cone" ("Argentine General," 1998).8 Brazil's
willingness to tolerate increased defense expenditures and arms procurement from
7 See Klare (1994) for the details on U.S. sales of high-tech armaments after
President Bush lifted the embargo on Chile (1990) and resumed sales to Argentina
(1992). U.S. arms sales have risen at the same time that a new security system in the
Southern Cone has taken hold.
^ Selcher (1992) reports that Brazil and Argentina signed a coproduction
agreement for aircraft, although the initiative never matured.


198
activity taking place in the region, as reported by participating countries. The data was
reported using formats developed in OAS Meeting of Experts in 1995 and the CSBMs
categories established in the Santiago Declaration. Note that only partial data is available
on the new measures established by the San Salvador Declaration.
The measures reported include a description of the event, its duration, its
geographic scope, and the branch of service and partner countries involved. Some
submissions contain additional information as well, such as initiation and completion
dates. Some measures are well known, easily verified and researched. Others are more
mysterious. Taken together, the inventories contain close to 2000 individual measures.
They range from the activities of permanent standing bodies, such as the Regional
Security System in the Caribbean, to exchanges of cadets between the military academies
of neighboring countries. Each measure is intended to contribute in some way to the
promotion of confidence and security.
As laid out in Table 4-4, each of the categories used by the reports corresponds to
one of the three sectors studied under the regime. The method employed here is to look at
overall data patterns by variation over time, over type of measure and sector reported and
over geographic area. The data does not support more detailed statistical analysis for
many reasons. First, it is dangerous to aggregate the data within categories, since there is
no standard threshold for reporting. For instance, the United States reported all of its
academic exchanges in one statement, while smaller countries reported each exchange of
individual officers. Second, it is difficult to compare the data across measures and
sectors, since the events that belong to each category are so different. Finally, the
comprehensiveness of the reporting, voluntary in any case, varies widely by country.


169
add new verification measures, such as demonstrations of newly acquired weapons and a
program of visits to military bases; develop a detailed plan of action for implementation
of the CSBMs in force; and address more forms of burden sharing, including
peacekeeping. All of these innovations, drafted by the Salvadorans with input from other
delegations, were ultimately rejected. Since many of the items rejected aimed to
strengthen reporting and verification procedures, it is fair to conclude that there are limits
to the degree of intrusiveness that states will tolerate from an international regime -
especially one that includes the United States with its well-publicized (and dramatized)
verification technologies.
Other hemispheric organizations have echoed the call for more transparency, as
was noted in the review of the Presidential Summits and the DMAs. The overlapping and
mutually reinforcing nature of the organizations helps strengthen the system. A good
example of how this procedure develops can be seen by comparing the Declarations of
San Salvador and Santiago, point by point. As Table 4-3 shows, many of the San
Salvador measures deepen or further develop those of Santiago.


110
aggression. Aggression is defined very broadly in the text and can be interpreted to
include internal threats. At the Caracas conference (1954), the United States would push
through a resolution over Mexican and Argentina objections which would include
communist influence in government within the definition of aggression (Council on
Foreign Relations, 1962).
In the view of others, the Rio Treaty was bom flawed, since its call for mutual
assistance is not binding. Neither does the treaty include a mechanism for the operation
of collective forces, although the Advisory Defense Committee (part of the OAS
structure) it created could presumably help manage such forces. Since the Committee has
never met, that function has never been exercised. Hartlyn, et al. (1992), claim that
through these omissions the Rio Treaty actually enshrines principle of non-intervention
and is an example how weaker nations can use international institutions to their
advantage (Hartlyn, et al., 1992, p. 11).
Ironically, the Treaty, intended as a pact to protect the hemisphere from external
aggression, has only been invoked to deal with crises between American states. Treaty
mechanisms have been used to address numerous conflictive situations, but only in the
Caribbean and Central America and never for a major conflict. The Treaty was
informally abandoned after Malvinas, when it could not invoke a collective hemispheric
response in the face of what many states in the region viewed as external aggression
(Vaky, 1993).


221
major powers might achieve a position of hegemonic influence over one of the buffer
states Uruguay, Paraguay and Bolivia.
Changes in military technology that favor offensive systems should exacerbate
security concerns. The Malvinas experience made the advantages of modem technology
and superior organization abundantly clear. We have already seen evidence that the
outcome of that war led Brazil to accelerate its arms modernization, in the event it was
challenged by a world-class military. Its indigenous arms industry developed aircraft,
battle tanks and sophisticated missile systems.1 Chile, which had backed Great Britain in
the conflict, also increased arms expenditures.
In contrast, Argentina entered a period of prolonged decline in terms of relative
military strength after the Falklands/Malvinas war. Under President Raul Alfonsin, levels
were lowered only to pre-Malvinas levels, but President Carlos Menem introduced drastic
reductions in force, which continued through the early 1990s and culminated with the
elimination of compulsory service in 1994. According to Escud and Fontana (1998),
Argentina has gone "beyond the point of no return" (pp. 63-65). Brazil and Chile reacted
to this trend by maintaining relatively flat arms inventories through the early 1990s.
It is in this context that the dramatic about face in Argentine foreign policy should
be understood. Argentina reached out to its principal competitors Brazil and Chile to
clear the agenda of potential security problems that its military might no longer be
capable of addressing. For instance, Argentina and Brazil put to rest competition over the
1 See Brigagao (1985), Dagnin (1985) and Franko-Jones (1998) for detailed
studies of the Brazilian arms industry.


6
(transparency requirements, cooperative arrangements and arms control).-^ At the heart of
the IASS lie two permanent organizations: the Organization of American States (founded
in 1948) and the Inter-American Defense Board (founded in 1942). The OAS is the
principal political organ of the system, with general oversight of conflict mediation and
collective security activities. The IADB for its part is charged with the military-technical
aspects of hemispheric security, to include defense planning, training, and
standardization. The organizational structure of the IASS is presented in Figure 1-1, for
reference at the end of this chapter.
The tension between these two organizations over the past half-century reflects
uncertainty about the appropriate role of the military in hemispheric security institutions,
the debate between broad and narrow definitions of security and the larger issue of civil-
military relations in the hemisphere. These themes, which we will revisit later in the
study, combine with the asymmetrical distribution of power in the IASS to make
organization of hemispheric defense particularly problematic. Efforts to organize
hemispheric security are almost inevitably associated with the United States, sparking
concern among Latin American states that a centralized system will front for the U.S.
security agenda.
3 Grossi (1997, pp. 242-3) sees the IASS as including the "interlinked treaties,
agreements and institutions with specific responsibilities on security," including specific
instruments like the Rio Treaty and technical bodies such as the Inter American Defense
Board. He characterizes the IASS as a "collective security arrangement" with a degree of
obligation below that of a formal alliance. Also, see Council on Foreign Relations
(1962).


121
including Argentina, which had contributed two ships to the blockade in the early 1960s
(Varas, 1985, 53). Two years after the Dominican intervention, at the 1967 OAS
conference in Buenos Aires, fears that the establishment of a permanent force was
seriously being considered led to the defeat of a U.S.-backed proposal to move the Board
under the OAS Charter, which would have rationalized and centralized a system that until
then had been piecemeal.28
Writing shortly after these events, Jose (1970) saw the eventual creation of a
permanent peace force as unlikely for three reasons: (1) a permanent, continuing
military presence would offend the non-military nature of the regional system and inter-
American relationships; (2) any effective instrument for collection action could violate
the principle of non-intervention, the cornerstone of the Inter-American System; and (3)
animosities, suspicions and jealousy between American States (especially but not
exclusively between the United States and the Latin American countries) hamper
concerted collective action in the security field (41-47). His observations can be applied
to any permanent military organization of am operational nature.
The historical truth is that the hemispheric security system has resisted every
previous attempt to make it more rational, systematic and ordered.29 Attempts to
28 In a diplomatic twist, the proposal was actually presented by Argentina, while
the U.S. watched to see whether it would be accepted. When it was clear that opposition
was strong, the U.S. withdrew its backing.
29 The logic of efficiency is as ineffective as it is inescapable. Consider the
Board's 1966 study, which accompanied the doomed resolution urging member countries
to place the Board under the OAS Charter during the 1967 Buenos Aires Conference, pp.
10-11: "At this juncture, we can point out that the Inter-American System has three


162
Table 4-1: CSBMs in the Declaration of Santiago
a) Gradual adoption of agreements for the advance notice of military exercises.
b) Exchange of information and participation of all member states in the United Nations
Register of Conventional Arms and the Standardized Instrument for Reporting of
Military Expenditures.
c) Promotion of the development and exchange of information concerning defense policies
and doctrines.
d) Consideration of a consultation process with a view to proceeding towards limitation and
control of conventional weapons.
e) Agreements on invitation of observers to military exercises, visits to military
installations, arrangements for observing routine operations and exchange of civilian and
military personnel for regular and advanced training.
f) Meetings and activities to prevent incidents and increase security for transport by land,
sea, and air.
g) Cooperation programs in the event of natural disasters or to prevent such disasters, based
on the request and authorization of the affected states.
h) Development and establishment of communications among civilian or military
authorities of neighboring countries in accordance with their border situation.
i) Holding of seminars and courses, and studies on mutual confidence and security building
measures and policies to promote confidence involving the participation of civilians and
military personnel, and on the special security concerns of small island states.
j) High-level meetings on the special security concerns of small island states.
k) Education programs for peace.
Transparency measures include advance notification on military exercises, which
aims to reduce uncertainty by alerting neighbors to maneuvers near border regions. But
no thresholds or specifications are established for size, duration, location or type of
exercise to be reported. No mechanism is identified for this purpose at the hemispheric
level, although the Board could serve as an appropriate venue for exercise


222
resources of the Paran river network with a 1979 agreement (Hirst, 1998). The two
countries also ended their dangerous nuclear competition.2 And Argentina and Chile
resolved some two dozen border disputes, beginning in 1984 with the signature of the
Peace and Friendship Treaty (Dominguez, 1998; Hirst, 1998).
Argentina also launched a diplomatic offensive with Great Britain, signing the
Declaration of Madrid in 1989, which created a working group to promote confidence
and security and developed a system of reciprocal information and consultation. One of
the measures adopted requires advance notification of air and naval maneuvers near the
disputed areas with more than 1,000 troops or 20 aircraft sorties (Cuevas and Nakata,
1997). Combined search and rescue exercises have been conducted. These measures,
which accompanied a steady improvement in relations between the two countries, may
well have established an important model for the region in the area of CSBMs.
Other developments helped push back the prospect of conflict in this subregion.
By 1990, with the exception of Bolivia and Chile, most border disputes had disappeared
as potential triggers for conflict. The Rio De La Plata Treaty (1973), the Amazon Treaty
(1978), the Treaty of Peace and Friendship (1984), and the renewal of the Antarctica
Treaty helped stabilize borders that still may have been unsettled. The 1986 UN
2 There has been a steady process of dtente between Brazil and Argentina in the
area of nuclear weapons and delivery systems. The 1985 Joint Declaration on Nuclear
Policy pledged both countries to use nuclear technologies for peaceful purposes only.
This declaration was followed in 1990 by the Foz de Iguazu Declaration, which
established bilateral inspection mechanisms under the auspices of the ABACC. The next
year, in 1991, Argentina and Brazil entered into a trilateral agreement with the IAEA.
Also that year, Chile joined Argentina and Brazil in the Joint Declaration for the Total
Prohibition of Chemical and Biological Weapons, thus completing an arms control
regime that covered all non-conventional weapons. See Krepon (1995).


344
Einaudi, Luigi R. 1996/7. The Politics of Security in the Western Hemisphere.
Parameters 26 (4): 13-25.
Elman, Miriam Fendius, ed. 1997. Paths to Peace: Is Democracy the Answer?
Cambridge, MA: MIT Press.
El Salvador. 1997. Informacin a la Comisin de Seguridad Hemisfrica de la OEA,
Sobre las Medidas de Fomento de la Confame y de la Seguridad, Realizdas por el
Gobierno de El Salvador: Perodo Junio 1995-Mayo 1997. San Salvador:
Ministerio de Relaciones Exteriores.
El Salvador. 1998a. Address to the Second Regional Conference on Confidence and
Security Building Measures in the Region.. San Salvador: Ministerio de Relaciones
Exteriores.
El Salvador. 1998b. Informe de Labores: 1997-1998. San Salvador: National Defense
Ministry.
El Salvador. 1998c. La Nacin Salvadorea: Su Defense, Seguridad y Desarrollo. San
Salvador: Ministry of Defense.
English, Adrian J. 1985. Armed Forces of Latin America: Their Histories,
Development, Present Strength and Military Potential. New York: Janes.
Escud, Carlos and Andrs Fontana. 1998. "Argentina's Security Policies." In Jorge I.
Dominguez, ed., International Security and Democracy: Latin America and the
Caribbean in the Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh
Press, 51-78.
Evans, Peter. 1992. "The State as Problem and Solution: Predation, Embedded
Autonomy, and Structural Change." In Stephan Haggard and Robert F. Kaufman,
eds., The Politics of Economic Adjustment: International Constraints,
Distributional Conflicts and the State. Princeton, NJ: Princeton University Press,
139-181.
Fauriol, Georges. 1992. The Future of the Western Hemisphere. In L. Eric Kjonnerod,
ed., Evolving U.S. Strategy for Latin America and the Caribbean: Mutual
Hemispheric Concerns and Opportunities for the 1990s. Washington, DC: National
Defense University Press, 1-20.
Fagan, Patricia Weiss. 1994. "El Salvador: Lessons in Peace Consolidation." In Tom
Farer, ed., Beyond Sovereignty: Collectively Defending Democracy in the Americas.
Baltimore, MD: Johns Hopkins University Press, 213-237.


120
Although resistance to this level of U.S. control was already forming, widespread
rejection would begin only in the aftermath of the U.S. intervention in the Dominican
Republic (1965).
The United States made efforts to multilateralize that action and was successful in
galvanizing the OAS to organize an Inter American Peace Force to replace U.S. troops
after the initial intervention. The Board quickly satisfied an OAS request to provide a
plan to support the deployment of the Inter-American Peace Force (IAPF) to the
Dominican Republic^ and sent a senior military advisor to accompany the OAS
Secretary-General. The IAPF was placed under Brazilian command and a number of
Latin American countries provided troops, including a Central American battalion
organized under CONDECA, with U.S. logistics support. The successful organization
and deployment of the IAPF led some to consider the advisability of a permanent force of
that nature.
But no act of assurance would be sufficient to convince Latin American states that
their involvement was anything more than an attempt to redress what had been essentially
a unilateral intervention on the part of the United States (Connell-Smith, 1968, pp. 80-
82). Resentment of U.S. tactics to wrestle an anticommunism resolution from the OAS in
Caracas in 1954 and its subsequent intervention in Guatemala boiled over after the
Dominican Republic. Several countries refused to continue isolating Cuba diplomatically,
Elements of the U.S. forces which had already intervened came under the Inter
American Peace Force authorized by a resolution adopted on the Third Plenary Session
held May 6, 1965 of the Tenth Meeting of Consultation of Foreign Ministers to change
current force to OAS force.


21
countries traditionally see hemispheric defense schemes and U.S. hegemony as "two sides
of the same coin ... the power of the United States to dictate Latin Americas foreign and
domestic affairs (Poitras, 1990, p. 123). The Haitian intervention, although it followed
years of negotiations within the OAS and was based on an emerging norm of
international relations (Tesn, 1996), did nothing to ally Latin American suspicions that a
new cooperative spirit in security meant cooperation on U.S. terms (Maingot, 1996). On
the other hand, conflict between two democratic states of the hemisphere, Peru and
Ecuador, both of which had participated in the development of confidence and security
building measures (CSBMs) shown to be effective elsewhere in the region, undermined
the new spirit of Latin American security cooperation. 18 These seminal events
corresponded with a turn away from new initiatives at the hemispheric level and a new
focus on subregional security.
At first glance, such a trend seems inconsistent with the integrating processes of
globalization. But given the asymmetrical power relations in the hemisphere, it is
plausible to argue that globalizing influences in fact accelerated the trend towards
subregionalism. Economic globalization, for instance has led not to consolidation of a
free trade area of the Americas (still a goal for the year 2005), but rather to new or
18 Much of the work on security institutions in this hemisphere focuses on
CSBMs, which are also are central to this study. Fundacin Latino-Americana de
Ciencias Sociales (FLACSO) of Chile has published two important volumes on the
subject, under the editorship of Franscisco Rojas Arevana (1996a, 1996b). Both volumes
adopt an institutionalist focus, describing the hemispheric concepts and institutions that
deal with security and illustrating their influence in a series of regional case studies.


228
At present there is no subregional analog to the Rio Treaty or the hemispheric
military organizations and security commissions. Instead, there are numerous, partial
organizations with overlapping membership and jurisdiction, ranging from bilateral
meetings for the exchange of information to multilateral institutions that manage burden
sharing commitments.
Bilateral contacts have been enhanced through meetings at various levels with
different components of the armed forces. It is likely that these contacts were initially
spurred by the hemispheric conferences of the armed forces. Both the Commanders
Conference and the Intelligence Conference, one of the specialized conferences within the
Army system, have facilitated side meetings between the military chiefs since the 1960s.
Brazil and Argentina completed their twenty-second annual meeting of
commanders in 1995 (Argentina 1995 CSBM Submission). Brazil and Uruguay have had
similar conferences each year, dating back to the 1980s (Brazil 1996 CSBM Submission).
Similar meetings have taken place between Argentina and Uruguay (Argentina 1997
CSBM Submission). These meetings are clearly not new phenomena.
But contacts since the 1990s have moved well beyond sidebar, bilateral meetings
between commanders convened at hemispheric conferences. In 1995, Argentina
institutionalized high-level consultation mechanisms with Brazil and Chile, followed by
similar agreements with Uruguay, Bolivia and Paraguay (Villagra, 1998).
Contacts at lower levels, which complement meetings of the high commands in
the region, are a hallmark of the 1990s in the Southern Cone. Argentina has an extensive
network of such meetings with Brazil and Uruguay. Argentina and Brazil began
meetings between the heads of their respective navies in 1995, with discussions that


227
institutions and influencing their rule making. But U.S. power might be present as a
balancing force. How do the countries in the subregion react to that possibility? Do they
strengthen institutions to exclude the exercise of U.S. power? Or does the United States
form part of the power calculus of the Southern Cone?
These research questions drive our analysis in this case study. They are designed
to illuminate the differences between the realist and institutionalist approaches.
Wherever possible, we will attempt to answer them clearly, so that a confident judgement
about the predictive power of the theories can be reached.
Institutionalization: Organizations and Rules
The organizations of the IASS are well supported by the major Southern Cone
countries. Argentina, Chile and Brazil have hosted or will host important hemispheric
conferences relating to security. Argentina was the site of the CSBM Meeting of Experts
(1994), the second DMA and a major counter-terrorism conference (1999). Chile hosted
the first regional CSBM conference (1995) and the second presidential summit (1998).
Brazils involvement has been more limited, although that country will be the site of the
fourth DMA to be held in the year 2000.
In addition, these countries have the best record of participation in the UN
registries. As reviewed in Chapter 4, the Southern Cone is the most active region in terms
of CSBMs declared to the IADB and the OAS Committee on Hemispheric Security.
Most of that reporting has come from Argentina, Brazil, Chile and Uruguay. Finally, all
countries in the Southern Cone are parties to the major security instruments and
institutions reviewed in Chapter 4.


Ill
What of the OAS Charter, which established the organization in 1948?16 The
Charter includes the principle of collective responsibility for hemispheric security and
collective action in Chapter 5, Articles 24 and 25. In addition, many of the basic
purposes of the OAS listed in the opening paragraphs of the Charter show a distinct
security component, such as strengthening peace and security; preventing potential
disputes and ensuring their peaceful settlement; and achieving an effective limitation of
conventional weapons so that the those resources can be devoted to economic and social
development. All of these security components are qualified by due respect for the
principle of nonintervention.
The organs of the OAS meet annually with the foreign ministers of the region in
General Assembly to consider resolutions developed by committees of the OAS
Permanent Council and to establish new mandates for action. The resolutions and
mandates are normally recommendatory in nature when directed to member countries and
mandatory when directed at specific elements of the OAS.
Security cooperation through the OAS has included some cases of combined
operations (a combined quarantine force against Cuba in 1962; combined
patrols/observation posts on the Costa Rican-Nicaragua border in 1955; a combined naval
patrol in Panama in 1959); one case of centralized action under a unified military
16 The Charter was subsequently amended by the Protocol of Buenos Aires
signed in 1967, which entered into force in February 1970; by the Protocol of Cartagena
de Indias, signed in 1985, which entered into force in November 1988; and by the
Protocol of Washington, signed in 1992, which entered into force in 1997. These
protocols expanded the role of the OAS in economic and social development and the
strengthening of democratic institutions.


127
During Falklands War, Venezuela extended diplomatic support and Peru provided
military aid to Argentina (Varas, 1985), while Brazil equivocated and Chile offered the
British use of communications facilities (Kelly, 1997). Chile also refused to support
Argentina in the OAS,38 unsurprising given the war scare a few years earlier over the
disputed Beagle Channel islands of Picton, Lennox and Nueva.
In Central America, a "shatterbelt" in the 1980s for Soviet and U.S. competition,
the role of U.S. power has been more apparent (Kelly, 1997). But even in this subregion
local rivalries count. Non-ideological conflict has been frequent between Honduras and
El Salvador and between Nicaragua and her neighbors to the north and south. When
Nicaragua began its arms build-up under President Anastacio Somoza, the neighboring
countries of El Salvador and Honduras increased expenditures, although a more distant
Guatemala did not.39
As in the previous century, the elements of threat continued to drive balancing
patterns. Changes in technologies that favor offensive operations since the 1980s have
undoubtedly exacerbated conflict. Certainly, Argentina's experience of the Falklands
38 Colombia, Trinidad and Tobago and the United States also abstained from
voting on the 1982 OAS resolution recognizing Argentine sovereignty over the islands
and calling for a cessation of British actions (Rengger and Campbell, 292).
39 Varas (1985, 83) does not believe, however, that perception of threat
adequately explains arms build-ups in Central America. He believes the principal cause is
internal conflict. Varas (1985) does find, however, that patterns of arms trade in South
America follow traditional ententes. Brazil exports arms to Chile but not to Argentina;
Argentina exports to neither, but does sell to Bolivia and Peru; and Chile has exported
arms to Ecuador (55). Resende-Santos (1996, 224-242) finds a balancing dynamic
behind patterns of modernization. Gupta (1997, 146), however, claims that Brazilian
modernization at the end of the military regime (1964-1985) occurred in a low threat
environment, notwithstanding the impact of the Falklands.


280
Fonseca. El Salvador has an active diplomacy with other Pacific Rim countries, such as
Chile, from which it purchased a computer simulation system for exercising battalion
level military operations.
In contrast, Honduran and Nicaraguan cooperation has matured over the 1990s.
Not only have the two countries engaged in joint patrols on the frontier, but both have put
forth a substantial effort to remove mines from their common frontier. This offers clear
evidence of a significant reduction in defense postures. The mines were laid along the
principal routes connecting the two countries. Early on, routes were cleared to allow
passage across the frontier. Now, the effort is primarily in the removal of mines in the
rural areas adjacent to the border, allowing farmers to return the land to productive use.
Without question, here is an example of coordination to reap absolute benefits both
those associated with increased economic productivity and the return of displaced
persons.
In the area of conventional arms, the Democratic Security Treaty commits
signatories to a "reasonable balance." Nicaragua, El Salvador and Guatemala
significantly reduced the size of their armed forces as an important part of the peace
negotiations that ended the armed internal conflict in those countries. 19 El Salvador, for
instance, reduced its budget by 50% and its armed forces downsized from over 60,000 to
20,000 troops, although it still spends more than twice as much as Nicaragua and
Honduras (Rojas, 1996b).
19 Although levels are still significantly higher than they were in the 1970s. The
subregion began its peace process with a heavy military burden (Torres, 1998).


172
Defense Postures
The rules guiding defense postures are less institutionalized at the hemispheric
level than those in the area of transparency. Here, we will consider the regional nuclear
weapons regime and efforts at conventional arms control.
The international nuclear regime contains one instrument that is based in Latin
America the Treaty of Tlatelolco, formally known as the Treaty for the Prohibition of
Nuclear Weapons in Latin America, which began to evolve after the 1962 Cuban Missile
Crisis. The following year, Mexico, Brazil, Chile, Bolivia and Ecuador began
negotiations to prevent the prospect of a nuclear confrontation in Latin America. From
this emerged the Treaty of Tlatelolco, opened for signature in 1967, which prohibits the
testing, production, possession or acquisition of nuclear weapons in Latin America.
Aside from the many binding articles that apply to the regional parties, the Treaty
of Tlatelolco also contains two protocols that bind consenting extra-regional states.
Protocol I obligates all states with territorial interests in the region to keep Latin America
a nuclear weapon-free zone (NWFZ). Protocol II requires all nuclear weapons states that
are party to the treaty not to use or threaten to use nuclear weapons against the
contracting parties to the protocol. The Treaty of Tlatelolco also includes an explicit
provision for "challenge inspections" (in addition to the safeguards of the IAEA) to verify
compliance with the treaty.
Numerous OAS resolutions have given the Treaty considerable support. At the
1992 General Assembly, the heads of delegation approved Resolution 1179, which
pledged member states to prevent all forms of proliferation of weapons of mass
destruction. A series of resolutions on Consolidating the Regime Established by the


171
Table 4-3 continued
Sector
Santiago -1995
San Salvador -1998
Observations
Defense
Posture
Hold talks to develop
procedures to
limit/control
conventional weapons.
Continue discussions to
advance conventional
arms control in the
region.
No new
developments.
Education programs for
peace.
Stimulate contacts and
cooperation between
legislators about CSBMs.
Adds another
governmental
sector.
Seminars/courses on
CSBMs, especially
involving military and
civilian personnel.
Hold seminars, course
and studies in diplomatic
institutes, military
academies, research
centers and universities
on CSBMs, security, and
disarmament, with
participation of civilian
and military officials and
civil society.
Lists specific
venues where
CSBMs
discussions are to
take place.
Broadens
participation to
include elements
of civil society.
In sum, then, the transparency regime is well developed. It is not binding and
does not include strong measures for verification, despite OAS experience in that area.29
Nevertheless, the verification mechanism that does exist a network of reciprocal
exchanges between neighboring states is operable. As we shall see, it is more effective
at the subregional than the hemispheric level.
29 The OAS Commission on Inspection and Verification (CIAV) in Central
America was created to verify the demobilization and disarming pursuant to the UN
peacekeeping missions that helped implement the peace agreement in Central America in
the early 1990s. Specific tasks included the physical verification of the destruction of
arms caches and reductions in the sizes of military forces, both state and guerrilla.


306
UNITAS program has offered opportunities for some naval coordination: both Ecuador
and Colombia join the United States for naval maneuvers off the Colombian Pacific
Coast (Ecuador 1995 CSBM Submission). Peru conducted naval operations in support of
civic action on the Putumayo river with Colombian forces, as well as similar operations
with Bolivians on Lake Titicaca (Peru 1996 CSBM Submission). Peru reported some
combined operations with Chile. For instance, Peruvian forces worked with Chilean
engineers on road construction in the south of Chile (Peru 1997 CSBM Submission).
Analysis
The high threat levels that predominate in the subregion have undermined
cooperation, not attracted it. The general pattern is squarely supportive of realism, even
though the balancing hypotheses proved untestable due to the pervasiveness of
transnational security problems.
Low levels of subregional institutionalization also conform to realist predictions.
Lack of hemispheric consensus on the use of military to fight the drug trafficking threat
means that there is not even a forum in which to discuss security cooperation in this area.
But where institutions are available and linkage to other issues is strong, cooperation has
emerged. Venezuela and Colombia have institutionalized security cooperation at a level
commensurate to the Southern Cone. Peru and Ecuador reached an historic accord with
the help of the guarantor states.
These positive trends may eventually lead to subregional cooperation, but that is
unlikely to emerge without sustained assistance from outside the subregion, principally
from the United States.


202
A closer look at categories B and C, totaling 23% of measures reported, shows
mixed compliance across the hemisphere. Participation in the United Nations Registry of
Arms Transfers was reported by Argentina, Brazil, Ecuador, Mexico, the United States,
Uruguay and Venezuela. UN sources show that others have participated as well during
this period (Chile, Peru and Nicaragua), but their reports were not included in the
declared data submitted to the OAS and IADB.
Participation in the UN Standardized Instrument for Reporting on Military
Expenditures was reported by Argentina, Brazil, Canada, Mexico, United States,
Uruguay, Peru, Chile, Ecuador and Venezuela. Other sources indicate that El Salvador
has also participated, although that information was not made available through the
CSBM reports.39
Defense white papers have been published by Argentina, Brazil, Canada, Chile,
El Salvador and the United States. While this level of participation is quite low, most
other countries have made some sort of comprehensive report of their defense strategies
at the numerous conferences reported in the data.
As noted, verification measures (category E) make up the majority of the
measures reported. Most of the measures reported reflect increased personal interaction
between military members of neighboring states. In particular, the data shows that almost
50% of the measures reported were visits and exchanges of observers and students. Most
39 There are several sources for information on participation in UN registries. In
addition to official UN publications, see volumes of World Military Expenditures and
Arms Transfers, published by the U.S. Arms Control and Disarmament Agency and The
Military Balance, published by the International Institute for Strategic Studies.


342
Organization in the Western Hemisphere. Syracuse, NY: Syracuse University Press,
47-90.
Cope, John A. 1995. International Military Education and Training: An Assessment
(McNair Paper 44). Washington, DC: Institute for National Strategic Studies.
Corrales, Javier and Richard E. Feinberg. 1999. Regimes of Cooperation in the
Western Hemisphere: Power, Interests, and Intellectual Traditions. International
Studies Quarterly 43 (1): 1-36.
Council on Foreign Relations. 1962. The Organization of American States and the
Hemispheric Crisis. New York: Harper and Row.
Cronin, Bruce. 1998. "Changing Norms of Sovereignty and Multinational Intervention."
In Joseph Lepgold and Thomas G. Weiss, eds., Collective Conflict Management and
Changing World Politics. Albany, NY: State University of New York Press, 159-
180.
"Creation of Regional Defense Body Analyzed." La Nacin (Buenos Aires internet
version), 21 July 1998, Foreign Broadcast Information Service (on line), 22 July
1998.
Cuevas, Luis and Jose Nakata. 1997. "El Fortalecimiento de la Confianza Mutua de las
Amricas." Revista del Colegio Interamericana de Defensa 22: 9-28.
Daadler, Ivo. 1992. The Future of Arms Control. Survival 34: 51-73.
Dagnino, Renato P. 1985. "A Industria de Armamentos: O Estado e a Tecnologia." In
Clvis Brigagao, ed., O Armamentismo e o Brazil: Guerra Deles. Brasilia: Editora
Brasiliense.
"Defense Minister Rejects Costa Rican Area Demilitarization Plan." Paris AFP (Central
America), 12 February 1998, Foreign Broadcast Information Service (on line), 19
February 1998.
Dent, David W. 1995. Latin American Policymaking: A Reference Handbook.
Westport, CT: Greenwood.
Department of Defense. 1995. United States Security Strategy for the Americas.
Washington, DC: Inter-American Affairs.
Desch, Michael C. 1998. "Why Latin America May Miss the Cold War: The United
States and the Future of Inter-American Security Relations." In Jorge I. Dominguez,
ed., International Security and Democracy: Latin America and the Caribbean in the
Post-Cold War Era. Pittsburgh, PA: University of Pittsburgh Press, 245-265.


257
Brazil and Chile. But those countries, which also cooperate with the United States, are
reacting primarily by increasing cooperation with Argentina, with minimum balancing
against an Argentine-U.S. axis. Their behavior indicates a strong desire to defend of the
integrity of the subregional security system. Balancing considerations, possibly behind
the rapprochement between the United States and Argentine, go no further than those two
countries.
Outside of this, in fact, support for realism is very thin. Realism may shape
cooperation on the margin, but it is not able to effectively constrain it.
By contrast, there is much stronger evidence that organizations and networks
facilitate increased security cooperation. It seems as if the multilateral network that
supports security cooperation is able to pursue absolute gains in almost every area related
to defense and security in the region. Burden sharing in peacekeeping exercises and
operations, security coordination in sensitive border areas, mutual support for patrolling
responsibilities, greater and greater inter-operability, especially with regard to
communications systems and radar these elements of legitimate cooperation are present
in the region.
The "nesting" of security within Mercosur is an important characteristic of the
subregion. The militaries clearly identify with the broader security concept that Mercosur
promotes, although Chile and Brazil less so than Argentina. Nevertheless, economic
linkages clearly promote cooperative security at least at this time in the consolidation of
the Mercosur project. The transparency that a common market generates is also an
important consideration.


226
drug trade flowing generally northward from Bolivia to Peru. Would cooperation emerge
in these areas? Or would cooperative arrangements follow a balancing dynamic to
compensate in some degree the problems of Brazilian strength and Argentine weakness?
There are a series of important research questions derived from the discussion of
the context of security competition and cooperation earlier. First, are there any
subregional security organizations? If so, do they help reap significant absolute gains -
especially in the areas of increasing transparency among Argentinas principal
competitors, Brazil and Chile; in the area of reduced defense expenditures; and in the
area of the limited transnational threats mentioned above?
Second, are the rules within these organizations comprehensive, inclusive and
binding? Are verification mechanisms employed to lock-in the absolute gains of
transparency and defense expenditures? Are agreements to address transnational threats
enduring? Do they require changes in national security priorities that may divert
resources to address regional concerns instead of national ones?
Third, is the actual behavior pattern rule-based or power-based? With regard to
transparency and verification, are the information flows stronger along balancing axes or
across them? Does substantial and effective military transparency ensue or are
information flows restricted and verification simply symbolic? Are defense postures
being reduced or is modernization and re-equipping ongoing? If a pattern of reductions is
evident, is it related to regime rules? Lastly, are countries devoting resources to share the
burden of addressing common threats?
A final consideration: What is the impact of the United States? In this subregion,
we do not expect to see the United States assume a role as a security manager, bolstering


88
interests are developed primarily from official sources. Sources for U.S. interests include
published data such as the U.S. National Strategy, National Military Strategy, the
Quadrennial Defense Review, and other informative sources. Latin American interests
are revealed through official documents and debates in institutional forums, which often
present well-developed national positions on a wide variety of security issues.
Regional military capabilities can be objectively determined through a number of
open source conflict profiles and balance of power studies. Although precise
measurement over time is difficult, rough profiles suffice to bring out
balancing/bandwagoning proclivities. Presence of enduring border disputes offers
evidence of aggressive intent. This mix of interests and capabilities provides one
explanation for patterns of security cooperation.
Institutions provide another. Observable characteristics include (1) degree of
administrative capacity to support multilateral decision-making and track compliance;
(2) issue agenda, particularly as compared to U.S. security interests; and (3) issue
linkages, especially nesting of security institutions within larger structures. The rules
produced by security organizations can be examined as to their degree of bindingness and
ease of verification. Sources for these factors derive from official reports and minutes of
meetings in institutions of IASS. Substantial new material from the archives of the
IADB, the OAS and the Defense Ministerials is introduced for this purpose. Staff reports
from the IADB on exercises, security studies and defense plans are also key to tracing the
evolution of institutions. Research monographs from senior Latin American officers
assigned to the Inter-American Defense College also add new perspectives on that


138
Assembly, in its Resolution 1179, recognized the opportunity and the need for increased
dialogue on and cooperation in security matters, among the nations of the hemisphere, in
light of the new international situation. This spirit led directly to a Meeting of Experts
on CSBMs (actually mandated by Resolution 1237 in 1993), which was held in Buenos
Aires late in 1994. The meeting ultimately produced a draft proposal of measures for
possible adoption in the hemisphere.
The project to hold a meeting of experts was undertaken by the Special
Committee. The debates that ensued during formulation of an agenda are very
instructive. Canada, based on its peacekeeping experiences, suggested focusing on
frameworks, mechanisms, and instruments for guiding and managing relations between
states as a basis for cooperative security. The Mexican contribution clearly objects to
the overall enterprise:
... we have as a starting point for our work an undefined and practically
nonexistent new concept of hemispheric security. Although reference
has been made to some new threats, by no means agreed upon by all,
these threats are not military in nature or origin. The delegation of Mexico
has some problems with the Meetings approach to its topic, because it is
proceeding to draw up a list, or catalog, of military confidence-building
measures without having previously identified the military threats that
must be reduced or neutralized. (OAS, 1993a)
This tension over what threats should be considered as hemispheric in nature and what
cooperative military action those threats should invoke is a consistent theme of the 1990s.
The Committee sponsored numerous specialized conferences, addressing a wide
variety of security themes such as terrorism, civil-military relations, small island security
(dealing primarily with transnational threats in the Caribbean) and education for peace.
The conferences dealing with small island security are worth mentioning here, because


194
Table 4-4: Hemispheric Cooperative Security Regime for Use of Armed Forces
Principle
Norm
Rules
Procedures
Transparency in
military matters
will reduce
uncertainty and
worst case
analysis by
demonstrating
peaceful intent of
neighbor, thus
promoting
peaceful
resolution of
conflict should
differences arise.
Be open
Announce exercises in advance
No specific
mechanism
Submit declared data on defense
expenditures. Exchange data at
OAS. Develop common
methodology for region.
UN Instrument
for Reporting
MILEX
Submit declared data on arms
transfers. Exchange data at OAS.
Improve and broaden
information.
UN Arms
Transfer
Register.
Submit white papers and other
official pronouncements of
doctrine and strategy, including
information on organization,
structure, size and composition.
No specific
mechanism.
U.S. and Chile to
sponsor
workshop.
Accept observers
and exchange
officers
Conduct exchanges in training
programs and issue invitations to
observer exercises.
By invitation to
IADB.
By arrangement.
Military burden
sharing resorts to
use of force as
ultimate sanction
and can contribute
to hemispheric
security through
humanitarian
operations.
Exhaust all
efforts to resolve
disputes before
use of force
Ranges from mediation to
censure to exclusion
Bogota Pact
Good Offices
Respond
collectively to
aggression
No operating rule for inter-
America defense, since RioTreaty
is regarded as obsolete
Some
standardization
for operations.
Remove
landmines
CA demining program
Established by
OAS and IADB
Enhance border
contacts
Develop communications
systems. Develop cooperative
activities.
Binational
communication
systems. Some
standardization.
Prevent
incidents;
increase security
to avoid
accidents or
incidents in
transportation.
Hold meetings and activities.
Develop cooperative program on
transportation of nuclear waste.
International
protocols.
Respond to some
transnational
threats: natural
disasters, small
island security
threats.
Little hemispheric consensus -
subregional problem which
requires subregional solution.
International and
regional disaster
relief procedures.


301
twice a year with Brazil. Meetings with Ecuador, Peru and Panama occur down to the
local commanders level and take place as often as three times a year. The main purpose
of these meetings is to exchange information about regional security problems -
especially guerrilla and drug trafficking activities (Colombia 1996 CSBM Submission).
In addition, Colombia holds annual meetings between its defense minister and
counterparts in Panama, Ecuador and Peru.
Contacts between Colombia and Venezuela under COMBIFRON, by contrast, are
frequent and continual. The system of frontier communications sponsored by that
organization includes radio links between ground, air and riverine units to coordinate
operations and exchange information. This network is complimented by cellular phone
between commanders at battalion level and above.
In addition to these substantial contacts, COMBIFRON also promotes verification
and conflict resolution. In 1997, when Colombia accused Venezuelan troops of illegally
detaining Colombian citizens in Colombian territory, the COMBIFROM verification
commission investigated the claim, despite attempts by Venezuelan defense leaders to
prevent such action. Ultimately, Colombia's claims were upheld and Venezuela
apologized for the transgression. The fact that the verification mechanism functioned as
planned is an important indication of the strength of this security institution.
Elsewhere in the subregion, the verification mechanisms under MOMEP bear
consideration. Although overseen by the guarantor states, a significant portion of the
operating costs of MOMEP were borne by Peru and Ecuador. Now, with the absence of a
physical presence of the guarantors, troops from Ecuador and Peru have assumed joint
verification responsibilities. That these functions are undertaken in a border environment


284
operation in Haiti. In the case of the Haitian operation, this participation was supported
in large measure with U.S. material assistance.
The "Fuerzas Aliados Humanitarian Operations" exercise, focused on disaster
relief, has also taken place in the subregion since 1996 (El Salvador, 1997). The exercise
follows the same basing pattern as the peacekeeping exercise, although participation has
been much wider. The inclusion of civilian officials due to the nature of the exercise
made it possible for Panama and Costa Rica to participate, and also facilitated
participation by Nicaragua, a country that only restored military-to-military relations with
the United States in 1999, with the return of a Nicaraguan delegate to the IADB.
The humanitarian/disaster relief exercises spawned an independent effort by the
Central American armed forces. El Salvador (1998 CSBM Submission) reports that in
August 1997, the first multilateral humanitarian exercise took place solely for the armed
forces of the subregion. The exercise included the deployment of air assets by
participating countries. In addition, an exercise directed by the CACAF in Nicaragua
exercised regional civil defense forces, with participation by the armed forces of member
countries. CACAF includes provisions for the activation of a combined task force in
support of the national emergency centers of countries affected by a natural disaster
(CACAF, 1998). This mechanism, currently under review in CACAF, was not used
during disaster relief operations pursuant to Hurricane Mitch in 1998 because each
country had to react to damage within its own borders (El Salvador 1998 CSBM
Submission), although El Salvador did provide some assistance to Honduras. In a related
area, a simulation exercise to prevent aircraft accidents took place in June 1998 in



PAGE 1

THE NEW POLITICS OF COOPERATIVE SECURITY IN THE AMERICAS BY WILLIAM B. FULLERTON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENT OF THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1999

PAGE 2

TABLE OF CONTENTS page ABSTRACT iv CHAPTERS 1. COOPERATIVE SECURITY IN THE AMERICAS 1 Principal Themes 1 Recent Trends in Hemispheric Security 11 The New Politics of Cooperative Security 20 2. THEORY 30 Introduction 30 Cooperative Security Regimes 32 Contending Paradigms 39 Contending Hypotheses 74 Operationalization 87 3. ASYMMETRY AND SECURITY 94 Introduction 94 The Role of Power in Hemispheric Relations 96 Balancing at the Subregional Level 125 Conclusion 129 4. HEMISPHERIC COOPERATIVE SECURITY INSTITUTIONS 134 Introduction 134 Institutional Evolution 136 Regime Rules 158 Regime Effectiveness 197 5. SUBREGIONAL COOPERATIVE SECURITY INSTITUTIONS 218 Introduction 218 Southern Cone 220 Central America 258 ii

PAGE 3

Andean Region Conclusions -^^^ 6. CONCLUSION 311 Review of the Evidence 3 1 1 Contribution to Theory 320 Limits of System-level Theories 325 A Modest Proposal 332 BIBLIOGRAPHY 338 BIOGRAPHICAL SKETCH 363 iii

PAGE 4

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor Of Philosophy THE NEW POLITICS OF COOPERATIVE SECURITY IN THE AMERICAS By William B. Fullerton December 1999 Chairman: Terry M. McCoy Major Department: Political Science Hypotheses derived from institutionalism are tested against the pattern of cooperative security observed in Western Hemisphere, especially since the end of the Cold War. The study addresses the hemispheric level and in three subregions: the Southern Cone, the Andean Region, and Central America. Findings confirm that security institutions have an independent effect in the areas of transparency in military matters, reductions in defense postures, and burden sharing. State power and interest play an important role, but power based theories alone are inadequate. At both the hemispheric and subregional levels, the processes of institutional politics produce rules that validated institutional hypotheses. The pattern of security behavior that emerges is generally rule based. The study makes three contributions to theory: (1) improved imderstanding of security politics in asymmetric systems; (2) the impact of relative gains on cooperation; and (3) the relationship of transparency, democracy and cooperative security. iv

PAGE 5

CHAPTER 1 COOPERATIVE SECURITY IN THE AMERICAS Principal Themes This is a study of the patterns of security cooperation and competition in Latin America since the end of the Cold War. Its focus is on changes in security behavior in the 1990s, or the "process of moving beyond praetorianism and the doctrines of counterinsurgency" (Child, 1996, p. 11), which have characterized the hemisphere's Cold War alliance system for the past half century, to a new mode of conflict management based on security cooperation and military transparency. The theoretical tools used to study this process derive from realist and institutionalist traditions in international relations, the first based on power and the second on international organizations. Many observers argue that a new security regime is emerging in the Western Hemisphere, based on an understanding that the new threats to hemispheric security are largely transnational in character and require interstate cooperation. Consensus on how to jointly address these new security challenges remains elusive. But despite the absence of central direction, an astounding array of transparency measures and cooperative endeavors in the security field has arisen to capture the imagination of policy makers throughout the region. The "procedures and mechanisms to ensure a secure, prosperous and democratic region" are in evidence all around us only awaiting final agreement on organizational architecture to "transform disparate intentions into cohesive, fiinctional, and comprehensive security arrangements" (Downes, 1 996, p. 8). 1

PAGE 6

2 Institutions and new rules regarding the use of military capabilities are emerging, but at the same time interstate rivalries and power politics remain important features of the foreign policies of all states in the region. Security cooperation itself is not new to the hemisphere, but rather a normal tool of foreign policy that occurs within the framework of the shifting informal ententes that characterize many security systems. A critical view of the "new cooperation" might point out that much of the activity is restricted to the conference rooms of international organizations that are far removed from military bases where weapons modernization is ongoing and from borders that remained armed and mined. The sources of conflict have not disappeared.^ They have endured despite previous attempts to engineer solutions that would replace power as the ultimate arbiter in the international relations of the region. What theoretical optic is available to sort through these conflicting tendencies? The Latin American experience has been hard on theories. The field of political economy, for instance, has seen the rise and fall and occasionally the rise again of theoretical approaches as disparate as modernization theory, hegemonic stability theory, Marxism, dependency theory, and bureaucratic authoritarianism. Explanations for regime change have covered the theoretical waterfront as well. Consensus explanations of macro-social change have been elusive. ^ For instance, border disputes, which have been the trigger for most militarized disputes in the past two decades, are still an important source of conflict (Mares, 1997). Downes (1998) cites the Secretary-General of the Organization of American States as stating in 1995 that thirty-one border disputes were still active. For a listing of the more than one dozen serious border disputes, see Rojas (1997).

PAGE 7

3 The field of security studies is divided as well. Realism remains the dominant paradigm for understanding patterns of security behavior in this hemisphere, relying on the concepts of power, national interest, and strategic balancing. The contender is institutionalism, which posits that institutions can facilitate communication, forge cooperative commitments, and encourage rule-based behaviors that go well beyond mere balancing. The basic research design borrows from Grieco (1988a), who develops a set of contending hypotheses, derived from realism and neoliberal institutionalism, which look both at the emergence of security institutions and the security behavior predicted by each theory. Both theories expect states to act in their own best interest. But, can states cooperate to reap the absolute gains implied by lower levels of conflict, reduced defense expenditures, and burden sharing to address transnational threats? The realist approach ranks the likelihood of significant security cooperation as very low, since states are more concerned with relative than absolute gains and, in any case, lack credible assurance that neighbors will respect rules restricting the use of force. Institutionalism offers more hope the dilemmas of distributive gains and defection can be at least partially ameliorated through international institutions. With hypotheses derived from these two approaches, I examine security cooperation separately at the hemispheric and subregional levels, looking at the evolution of security institutions and observable patterns of security behavior relevant to those institutions. Comparison of the empirical record with explanations offered by systemlevel theories based on realism and institutionalism shows the latter to offer a more reliable framework for understanding security behavior in the region.

PAGE 8

4 Contrary to Grieco, I argue that institutions do matter, although not always in ways that promote cooperation. When interests are largely compatible and power evenly distributed, institutions can be effective; institutional mechanisms facilitate cooperation in a wide variety of ways. However, in the absence of clear consensus, especially under conditions of asymmetry, institutions may not foster cooperative behavior. That is not to say they are unimportant. Indeed, it is in these situations that institutional processes can have their most significant impact over power and interests. As we shall see, the institutional politics of hemispheric security institutions can work against the interests of the most powerfiil states, particularly the United States.^ While military balancing against the United States may not be an option, smaller states can develop a diplomatic line of resistance that rallies opposition and prevents U.S. cooptation of institutional tools. This "institutional balancing" is most likely when the security agenda is complex and issue linkages are available conditions characterize the 1990s as a result of the processes of globalization. A divided security agenda emerges that stunts the growth of security institutions at the hemispheric level. On the other hand, in subregional areas, where mutual interests are more easily identified, a rough balance of ^ The power disparity between the United States and the rest of Latin America is so great that comparisons are not always meaningful. At the beginning of the decade, the United States had more men and women in uniform and spent about twenty times more on defense than all of Latin America. Defense spending accounted for a larger percentage of the U.S. economy (which was roughly three times that of Latin America) than any other country in the hemisphere (with the possible exception of Cuba). Similar comparisons of equipment, organization, technology, and training levels deepen the divide. See Poitras (1990).

PAGE 9

5 power obtains between partners and the number of players is low, the same trends that make security cooperation problematic at the hemispheric level will instead promote it. These phenomena are best explained through institutionalist analysis. While institutionalized cooperation based on mutual interest can sometimes be explained by realism "defensive positionalists" argue that states may rationally adopt institutional tools to facilitate cooperation under some circumstances (Jervis, 1986) realism carmot account for institutional outcomes that contradict the balance of power and interests, particularly in the area of security. But as with all systemic models, there are gaps in both the realist and neoliberal explanations of the security behavior examined in this study. In the concluding chapter, I suggest that those gaps are best corrected by tools drawn from subsystemic institutionalism, which link external security behavior to the character of civil-military relations within individual countries. Interstate behavior may well be the outcome of a two-level game, in which both states and domestic actors compete and cooperate to realize their interests. The scope of the study encompasses the Inter-American Security System (lASS), which consists of hemispheric security organizations (the Organization of American States, the Inter-American Defense Board and College, the presidential summits and defense ministerials), security-related legal instruments (treaties, conventions and declarations), and institutionalized rules pertaining to the defense forces of the region

PAGE 10

(transparency requirements, cooperative arrangements and arms control) .3 At the heart of the lASS lie two permanent organizations: the Organization of American States (founded in 1948) and the InterAmerican Defense Board (founded in 1942). The OAS is the principal political organ of the system, with general oversight of conflict mediation and collective security activities. The lADB for its part is charged with the military-technical aspects of hemispheric security, to include defense planning, training, and standardization. The organizational structure of the lASS is presented in Figure 1-1, for reference at the end of this chapter. The tension between these two organizations over the past half-century reflects uncertainty about the appropriate role of the military in hemispheric security institutions, the debate between broad and narrow definitions of security and the larger issue of civilmilitary relations in the hemisphere. These themes, which we will revisit later in the study, combine with the asymmetrical distribution of power in the lASS to make organization of hemispheric defense particularly problematic. Efforts to organize hemispheric security are almost inevitably associated with the United States, sparking concern among Latin American states that a centralized system will front for the U.S. security agenda. 3 Grossi (1997, pp. 242-3) sees the lASS as including the "interlinked treaties, agreements and institutions with specific responsibilities on security," including specific instruments like the Rio Treaty and technical bodies such as the Inter American Defense Board. He characterizes the lASS as a "collective security arrangement" with a degree of obligation below that of a formal alliance. Also, see Council on Foreign Relations (1962).

PAGE 11

7 Geographically, the hemispheric case study covers the member states of the Organization of American States, which includes all independent states of the hemisphere minus Cuba. The subregional case studies focus on three subregions that share strong common traits that permit identification as regional sub-systems: the Southern Cone, Central America, and the Andean Ridge. Organizations, legal instruments and security rules particular to these subregions are analyzed in turn. While a part of the hemispheric case study, the English speaking Caribbean is not addressed as a separate subregion because of its markedly different history, language, culture and political tradition, as well as significant differences in its security environment. Only a handful of Caribbean states are signatories to the Rio Treaty and other traditional security arrangements of the region. Few are members of the Inter American Defense Board (lADB). Only recently have the Caribbean states become members of OAS. The subregion is important and merits study, especially in light of the significant security cooperation present there, but cannot be easily compared to the other subregions.^ The time frame is post1990, a start point that coincides roughly with the end of the Cold War, the beginning of recent efforts to integrate trade and the end of authoritarianism in the region. A shift in U.S. policy also became evident early in the decade, from a security agenda focused on the Soviet and Cuban threats to a new set of priorities that included controlling proliferation of weapons of mass destruction, supporting new democracies and addressing transnational security threats along with a 4 Ivelaw Griffith's many contributions are important. In addition, see the recent volume on Caribbean security edited by Michael Desch (1998) features pieces by Rosenau, Dominguez, and Griffith, among others.

PAGE 12

8 renewed willingness to work with multilateral institutions. As an historical point of comparison, previous security institutions of the hemisphere are covered in Chapter 3. The lack of a fixed end date for the study complicates efforts to draw hard conclusions about the patterns of security cooperation and the effect of regional and subregional institutions and rules regarding the use of force. For instance, there is little that can be said about the robustness of the new institutions and rules in the absence of a significant extra-hemispheric challenge or change in power distribution globally or regionally. Such changes are unlikely in the short term. But we can still analyze how institutions emerge, how they reinforce norms and create rules and how they affect behavior patterns. The topical focus is on the use of the armed forces in the foreign policies of states in the Western Hemisphere. Thus issues of transparency in military matters, reciprocal verification of defense postures, arms restraint and security cooperation (involving militaries), as well as conflict, are within the scope of the study. Security cooperation that might be present between police forces or other branches of the government is only tangentially discussed. Likewise, domestic aspects regarding the role of the armed forces within states are considered only briefly. This chapter continues with a short discussion of recent developments in security relations in light of the end of the Cold War and the onset of globalization, followed by a more fully developed exposition of the main argument of this investigation. Chapter 2 discusses the theoretical literature in order to develop theories and hypotheses. These hypotheses are then situated in a research design that will test their relative explanatory power. Specific research questions will be developed in later chapters, after the requisite

PAGE 13

9 empirical settings are examined. Chapter 2 also develops a working definition of cooperative security, with factors that can be operationalized for the detailed analysis to come. Institutional factors relate to organizational assets (administrative capacity, issue agendas and linkages) and behavioral rules (degree of bindingness and verifiability). Cooperative security behavior is measured in terms of military transparency, reductions in defense postures and combined security operations (burden sharing). An historic case study the institutional politics of the lASS during WWII and its aftermath is the subject of Chapter 3. A convergence of interests, an uncomplicated security agenda, and the interests of United States and its ability to offer assurances and side benefits to key member states help explain the institutional design of the lASS and the cooperative security behavior characteristic of that epoch. When the context changed in ways that complicated interstate relations, divergent interests led to a process of institutional decay. ^ Institutional balancing replaced "institutional bandwagoning."^ This historical review offers a case study of the enduring political processes present in hemispheric security relations, showing when cooperation is likely and when it is likely to fail. ^ This general finding is consistent with Desch (1998), as well as a number of other cooperation "pessimists." But I do not argue that hemispheric institutions are simply a tool of the United States, as Desch does. The bargaining dynamics are substantially more complex. 6 Corrales and Feinberg (1999, p. 8) employ this term, which they credit to Hurrell. U.S. acts of reassurance helped maintain "institutional bandwagoning" towards a Free Trade Area of the Americas in the early 1990s. But with the loss of fast track authority and a cooling of enthusiasm after the Mexican peso crisis, dynamism has shifted to subregional agreements.

PAGE 14

10 The second case study focuses on the main theme of the study hemispheric security cooperation in the 1990s. While the dynamics of institutional balancing and bandwagoning are similar, changes in the international context radically altered some of the underlying conditions for cooperation present in the earlier epoch. Efforts to consolidate a new hemispheric security regime have largely been frustrated, because its construction falls along the fault line that separates U.S.-Latin American priorities for regional security. Despite this failure, however, cooperative security behavior is evident in the region, going well beyond that prescribed by institutional rules at the hemispheric level. Furthermore, significant geographical variations emerge, such as the degree of military transparency and burden sharing in the Southern Cone; the depth of defense reductions in Central America; and levels of security cooperation in border areas within the Andean Pact. While some of the cooperative activity present conforms to the dynamics expected by realism, subregional institutions play an import£int role. Since the deviations vary by geographical area, we examine subregions in Chapter 5. Subregional comparisons allow a more thorough test of our contending hypotheses and inform our conclusions with a greater degree of confidence. In the case of the Southern Cone and Central America, hemispheric security institutions are replicated and strengthened at the subregional level, while in the Andean Ridge countries, little security institutionalization exists. Furthermore, in all three subregions, the relative importance of the United States differs significantly. Thus the case studies allow a clearer look at the role of both institutions and power and interests. Here, too, institutionalism offers the better explanation, but here, as at the hemispheric level, it is still incomplete.

PAGE 15

11 In the concluding chapter, we look at the limitations of the systemic approaches, both at the hemispheric and subregional levels, and consider extensions and alternative explanations. An institutional approach that connects domestic politics with hemispheric security institutions looks promising, although its explanatory power may be limited to special circumstances. We explore the approach and suggest a research design for further investigation. The chapter concludes with a modest policy recommendation. Recent Trends in Hemispheric Security Is the emergence of security transparency and cooperation a significant break from past interstate behavior in the hemisphere? The level and frequency of interstate conflict in the hemisphere has historically been low when compared to other regions. By most measures defense expenditures, size of armed forces, numbers of conflicts, battlefield deaths Latin American is the least militarized major region in the world.^ Security cooperation has not been uncommon and third party mediation has sometimes helped limit conflict that does arise. In this environment, slight increases in security transparency and cooperation may not be considered remarkable, especially in light of integrative processes currently ongoing in the political and economic arenas. This disclaimer notwithstanding, the use and threat of force in Latin American foreign policies has been a frequent feature of the almost two centuries of independence ^ This characterization of Latin America has broad support. See Dominguez (1998), Poitras (1990), Kelly (1997), Van Klaveren (1992), and Ramsey (1999). Desch (1998, 256) notes that Singer and Small found that the Western Hemisphere had the lowest number of wars per year (0. 1 1 ) and the lowest battle deaths per year (3 1 00). But it does not follow that the region is free of security competition. In the early 1980s, the region accounted for almost 20% of the Third World's total arms expenditures (Varas, 1985, p. 38).

PAGE 16

12 for the region. Mares counts in the twentieth century alone more than two hundred cases where states "either threatened or used military force or were the subject of such threats or force" (Mares, 1997, p. 195). While interventions involving the United States account for a significant portion of this number, most involve militarized disputes between neighboring states of Latin America that bear the same characteristics of conflict everywhere. The twentieth century disputes that Mares (1997) cites are taken largely fi:om the Militarized Interstate Disputes (MID) database. While many fall short of war, as defined by the standard 1 ,000 battlefield deaths, at least eight conflicts qualify by that criteria: the Second Central American War, 1906; Third Central American War, 1907; Dominican Republic-Haiti, 1937; Bolivia-Paraguay, 1931-35; Peru-Ecuador, 1939-1941; El Salvador-Honduras, 1969; Argentina-Great Britain, 1982; and Peru-Ecuador, 1995. Other disputes which were not counted came very close to the cutoff criteria. The PeruColombia conflict in Leticia in 1932, for instance, resulted in 852 deaths. Against this historical backdrop of violence and mistrust, developments in the field of security cooperation in the 1990s do indeed present an empirical contrast. In the Southern Cone, Argentina and Chile have participated in joint exercises involving their naval and air forces deployed in the Beagle channel, the object of a near war in 1978. The two countries also sent ground forces to work under Brazilian command on the border of Peru and Ecuador as part of a military observer mission. And in 1999, Presidents Carlos Menem and Eduardo Frei, in a reenactment of the 1899 "abrazo" which led to the Pactos de Mayo in 1902, signed a declaration in the austral port of Ushaiua

PAGE 17

13 pledging even greater security cooperation and military transparency. ^ That year, the last of two-dozen border disputes between the two countries was resolved. Argentina recently conducted joint military exercises with Brazil and Uruguay at a 19th century battlefield of what was then Brazil's eastern province of Cisplatine.^ Brazilian, Uruguayan and Chilean officers participate in instruction on peacekeeping activities at an Argentine training center at an old base near the same previously conflicted sector. All countries of the region to include Bolivia and Chile, still with no formal military ties have signed a declaration that establishes the Southern Cone as a United Nations Zone of Peace. The declaration commits its signatories to the removal of mines in the region significant given the recent publicity over the minefields that still exist along the disputed international border of Chile and Bolivia ("Foreign Ministers Spar," 1997). In the Andean Ridge, Colombia and Venezuela have established a border commission that coordinates joint military operations. Cooperation between those two countries extends to naval operations, despite an unresolved border dispute in the Gulf of Venezuela. And, most significantly, Peru and Ecuador have ratified a peace agreement Argentina, Brazil, and Chile signed a disarmament agreement in 1 902 known as the Pactos de Mayo. Under its terms, Argentina and Chile were compelled to resell ships to foreign arms suppliers (Varas 1985). Somewhat ironically, the original agreement was less an expression of trust and confidence than an act to solidify the balance of power system that emerged in 19* century South America (Burr, 1955). ^ War began in 1 825 between Argentina and Brazil over the Cisplatine province (modem day Uruguay). It ended in 1828 under British mediation leading to the independence of Uruguay.

PAGE 18

14 that promises a definitive end to a conflict which has engendered two wars and a series of militarized disputes. In Central America, Honduras and Nicaragua are actively demining their common border, under the auspices of the Organization of American States and the InterAmerican Defense Board. El Salvador is participating with the deployment of Salvadoran technical experts to Honduras. El Salvador, Honduras and Nicaragua conduct joint patrols in the Gulf of Fonseca. The armies of those countries, along with Guatemala, have formed a subregional security organization to promote cooperation. And all countries of Central America are signatories to the recently enacted Democratic Security Treaty, which contains many aspects related to cooperative security and military transparency. Throughout the hemisphere, we find hundreds of examples of close coordination to address common threats, unprecedented transparency in military matters, arms control agreements, and significant changes in force structure, that demonstrate a new spirit of cooperation in the area of security. The incidence of militarized conflict has dropped to near zero.^^ The hemisphere has seen periods of institutionalized security cooperation in the past, generally promoted by the United States in the face of a systemic threat, but cooperation in the 1990s is occurring in an environment that lacks a consensus threat. Furthermore, patterns of cooperative behavior reach across historic rivalries. The United 10 Wallensteen and Sollenberg (1998) compare incidence of Latin American conflict from 1989-97 with other major regions. They report eight militarized disputes in 1989 (out of a global total of 47), and only 2 (of 33) in 1997 (internal conflicts in Colombia and Peru).

PAGE 19

15 States is uninvolved or irrelevant to much of this activity. This empirical puzzle merits serious attention. What is the cause of this palpable change in the tenor of interstate relations in the region? One possible explanation lies with factors commonly associated with realism. Consider for a moment what the end of the Cold War has meant from the perspective of U.S. strategic interests. As late as 1987, Soviet assistance to Nicaragua, heavy military involvement in Cuba and extensive contacts in Peru were viewed as serious threats to U.S. interests in the region. Argentina and Brazil, in the late 1980s, avidly pursued nuclear technologies and ballistic missile delivery systems. ^ ^ As late as 1989, Cuba deployed the most powerful military forces in the region after the United States, Canada and Brazil and provided direct support to guerrilla movements in El Salvador and Guatemala. ^ 2 j\yQ Salvadoran conflict was still contested at the beginning of the decade. By the mid to late 1990s, all of these potential threats to U.S. interests had disappeared. The disintegration of the Soviet Union, the collapse of Cuba's military forces, the end to the Central American civil wars and the abandonment of nuclear and missile technology programs in the Southern Cone left the hemisphere without a strategic ^ Into the 1 990s, the Tlatelolco Treaty banning nuclear weapons from Latin America had not been accepted by the four countries most likely to proliferate: Argentina, Brazil, Chile and Cuba (Varas, 1 994). 12 Dominguez (1998) notes Cuban support of movements in El Salvador, Guatemala and Chile.

PAGE 20

16 threat to the United States, other than the transnational challenges of drug trafficking, organized crime and terrorism. These threats represent the "dark underside of globalization and interdependence" (Godson and Williams, 1998, p. 66). The same processes that accelerate the flow of goods, people, ideas and money through the global economy, facilitate criminal transnational networks, mass immigration flows, and volatile capital movements (Munoz, 1996). To the extent that these threats affect the security of the states, they can be considered with realist tools. Transnational threats may not be core survival issues for the United States, but they do have an impact on security. The threats to security caused by cross-border flows of contraband (principally drugs), people (immigration and terrorism) and other phenomena (environmental degradation, natural disasters, etc.) affect the U.S. territory both directly (drugs, illegal immigration and terrorism) and indirectly (by undermining newly consolidated democracies and creating an environment of instability which exacerbates transnational threats). Since unilateral action against such threats is not costeffective, the United States might rationally seek ways to influence behavior through bilateral agreements or the creation of security institutions that foster cooperation and 13 See Perry (1997), Wilhelm (1998). General Wilhelm commanded the U.S. Southern Command in the late 1990s. There are numerous official publications that provide detailed lists of transnational threats of concern to the United States. Often cited are the transnational threats of drugs, terrorism, trafficking in arms, migrant flows, and internal instability. The Department of Defense, "United States Security Strategy for the Americas" (1995), offers such a similar catalog. See Einaudi (1996/7), Lowenthal (1992) and Van Klaveren (1992) for good treatments of the topic. Villagran de Leon (1993) offers a review of OAS activity in these areas.

PAGE 21

17 burden sharing in these areas. A pattern of security cooperation consistent with this general optic would largely reflect U.S. interests. Another possible explanation goes beyond a focus on state power and interest, drawing on the proliferation of new international organizations and the growing transnational linkages characteristic of a globalization of economic, cultural and political issues. The globalization literature emphasizes the important role of new norms and institutions in organizing the "turbulence" of world politics (Rosenau, 1990; Vaky, 1993; Munoz, 1996). Despite the recent bearish turn in the United Nations (after euphoric predictions following the end of the Cold War), the general belief is that institutions can help, especially if carefiilly designed with incentives that appeal to national interests. These institutions are thought to be easier to create in regional organizations, such as the OAS (Lepgold and Weiss, 1998; Lambom, 1998).15 Corrales and Feinberg characterize the region in the 1 990s as passing through a high point of what they term "hemispherism": "the active attempt by the nations of the hemisphere to redirect their foreign policies in favor of closer and coordinated cooperation with one another" (Corrales and Feinberg, 1999, p. 2). Unprecedented cooperation in the 1990s is based on acceptance of new norms, creation of new 14 For example, some argue that a new norm on democracy and human rights appears to be challenging the old, statist norm of sovereignty and non-intervention (Cronin, 1998). 1 ^ The importance of organizations as ways to institutionalize norms is emphasized here. Ideas untethered from institutions are unlikely to have lasting impacts. Institutionalist treatments of Kant's "perpetual peace" are not vulnerable to assaults such as Carr's (1964) attack on WWI idealism.

PAGE 22

18 institutions, and the inclusion of new items on the inter-American agenda. The U.S decision to work through hemispheric institutions and willingness on the part of Latin American states to engage in "institutionalized bandwagoning" has been key to the success of cooperative initiatives (Corrales and Feinberg, 1999, p. 8). A renewal process at the OAS 1^ was initiated in 1991 through the adoption of OAS Resolution 1080, "The Santiago Commitment to Democracy and the renewal of the InterAmerican System.''^^ The resolution established a "firm resolve to stimulate the process of renewal of the Organization of American States, to make it more effective and useful in the application of its guiding principles and for the attainment of its objectives." While the renewal process is in large part an administrative reform to increase organizational efficiency, responsiveness and transparency of resource allocation, the movement includes rethinking and redefining several of the core fimctions of the OAS. For instance, economic development has been broadened and redesignated under the rubric of sustainable development and support for democracy has been strengthened by the creation of institutions to monitor, verify and improve governance in newly consolidating democracies. Corrales and Feinberg claim that a new historical peak of hemispherism was reached in 1 994, at the Miami presidential summit. At that meeting, the presidents of the 16 See OAS (1995a) for a review of that process. 1 ^ OAS Resolutions are listed by resolution number and title when that additional information adds to the argument. OAS conferences are referred to by title or subject matter and date. OAS publications are available through the Office of the Secretary General, Organization of American States, Washington, DC.

PAGE 23

19 Americas committed their countries to: "cooperate toward an impressive array of goals in politics (advancement of democracy, good governance), in economics (free trade in the hemisphere, financial coordination), and in security (the fight against the drug trade, the collective defense of democracy)." As Corrales and Feinberg indicate above, hemispheric security and conflict resolution have been caught up in the reform movement. As a part of that process the ministers of foreign affairs and heads of delegations, through OAS Resolution 1 080, "expressed their decision to start a process of consultation on hemispheric security in light of the new situation in the region and worldwide, from an up-to-date and comprehensive perspective on security and disarmament." The renewal process initiated change in the lASS in several areas: (1) reorganization of the institutions that address security; (2) redefinition of the principles that underlie hemispheric security; and (3) an ambitious plan of action that includes concrete measures for adoption by member states that constrain use of force and promote a vigorous defense of democracy. In 1993, the President of the OAS Committee on Hemispheric Security, Dr. Heman Patino Mayer, presented a new model of hemispheric security "cooperative security" to replace the flawed collective security arrangement established by the Rio Treaty. Under "cooperative security," countries would join together to address potential threats instead of waiting until open conflicts developed. Patino recognized that such a model was based on the premise of shared interests, shared threats and a willingness to cooperate to address these threats, using all elements of national power, to include the armed forces. In his view, the hemisphere was ready (OAS, 1995b).

PAGE 24

20 There is substantial evidence that elements of this vision have taken root in the hemisphere. Institutionalized cooperation is emerging at all levels locally, subregionally and regionally. Its constituent parts overlap in places and leave voids in others, and mechanisms of coordination between levels are underdeveloped. Despite these limitations, this network of security institutions is becoming an important factor in security relations between states of the region and may be directly responsible for much of the security cooperation observed in the 1990s. This study is designed to evaluate that challenge to the realist paradigm. The New Politics of Cooperative Security Just as regional trade integration has slowed at the hemispheric level, giving way to subregional pacts, security cooperation demonstrates a similar trend. The evidence shows that cooperative security generally coalesces at the subregional level, especially where security can be nested in other cooperative regimes, such as Mercosur {Mercado Comun Sur) in the Southern Cone and the peace process in Central America. But the security institutions that have emerged at the hemispheric level as a result of this process are far from fialfilling the vision of Patino Mayer. Understanding of how the renewal process actually unfolded in the lASS requires a theoretical model that can encompass the dynamics of power, interest and institutional behavior. The U.S.-led intervention in Haiti (1994) and the shock of the Ecuador-Peru conflict (1995) confronted states in the region with two guideposts to the future: the first, an example of outcomes likely under an effective hemispheric security regime and the second, an example of outcomes likely under a failed regime. Latin American states, for the most part, were not satisfied with either scenario. On the one hand, Latin American

PAGE 25

21 countries traditionally see hemispheric defense schemes and U.S. hegemony as "two sides of the same coin ... the power of the United States to dictate Latin America's foreign and domestic affairs" (Poitras, 1990, p. 123). The Haitian intervention, although it followed years of negotiations within the OAS and was based on an emerging norm of international relations (Teson, 1996), did nothing to ally Latin American suspicions that a new cooperative spirit in security meant cooperation on U.S. terms (Maingot, 1996). On the other hand, conflict between two democratic states of the hemisphere, Peru and Ecuador, both of which had participated in the development of confidence and security building measures (CSBMs) shown to be effective elsewhere in the region, imdermined the new spirit of Latin American security cooperation. ^ ^ These seminal events corresponded with a turn away from new initiatives at the hemispheric level and a new focus on subregional security. At first glance, such a trend seems inconsistent with the integrating processes of globalization. But given the asymmetrical power relations in the hemisphere, it is plausible to argue that globalizing influences in fact accelerated the trend towards subregionalism. Economic globalization, for instance has led not to consolidation of a fi-ee trade area of the Americas (still a goal for the year 2005), but rather to new or ^ Much of the work on security institutions in this hemisphere focuses on CSBMs, which are also are central to this study. Fundacion Latino-Americana de Ciencias Sociales (FLACSO) of Chile has published two important volumes on the subject, under the editorship of Franscisco Rojas Arevana (1996a, 1996b). Both volumes adopt an institutionalist focus, describing the hemispheric concepts and institutions that deal with security and illustrating their influence in a series of regional case studies.

PAGE 26

22 revitalized subregional trading blocks: Mercosur, the Andean Pact, the Caribbean Common Market (CARICOM) and the Central American Common Market (CACM). In the area of security, subregionalization was driven by a number of related factors. In the first place, security issues were thought to be less salient with the end of the Cold War. The unifying theme of EastWest confrontation had disappeared, given way to a new focus on economic relations. In particular, Latin American security issues were thought to be of marginal importance, given the lack of U.S. strategic concerns in the region. This new "marginalization" opened space for subregional initiatitives (Druetta, Tibiletti, and Donadio, 1991). In the second place, the "dark side" of globalism, with the emergence of transnational threats, complicated the regional security agenda significantly, leading to a natural breakout of specific subregions based on particular security challenges (Stanely, 1999). Hence, while the Southern Cone focused on overcoming a rather standard security dilemma in order to move economic integration forward. Central America and the Caribbean were confronted with the transnational challenges of contraband and criminality. In the Andean Ridge, countries faced a witch's brew of tense interstate relations and deeply rooted transnational security threats. Transnational threats are often considered likely triggers for cooperative security, since solutions require multilateral approaches. But U.S. attempts to include transnational threats on the hemispheric security agenda have been met with resistance. Latin American countries are nervous about a broadly defined security agenda that includes many threats of a non-military nature and are unwilling to allow the United

PAGE 27

23 States to dominate a process of "securitization" that defines hemispheric threats (InterAmerican Dialogue, 1997). The asymmetry in power relations driven home by the 1989 invasion of Panama to address the results of a "narco-regime" make hemispheric consensus difficult to arrange around these sorts of issues. The "internationalization of politics" is another feature of the impact of globalization (Munoz, 1996b). Its first order effect the awakening of ethnicities and nationalities in Eastern Europe did not have a major impact in Latin America, where the business of nation building is largely complete. But new norms in the area of human rights and democracy found ready acceptance in the region, which was just completing its transition from authoritarianism as the 1 990s began. With democracy as a unifying theme and the absence of "alliance handicaps" stemming from ideological differences (Liska, 1 962), the possibilities for hemispheric convergence seemed bright. But as applied to security relations, the picture is more complex. In the early 1990s, ground work was laid for the "collective defense of democracy": OAS Resolution 1080 adopted in Santiago (1991) and the Washington Protocol amending the Charter strengthened automatic responses within the OAS to ensure a strong reaction in the event that democratic processes were interrupted in a member state. But as demonstrated during the Haitian crisis, the political will to violate sovereignty in the defense of democracy proved impossible to muster (Maingot, 1996). Mexican opposition to U.S. policy in this incident was strong. Mexican policy remains anchored in the principle of 19 This is the term employed by Buzan, Waever, and de Wilde (1998) to explain how an economic, social or cultural issue becomes identified as a threat within a given regional security complex.

PAGE 28

24 non-intervention, which also forms the basis of its Cuba policy (Gonzalvez Galvez, 1998). In 1999, when the United States pushed for an even stronger mechanism to "defend" democracy, resistance from key Latin American states killed the initiative.20 Sovereignty and non-intervention, as well as democracy, are norms of OAS Charter.^l Thus at the regional level, democracy has not been as significant an integrating factor as some expected, in part because of the presence of marked power asymmetry. But at the subregional and unit levels, the impact of democratization on security policies has not been any easier to characterize. While most analysts agree that democracy has an impact on security, few agree on precisely what that impact is (Hirst, 1 996). The consolidation of democracy in Latin America has taken a variety of paths and its foreign policy expression is equally varied. Perhaps more significant than the simple adoption of democratic forms of government is the general rise in transparency between countries, based on enhanced information flows, open markets and open political systems. This transparency has its major impact in security relations between neighboring states. For all of these reasons, conditions for subregional security cooperation were enhanced in the 1990s. In subregions with substantial economic interaction, shared civil20 At the 1999 OAS General Assembly in Guatemala City, the United States authored language to supplement Resolution 1080. An automatic OAS response would be initiated upon detection of warning signs that democracy was failing, instead of waiting for an actual disruption. 21 The "guiding principles" of the OAS are numerous. As stated in the Charter, they include respect for international law, human rights, sovereignty, noninterference in domestic matters, rejection of wars of aggression, and a recognition of the importance of social development, justice and education, among others (OAS Charter, Articles 2 and 3).

PAGE 29

25 military traditions, and common security threats, significant security cooperation has emerged. Often the trend towards more cooperative security was facilitated by specific institutions that allowed defense policies to gradually adjust from confrontational to cooperative strategies. In many cases, these subregional institutions have filled gaps in the overarching hemispheric model, which has remained incomplete and rift by divergent interests. In Mercosur, for instance, links between economic and security institutions appear to be mutually reinforcing. This is not to say that economic integration always promotes cooperative security. The Andean Pact, created in 1969, initially triggered a dramatic improvement in relations between Ecuador and Peru, whose leaders signed a number of bilateral agreements for economic cooperation, but militarized clashes between the two countries increased dramatically from 1977 onward (Mares, 1996/7). Colombia and Venezuela engaged in an arms race during the height of the Andean Pact's success.22 El Salvador and Honduras fought a war in 1 969 at the peak of Central American integration, over the status of Salvadoran nationals in Honduras who were in a sense the physical embodiment of that economic integration. 23 Mares states that "Venezuela had previously reinforced border defenses after Colombian guerrillas crossed. But the 1 987 appearance of a Colombian navy vessel in Venezuelan-claimed waters provoked mutual military mobilizations. Similar incidents in the early 1970s also fiieled arms acquisitions" (Mares, 1997, p. 213). 23 Perhaps these economic pacts leave behind institutional traces. Van Klaveren (1992) asserts that CACM, almost moribund in the 1980s, nevertheless provided a framework to begin the Esquipulas process that culminated in the Central American peace accords.

PAGE 30

26 If the processes of globalization in the 1990s have favored subregional security integration, what are the prospects for a hemispheric counterpart? Based on the pattern of institutional security politics in asymmetric systems, they are poor. Varas (1992) categorizes U.S. security behavior in the region as varying between "hegemonic control" and "coercive control." These labels exaggerate the influence of the United States, but nevertheless offer a key to understanding the new security politics. What Varas means by "hegemonic control" is a U.S. policy of using regional institutions to give a multilateral flavor to its security policies. When this course of action seems more cost-effective than unilateral action ("coercive control"), the United States attempts to create or remake security institutions that reflect its own interests in the region. Under the best of circumstances, this is not easy. Because institutions tend to restore symmetry through common decision-making procedures, opportunities for small countries to "voice" their interests, and facilitation of issue linkage new forms of political behavior are possible: "institutional bandwagoning" (support to U.S. initiatives) and institutional "balancing" (resistance to U.S. initiatives). The United States is forced into an assurance game to maintain support, requiring constant displays of multilateralism (acquiescing to majority decisions, accepting binding obligations, refraining from unilateralism), as well as significant side payments to potential balancers (e.g., Mastanduno security "outlets" and other benefits). This is because the U.S. commitment to multilateralism is rarely credible on its own merits. The "shadow of the future" for Latin American states is the certainty that the United States will defect or change the rules of the game when necessary to secure its own interests.

PAGE 31

27 However, as long as there is a mutual interest in the gains to be secured through security cooperation (e.g., defense of the continent against the Axis countries in WWII), institutional support for U.S. initiatives is likely to be forthcoming. Over time, a natural decay may set in, as more issues are brought under the purview of the institutional arrangement, leading to a more complicated agenda that is harder to support and easier to subvert. At some point, a balancer may emerge at different times in U.S.-Latin American relations, that role has been played by Argentina, Brazil and Mexico. Since in an assurance game, cooperation quickly unravels once defection is probable, institutional balancing can quickly develop into significant resistance. The conditions wrought by globalism in the 1990s are particularly auspicious for institutional balancing. The complicated security agenda makes it much more difficult to define a set of security objectives that are equally important to all member states. This same phenomenon makes it easier to link issues in ways that benefit smaller states. Finally, the expansion of the inter-American system, with the incorporation of Canada and the English speaking Caribbean, has added to the pool of potential balancers and exacerbated the well known cooperation dilemmas facing large groups. If all of this simply meant that multilateralism is difficult and likely to cause the U.S. to resort to unilateral action or engage in states bilaterally ("divide and conquer" tactics), then the institutional politics described above would be of little import. But in fact, the outcomes that the process generates are important; they have an independent impact on cooperative security behavior and the behavior that results is not always consistent with U.S. priorities.

PAGE 32

28 The logic of security politics at the hemispheric level is a special case that applies to institutions that encompass asymmetrical power relations. As such, it can be derived from the more general form of institutional theory. That theory, along with realism, is developed in the next chapter. Both provide a set of hypotheses with which to examine the successful security cooperation at the subregional level, as well as the limited cooperation at the hemispheric level.

PAGE 33

29 c s OI CS CO
PAGE 34

CHAPTER 2 THEORY Introduction Following a brief introduction, this chapter develops a theory-based description of the components of a cooperative security regime, made up of institutions that facilitate cooperation by providing organizations for negotiation and rules that prescribe specific behaviors. 1 Once that important mark is established, the chapter investigates the realist and institutionalist traditions in some depth, particularly with regard to the obstacles and opportunities these paradigms posit for security cooperation between states. From that examination, contending hypotheses emerge to explain the patterns of security cooperation present in the hemisphere. Finally, the chapter concludes with a discussion of the operationalization of key explanatory factors. Realism and neoliberal institutionalism both address the problem of security cooperation and conflict in the international system. The theoretical frameworks share many of the same assumptions of how the world works. Both assume that states are the principal actors in an anarchic global "system." In the absence of worid government, states must rely primarily on their own resources to ensure survival. Since the main threat to survival is aggression from other states, international relations, especially in the 1 Mere membership in international organizations does not constitute participation in a regime because independent decision-making is in no way constrained (Stein, 1990). 30

PAGE 35

31 area of security, tend to be conflictual. Cooperation is difficult to achieve, even if all participating states would be better off in absolute terms. The simple fact that a cooperative strategy exists that might reap absolute benefits for participating states does not mean that states, acting rationally, will adopt such a strategy. The dilemmas of cooperation the collective action (free rider) problem, cheating and relative gains must be overcome for states to reap the absolute benefits of cooperation. Institutionalism admits the possibility that multinational institutions comprised of states and non-governmental organizations can foster cooperation. Realism denies institutions any independent role. Since the two theories differ strongly on the relative importance of institutions, these offer solid analytical traction for examining cooperative security behavior and the links between that behavior and norms regarding the use of force. However, neither the realist nor institutionalist approaches in their elementary form explain patterns of security competition and cooperation in the Western Hemisphere. Both theories need to be modified to reflect the marked power asymmetry present in the region. Power-based approaches must recognize that the United States, with its long time horizon and generous resource base, will at times bear the costs and promote the agenda of multilateral institutions, even in the face of short-term, tactical reversals. By the same token, institutional approaches must recognize that no multilateral regime can effectively constrain the United States from privileging its global or domestic interests in ways that may offend institutional norms. From a theoretical standpoint, the net affect of this great power asymmetry is to mute the differences between the two theoretical approaches, making it more difficuh to

PAGE 36

32 design research questions that ably weigh the available evidence. In cases where institutions appear to succeed, success might in fact be due to institutional effectiveness or U.S. support. On the other hand, the failure of an institution to constrain the United States says little about the impact of institutions on state behavior in less extreme cases of power asymmetry. One way to escape this complication is to drop down one level of emalysis to the subregional level. Within the geographical confines of Central America, the Andean Ridge and the Southern Cone power is distributed more evenly. At this level, sharp distinctions appear in the behavior patterns predicted by realism and institutionalism. This becomes evident in the analysis of the subregional case studies in Chapter 5. At this point however, it is useful to consider what a cooperative security regime might consist of, in order to better focus discussion of the two theoretical paradigms. Cooperative Security Regimes The purpose of this section is to describe an ideal type that represents a potential end point for the developing set of security institutions in the region. We begin with a discussion of cooperative security, and then develop the constituent parts of what would make up a cooperative security regime. The end result of this discussion is a set of variables that can be operationalized and used to measure the degree to which such a regime might be evolving in the hemisphere. To develop the concept of cooperation security, we will rely on the work of Janne Nolan (1994), Antonia and Abram Chayes (1994), Joseph Nye (1991), and Ivo Daadler (1992). According to Nolan et al. (1994), cooperative security is "designed to ensure that

PAGE 37

33 organized aggression cannot start or be prosecuted on any large scale." It seeks to accomplish this purpose "through institutionalized consent, rather than through threats of material or physical coercion," an important element being the creation of "collaborative rather than confrontational relationships among national military establishments" (pp. 45). A cooperative security regime supports these objectives through a set of institutions, with norms that specify behaviors with regard to use of force.2 Sometimes the norms simply proscribe force (with the caveat that use of force in self-defense is always permitted); sometimes they encourage cooperative security arrangements to address certain shared threats; sometimes they create the expectation of collective responses to aggression. An important point is that cooperative security requires a proactive attitude on the part of participating states to deal with emerging threats before they become open acts of aggression. Cooperative security is not collective security. Its relationship to collective security is like that of preventive medicine to surgery (Nolan et al., 1994). As Nolan et al. (1994) point out, a "rich fabric of constraints" has already appeared in some parts of the world that offers instructive examples of elements that might help construct a cooperative security arrangement: limits on military operations, such as the various CSBMs in place in Europe and the Middle East; agreements for averting wars, such as measures covering accidents, hot lines, and crisis centers between the superpowers; limits on force ^ For Daadler (1992), cooperative security is based on a "set of principles, norms and rules that govern the military dimension of interstate relations" created through an "intensive dialogue and habitual cooperation" (p. 52).

PAGE 38

34 size, weapon types, and operational practices, such as those contained in the Strategic Arms Reduction Talks (START), Conventional Armed Forces in Europe (CFE), and Intermediate-Range Nuclear Forces (INF) agreements; and the nuclear, biological, and chemical weapons and ballistic missile nonproliferation regimes cooperative verification and transparency measures, such as the data exchanges and on-site inspections required by various arms control agreements, the UN arms sale registry, and the 1990 Open Skies agreement, (p. 7) All of these measures have been described as CSBMs.^ In this hemisphere, in the 1 990s, there are rules already in place associated with the composition of forces, such as prohibitions on weapons of mass destruction, arms control/restraint, ceilings on troop strength and constraints on military exercises and deployments, and restrictions on certain other military activities. There are also rules relating to transparency, such as exchange of information about military activities (e.g., exercise notification), capabilities (e.g., participation in arms transfer registries), and doctrine (e.g., publication of so-called white papers). There are mechanisms to promote coordination in security matters, including meetings between high-level representatives of the defense ministries, exchanges of personnel and military units, and participation in conferences, symposiums, etc. Some measures are designed to develop cooperation in certain joint endeavors. They include measures that promote interoperability, such as ^ It is easy to see that such an expansive catalogue of CSBMs may cause some to decry the vagueness of the concept of CSBMs one author calls them "the fastest growing business of the post-Cold War era" and points out the lack of accepted theory to explain their growth or impact. Without a doubt, the term CSBM has become a watchword for all sorts of new security relations throughout the world and includes procedures that go well beyond the "modest measures" agreed to at the 1975 Helsinki Conference, where the concept was first coined. These observations come from Desjardins (1996), a critic of generalized acceptance of the value of CSBMs. Desjardins' work is an invaluable corrective for tendencies prevalent in the literature to expand the definition of CSBMs and exaggerate their value.

PAGE 39

35 development of common military doctrine, establishment of a common military education system, and participation in joint training operations. There are discussions in some subregions regarding procedures to create multinational units, share logistics and purchase common hardware, and adopt unified command structures. How can we make sense of these disparate elements? The recent work in the area of cooperative security shows that there are several important features to consider.^ First, a cooperative security regime should have the administrative capacity to facilitate negotiation of rules, track compliance (through data collection), evaluate results (through data analysis) and report to members. Second a cooperative security regime should have binding rules regarding use of force. Use of force may only be employed as a last resort (after exhausting all other means), must be applied multilaterally (except in the case of self-defense), and can only be used in support of multilateral, consensus objectives (Nolan et al.,1994). Clearly, there must be prior agreement on what sorts of crises should invoke a military sanction. Third, rules must be produced through an inclusive decision making procedure. The Chayes (1994) assert that "a regime is inclusive to the extent that the states affected by it have a reasonable opportunity to participate in its processes and it does not unfairly discriminate against any of its members" (p. 72). These two general requirements for rules that they be binding and produced via an inclusive decision making process are important characteristics that should be present regardless of the particular aspect of security cooperation considered. This derives from Nolan (1994), the Chayes (1994) and Daadler (1992).

PAGE 40

36 Other characteristics help define specific security behavior. First, capabilities of member states must be transparent. Information must be reliable (verifiable), relevant and timely, providing early warning of defection. Second, military postures should be kept at low levels, with limited armaments capacity and constraints on exercises, ceilings on forces, restrictions on deployments (Daadler, 1992). Third, a cooperative security regime must have operational capacity to respond to crises at varying levels, through good offices, formal mediation and ultimately through collective response and burden sharing agreements. This operational capacity need not include a command element (as is the case with NATO), but some mechanism should be available for organizing a collective response to violations of the regime norms. These three specific injunctions yield the behavior clusters that form the key components of rule-based security behavior: transparency and verification; reduced defense postures; and burden sharing (collective capacity to sanction offenders). There are clear absolute gains from cooperation in these three behavior clusters.^ Military openness allows states to make a more informed choice about their security needs. Their calculations can thus avoid a "worst case" analysis requiring them to expend more resources than necessary. Reduced defense postures offers a similar absolute gain. With a lower requirement for defense, states can spend resources on other priorities, such as economic development and social programs. Regional burden sharing takes this one 5 The presence of absolute gains is an important prerequisite for cooperation under the assumptions of institutionalism. See Keohane (1984), Keohane and Martin (1995) and Hasenclever, et al. (1997). Military relations among states must be seen as a problem of "common security" (Nye, 1991, p. 162).

PAGE 41

37 step further. If states pool their capabiHties or agree to speciahze in roles where their individual capabilities are already substantial and devote those capabilities to the region as a whole, all states enjoy the benefits of more efficient allocation of security resources. In addition, there are some regional security problems which may require cooperative approaches for their solution. Problems such as trafficking in contraband, illegal immigration, environmental degradation are either impervious to unilateral action or extremely costly. This last point brings us back to the beginning. What sorts of threats or acts might trigger multilateral military action under a cooperative security regime? Presumably, an aggressor state that openly attacked a member or members of the regime would invite a military response. But what about the Haitian crisis? Does the disruption of democracy, the violation of human rights or the threat of instability also merit military intervention? What about uncontrolled trade in contraband — trafficking in drugs or small arms, for instance? If a state becomes a regional base for "narco-terrorism," is military intervention then justified? What about use of the armed forces to cooperate in nontraditional roles, such as natural disaster relief, environmental protection, peacekeeping and other humanitarian operations? The short answer to this is that an urgent security threat is whatever the regime members say it is. In this study, we will focus on two conceptual clusters that can be derived from this discussion: institutions and behaviors. Institutions are "related complexes of rules and norms," which vary in terms of their level of "institutionalization," ranging from

PAGE 42

38 informal conventions to formal intergovernmental organizations Their rules "prescribe behavioral roles for actors" in accordance with general norms (Keohane, 1989, p. 163). Together, organizations, norms and rules regarding the use of military force make up what we identify as the cooperative security regime in the hemisphere. Tracing the evolution of these elements and comparing them to underlying power and interests of states in the region especially along the fauh line of U.S.-Latin American relations offer a constructive comparison of realism and institutionalism. I treat institutions as conceptually distinct from observable behaviors in areas of transparency and verification, defense postures and burden sharing.*^ Those behaviors demonstrate a pattern of security cooperation that can be compared to that promoted by security institutions. Such an exercise offers a second test of the explanatory power of the contending theoretical paradigms. This research requires, then, data that can support discussion of the evolution of institutions vis-a-vis power and interests, as well as data that substantiates the depth and scope of cooperative seciuity behaviors. These are the principal methodological challenges of the study. The remainder of this chapter examines both the realist and institutionalist paradigms in light of their explanatory power regarding cooperative security behavior, first at the hemispheric level and then at the subregional level. The dilemmas of See Keohane (1989) for a longer discussion of typologies and factors related to this discussion. Nye (1991) offers a similar definition: "persistent sets of norms that shape expectations and constrain behavior" (p. 1 55). ^ Not all forms of institutionalism adopt this perspective, as we shall see when reviewing "reflective" approaches to the study of institutionalism.

PAGE 43

39 cooperation, especially in the context of the current theoretical debate over absolute and relative gains, figure largely in this examination. Contending Paradigms Realism Realism, tracing its origins back to Thucydides, has enjoyed a long run as the principal paradigm for understanding interstate behavior, especially as relates to security, classical realism, which took its modem form under Hans Morganthau (1973) after World War II, adopts the state as its basic unit of analysis and the twin factors of power and national interest as the best explanations for state behavior in the international system. ^ The tradition, while rich and varied, coalesces around a set of mainstream assumptions. Keohane (1989) lists three which together make up what he calls the "hard core" of Classical realism: (1) states are the principal actors in world politics; (2) they act as unitary, rational actors; and (3) seek power and define their interests in terms of power.^ Because power is a relational concept, international politics becomes a "struggle for power" (Morganthau, 1973). States tend not to trust one another, tend to rely on their own capabilities to ensure their survival and strive to maintain their military capabilities Keohane (1986) offers a useful review of the development of this school of thought. 9 Mearsheimer (1994/5) lists five: (1) states operate in an anarchic system; (2) the principal threat to a state is another state; (3) intentions can never be known with certainty; (4) survival is the highest goal of states; and (5) states are rational actors. By rational, we mean that states select strategies based on consistently ordered sets of priorities (Keohane, 1986).

PAGE 44

40 relative to other states in the system. The result is constant security competition, whereby one state's efforts to enhance its security degrades the security of others. 1^ Waltz (1979) refined classical realism with a systemic approach, carefully laying the baseline assimiptions of neorealism: States are unitary, rational actors attempting to maximize power in relation to other states in an anarchic, self-help system. His assumption that states are unitary actors meant that detailed analysis of the source of preferences was unnecessary. State interest could be reduced to the need to ensure survival. Thus, as rational actors states will seek strategies to ensure survival in the interstate system. That system is anarchic in the sense that no world government exists to manage security relations between states. States operate in a self-help system meaning they can only rely on themselves to ensure survival. Consequently, states vie inexorably to maintain a position of relative strength vis-a-vis their rivals. Perhaps because of its privileged position, realism has been subjected to periodic challenges that depart from its basic assumptions and attempt to construct an alternate view of international relations. The principal challengers since the advent of the Cold War have been variations of liberalism, ranging from the fimctionalism of the 1950s to the regime theory of the 1970s, to what is now known as neoliberal institutionalism.^ 1 Realism, as we shall see, is not sanguine about the possibility that institutions can change the basic conditions in the international system in ways that promote cooperation. 10 Jervis (1978) found that some cooperation was possible in a defensedominated world. See Glaser (1997) for a recent answer to Jervis' challengers. 1 1 Keohane (1989) adopts this term, from the "liberal institutionalism" coined by Grieco (1988a).

PAGE 45

41 Institutions simply reflect the underlying power and interest of participating states. Institutions are arenas for acting out power relationships (Mearsheimer, 1994/5) or representations of power aggregation in the form of alliances. This does not mean that security cooperation never occurs, just that it is closely circumscribed and not likely to result from institutions. To better understand why cooperation under the assumptions of realism is so difficult to achieve, it is necessary to examine more closely the theoretical literature on the standard dilemmas of interstate cooperation. Olson (1964), and others, laid out a "logic of collective action" to show why "collective goods" are normally supplied at suboptimal levels. Because collective goods are "public" or non-discriminatory in the sense that no state can be excluded from reaping the benefits, states have powerftil incentives to shirk and free-ride. ^ ^ States will tend to allow others to bear the costs for the provision of collective goods because the benefits accrue more or less equally to all. Clearly, if each state follows this logic, collective Classical realists such as Aron (1973) and Morganthau (1973) recommended formation of alliances as stabilizing, balance-of-power mechanisms. Alliances are institutions, based on Snyder's (1990) definition: "formal associations of states for the use (or non-use) of military force, intended for security or aggrandizement of members" (p. 104). But latter day realists have disagreed on the impact of alliances, as we will discuss shortly. 13 Olson also listed as a key characteristic 'jointness of supply,' meaning "an individual's consumption does not preclude consumpfion by others" (Gowa, 1989, p. 1245). Public goods theorists also employ these concepts (Kindleberger, 1973).

PAGE 46

42 action will normally not be possible in the international system and as a consequence collective goods will not be supplied. Of the three clusters of behavior we are most concerned with military transparency, defense postures and burden sharing transparency and the enhanced regional security produced by burden sharing most closely approximate collective goods. Information about military activities, once made public, is available to all and its utility does not diminish as the number of states with access to the information increases. Enhanced regional security is more problematic more difficult to organize and much easier to privatize. But in its pure form, regional security also approximates a public good. Thus both of these "goods" are subject to free rider problems, which means, that absent other factors, we should expect to see transparency and regional security provided at suboptimal levels. Another dilemma of cooperation is nicely illustrated in the context of the security dilemma described earlier. There are two distinct problems captured by the security dilemma: cheating and relative gains. Cheating problems are often illustrated by game theoreticians through the Prisoners' Dilemma. brief, two prisoners, accused of collaborating in a criminal activity carrying a stiff sentence, are held in separate cells. Their jailers have only 14 See Stein (1990) for a summary of this reasoning and a list of its major contributors. 1 5 The numerous models used in game theory to describe the strategic behavior of states are presented in Oye (1986).

PAGE 47

43 enough information to convict on a less serious charge. Both are offered freedom in exchange for evidence against their partner. If only one provides information, the other will receive a harsh sentence. Absent reliable information and assurances that the partner will remain silent, both prisoners are likely to offer evidence against each other leading to imprisonment. In short, they rationally select strategies that lead to suboptimal outcomes. 16 This problem potentially affects all three of our behavior clusters. In the case of transparency, states that report information about their military activities may expose critical weaknesses. If potential rivals do not also report their activities, then the reporting state may increase its vulnerability by compliance. Thus states may not provide information or may provide partial or misleading information. Reduction of defense postures is also affected by the possibility of cheating. States that do reduce their postures will increase their vulnerability if potential rivals do not reciprocate. Thus states may not reduce their postures, or may offset those reductions by seeking alliances engaging in competitive balancing. Burden sharing faces a similar dilemma, to the extent that it is based on the pooling of capabilities and specialization of roles. States that design their forces based on the assumption that others will provide support in certain areas (sea patrols, air defense, etc.) become vulnerable if others do not uphold their commitments. Thus states will tend not to cooperate or may offset vulnerabilities by building military capabilities, thus affecting the second behavior cluster. ^ ^ See Oye (1996) for a description of the Prisoners' Dilemma and related games. See Snyder (1990) on burden sharing concerns.

PAGE 48

44 The theoretical Hterature attempts to address the problem of defecting as illustrated by the Prisoners' Dilemma through a variety of means. To the extent that security cooperation is not a "one-move game," states can signal their cooperative intent by small incremental steps designed to elicit a positive response. If cooperation is not forthcoming, a state can retaliate. Because of the "shadow of the future" gains from future cooperation states are likely to cooperate if they do not discount future gains too heavily and the last move of the game is unknown. Axelrod (1984) allows for cooperative activity between two partners following such a "tit-for-tat" strategy. If communication is enhanced through iteration, some cooperative behavior could theoretically emerge. But in the security field this would rarely amount to more than small incremental steps related to specific security problems, such as protocols to reduce tension caused by unforecasted troop movements near disputed border areas. These steps would have to be reciprocal and easily reversible (Oye, 1986). In any case, even if the problem of lack of communication and uncertainty of intent were somehow solved, the security dilemma contains the seeds of a second dilemma to cooperation, the problem of relative gains. While inherent in the security dilemma, the relative gains problem is not obvious in the Prisoners' Dilemma. Even if information and commitments are somehow made reliable, states will gain in different measures from cooperation. Grieco (1988a) clearly lays out the problem of relative gains from the realist perspective, his argument rooted in the core assumptions of realism. He recognizes that under some circumstances relative gains matter less and states may be able to cooperate

PAGE 49

45 to realize absolute gains. But in the area of security, because of core concerns regarding survival and easy translation of relative gains into military capabilities, this possibility is limited. 18 Matthews (1996), drawing heavily on the offense-defense balance developed by Jervis, shows the impact that future, cumulative gains might have on a state's strategy to cooperate in the area of security: "If a relative gain in a current round of interaction creates advantages that allow additional gains in future rounds, relative gains will be more important" (Jervis, 1978, p. 1 14). In an offense-dominated world, first round wins can confer enormous positional advantages. If defense is dominant, recovery is possible and hence cooperation more likely. However, as Mearsheimer (1994/5, p. 23) reminds us, in practice it is very difficult to distinguish offensive weapons from defensive ones.^Other criticisms are more general. Snidal (1991) shows that in large groups, relative gains do not offer important obstacles to cooperation, since "each cooperative dyad can achieve relative gains against the other actors" (p. 394). But Snidal 's overly stylized model is sharply criticized by Grieco (Snidal, Powell and Grieco, 1993) for its assumption of constant returns from cooperation. There are apparently no empirical examples of such a system (Mearsheimer, 1994/5). In any case, not all "cooperative dyads" are equal, at least with respect to security issues. It matters if a state's neighbor 18 See Grieco, (1988a, 1988b) and Powell (Snidal, Powell and Grieco, 1993) for a discussion of various factors which affect the acuteness of the relative gains problem. Jervis (1999) and Keohane and Martin (1995) note a convergence on the issue of when relative gains will matter. 19 See also Jervis (1999).

PAGE 50

46 reneges on an arms control agreement. That a distant partner did not may offer little solace. In the case of transparency, some states may rely on secrecy more than others. A weak state may wish to present misleading displays of strength, while a strong, aggressive state may wish to hide its strength. A strong, status-quo state, on the other hand, would have no reason, based on relative gains, to fear transparency. In fact, transparency could enhance deterrence in this case. Thus, even in the absence of cooperation, transparent behavior on the part of these states might be observed. A different argument, based on the nature in which military information is handled in various states, exposes a second relative gains dilemma related to transparency. In some states with open systems and mature civil-military relations, much information about military activities is already public. That is not the case in other states in fact military information may not even be available to representatives of the government. Thus, some states do not incur additional risk by endorsing military transparency, while others do. This dilemma is well documented in studies of transparency measures attempted between members of the North Atlantic Treaty Organization (NATO) and the Warsaw Pact.^O In short, states that have much to lose from transparency may refuse to provide information or may provide misleading information intending to dissimulate weaknesses or intent. With regard to reductions in defense postures, the most obvious relative gains problem is the impact on arms producers, principally the United States but also including 20 See Desjardins (1996).

PAGE 51

47 Brazil, Argentina and Chile (three countries who together account for 90% of intra-Latin America arms transfers). Other relative gains problems associated with defense reductions are present as well. Some types of reduction (mechanized arms, aircraft, etc.) will fall more heavily on states with modernized forces, while others (troop ceilings, limits on maneuvers) may primarily affect states with larger, but less technologically advanced forces. Some states may rely more heavily on defensive systems (such as landmines or anti-aircraft missile systems) than others. Thus reductions in some areas may not be possible, or may be offset by increases in other areas. The net effect would be marginal reductions, at best. The issue of regional burden sharing raises a different question. To what extent will a state compromise its security autonomy for advantages gained by cooperation, accepting increased vulnerability in exchange for gains associated with efficient allocation of regional resources to address conmion problems? Regional burden sharing allows all states to use their defense assets more efficiently. But some states will benefit more than others by the reduction of regional threats. Clearly a state without a coastline receives no immediate benefit from agreements to share sea-patrolling duties. In cases of more complex, transnational threats, such as illicit trafficking in narcotics, calculating relative gains is much more difficult. How are the benefits of reduced drug trafficking distributed among producer, transit and consumer states? Another relative gains problem relates to the training advantages of taking part in combined operations. These exercises and operations require intense interactions that provide access to state-of-the art military techniques and technology, perhaps most

PAGE 52

48 valuable to less advanced states. These considerations make burden-sharing arrangements difficult to achieve under the assumptions of realism. Taken together, the three dilemmas to security cooperation discussed here freeriding, cheating and equitable distribution of benefits combine to form a powerful deterrent to cooperative security behavior. Table 2.1 summarizes the problems associated with cooperation.

PAGE 53

49 Table 2-1 : Dilemmas of Security Cooperation SbPPiii*ifv Collective Good Cheating Relative Gains X I IFLFlvlll MllU Prnhlem and Problem and Nnn-C^nnnerativ£ Non-CooDcrative NonRehavior Cooperative Rphiivior Miiiiary 1 raiisparenLy illL'ICa.oCU dULJi UAlilidl^O a viilnprpihilitv if V uiiiwi ciL'iiii y 11 mav relv on LaiCUldllOn — LUIICLLIVC gUUU. \x/pjiKnpccpc 3rp avoiQS worsi cxpubcu. fVifin otViPT*c fn wciow diicii y Stronp inopntive hide tn firpp HHp \.\J 1.1 V/W IIUW* weaknesses or Hispiiisp ^tfitPQ mav Tint aoorp^^i VP a^^i wool V w piuviuc XXlldillUilo. iliiUllllallUll \Jl luay proviuc Since t^QTtl5ll r\t* pal Lictl Ui LX aiXopoX CXXC J tY^O\7 Mi^t iixdy no I DC lIliUrillaLliJil. xCCiprULaLCU, IL ic iinliVplv fr\ Xo u.xxxxiv.wxy lu flpvplnn UW V WIUIJ. IVlUl c iS\J CUilCLllVC lllwrCd-dCU OUIIIC lypCo yJl ^uuuo pruuiciii, vuiiicrauiiiLy ii xcuucLiuii nidy Pn
PAGE 54

50 Table 2-1-continued aecuniy /\DaUlUlC v>UllvdlVc vtUUU Behavior A roDicni itnu r ruuivin itiiu Rphiivini* Rehavior — : — Regional —r-z More IvcglOnal ScCUTliy oUIIlC dlalCo ourucii CiiivXClll Ccui auuiUAiiiiaic u viilnprahilitv it V uiiiwi duiiii y ii mav oain mnrp iiiuy ^ciiii iiiviv onaring aiiocaiion oi MdLCd LU UU IlUl irUIIl llllCIaUllUIl lor security Uphold ana some Kegionai resources Lomini imcnis. mcmDerb niay fTQiti rtir\fA "nr^^rvi ^alll IIIUIC ilUIIl increased OllUIlK lllL'CiiLiVC lu iicc nuc. oCCUrilj UX soiuiions xo States may not transnational CUUpCXaLC Ul lllay piuuiciii^. offset vulnerabilities Cooperation in through increase in many areas may defense posture or not be possible. seeking alliances (balancing.) Thus, under the assumption of realism, the logic of collective action and the security dilemma demonstrate that sustained security cooperation is difficult to achieve. And yet we know that realism admits some forms of security cooperation. To better understand what cooperative security behavior might emerge under realism, it is necessary to look at the questions of sensitivity to relative gains and the dynamics of alliance formation. Grieco (1988b) develops the concept of sensitivity to relative gains (his "k" factor), arguing that, among other criteria, states faced with increasing levels of threat, an

PAGE 55

51 historical adversary and a relative decline in power will have a high sensitivity to relative gains. Threat, as laid out by Walt (1987), includes the systemic elements of capability, proximity, and state of technology (offense-defense), as well as the classic realist factor of aggressive intent. This last element enriches Waltz's systemic approach with a unitlevel variable, at the cost of some loss of parsimony. Nevertheless, to the extent that the presence of active border disputes can be taken as an indicator of aggressive intent, it is relatively easy to measure and incorporate into our analysis. Thus, in bounded subregions composed of proximate states with a history of conflict, enduring border disputes and recent changes in power distribution, relative gains dilemmas should be difficult to surmount. This will be especially so during times of technological and organizational change which favor offensive operations, such as the current "revolution in military affairs," based on unprecedented mobility, precision weaponry and information technologies (Mares, 1996/7; National Defense University, 1995 and 1996). Conversely, in subregions with a pacific history, low threat levels and stable power relations, relative gains sensitivity will be lower and some cooperation may be possible, using standard "cooperation under anarchy" techniques (Oye, 1986). Grieco (1988a) also argues that high levels of economic interaction in these situations will tend to make relative gains problems more severe, assuming economic advantages can be transformed into military capabilities (Grieco, 1988b). Taken together, these factors help explain patterns of security cooperation and conflict between states.

PAGE 56

52 From Walt (1987), based on his theory of alliance formation ,21 we would expect to see balancing behavior predominate states will seek partners to rally against states they perceive as most threatening. Balancing patterns should be readily observable in the behavior clusters developed to explain the make-up of cooperative security regimes: transparency/verification, defense postures and burden sharing. In his empirical work, Walt deals with the broader concept of "alignment," of which alliances are a subset. Alignments need not be formalized (Snyder, 1990). The "tightness" of such arrangements is the subject of much debate, but the presence of a common enemy is probably the most important factor. 22 of course, high levels of threat may not always induce "external" balancing; arming, or "internal" balancing is also a rational response to threat (Morrow, 1993). In any case, alliances are still subject to the same problems inherent in less formal forms of security cooperation.^^ Walt, studying alliance formation in the Middle East, found that regional balancing and bandwagoning were not closely related to superpower penetration even during the Cold War. He determined that security behavior can best be explained by the dynamic of competitive balancing. The study of military conflict and cooperation in the Middle East remains attractive to security analysts. For a recent look at regional security cooperation during the early 1990s, see Jentleson and Kage (1998), who find that realist factors best explain the trend. 22 See Christensen and Snyder (1990) and Weitsman (1997) on the "tightness" of alliances. 23 Snyder (1990) describes the alliance "games" as burden sharing (concern that costs are borne equally), abandonment (fears that commitments will not be upheld when a state's security is threatened) and entrapment (fears that binding commitments will draw a state into a conflict it would have rather avoided).

PAGE 57

53 Grieco does not believe that relative gains are a significant problem in strongly asymmetric power situations (Hasenclever, et al., 1997). To gain a better idea of what patterns of security cooperation and conflict might be under that condition, we return to Waltz. The main variable left after Waltz's depiction of the international system is the distribution of capabilities within that system. Waltz focused on the different behaviors expected under conditions of bipolarity and multipolarity.^^ Bipolarity was deemed more stable, because of the clarity that such a system offers and the subsequent predictability of great power behavior. Interventions by the United States in Central American and the Caribbean during the Cold War are consistent with this theoretical model. Great powers in a bipolar world also engage in cooperative behavior with smaller states, as great power competition for influence and positional advantage penetrates local conflicts throughout the globe. Cooperative arrangements entail some level of risk, since a prospective partner could defect once conditions change, but great powers can accept risk in their relations with smaller states since local, tactical reversals do not affect the overall balance of power.25 Smaller states do not have this luxury (Waltz, 1979, p. 194), and hence would be unlikely to cooperate with each other in the absence of great power leadership. Hence, the Cold War pattern of security behavior was marked by security 24 Unipolar systems were not specifically addressed by Waltz. But we can infer from his discussions on the disadvantages of world government that he would have predicted them to be unstable because a powerfiil center invites revolution. 25 The longer time horizon of large states with little direct threat to survival fi-om smaller states is an argument that appears in numerous works. See Stein (1990, Chapter 4, esp. 109-1 10) for a representative argument.

PAGE 58

54 competition between the United States and the Soviet Union and security cooperation within each aUied camp. A strong current in the theoretical Hterature develops this point further. Olson (1964) had already shown that under special conditions the provision of collective goods may be possible notwithstanding strong incentives to free ride. If one state, or a small group of states, determined that the provision of such a good provided private benefits that outweighed the costs of provision, then a collective good could be produced. The great powers would bear the costs. Working from this framework to address problems of international political economy, Krasner (1976) and Gilpen (1987) accomplished much theory-building in the area of hegemonic stability and produced strong hypotheses and a compelling research agenda that proved very popular for a time. Under hegemonic stability theory, a large power establishes rules for cooperative behavior for its own benefit and bears the costs of maintaining those rules. This approach is consistent with Walt's view that great powers can accept the costs of short-term tactical reversals. In recent years, this approach has suffered a decline, in part as a result of its own overreaching. As Lake (1993) points out, a great power or small group of great powers may not necessarily choose to exert its influence. In any case, great power behavior can be either benevolent (tolerant of free riding behavior) or coercive (payment extracting). Coercive behavior will depend partly on the resources available for coercion and the likelihood that coercion will be effective. Even if coercion can be cost effective, however, it does not follow that a hegemon will adopt coercive practices. Mastanduno (1997) demonstrates that a hegemon might

PAGE 59

55 choose to provide positive incentives for cooperation even if coercion can be cost effective in the short run. It is possible that a great power such as the United States might choose to underwrite significant security cooperation at this particular junctiire in history. Mastanduno argues that the United States, which enjoys privileges of the "unipolar moment," faces strong incentives to reassure the other states in the system of its peaceful intent. In making this argument, Mastanduno relies on the hypotheses developed by Stephen Walt to explain alliance behavior. In point of fact, Waltz had little to say about security behavior in a unipolar world, which characterizes to a great extent conditions in the Western Hemisphere during the 1990s. Waltz believes that any "unipolar moment" will be short lived because of the propensity of other states to revolt against overweening power at the center. In any case, the kind of security cooperation that Waltz envisioned was simply a byproduct of superpower competition. Absent that competitive dynamic, cooperation with prospective allies should diminish. Mastanduno borrows from Walt to argue that the United States would rationally take steps to appear less threatening to other states, in order to preclude their forming a balancing alliance that would act as a counterweight to U.S. power. Thus the United States could be expected to work through multilateral institutions, promote transparency to signal peaceflil intent and offer security outlets to potential rivals (burden sharing opportunities such as leading roles in peacekeeping operations). While this hypothesis may not square with current U.S. policy in Central Europe and Russia, it offers a theoretically sound prediction of U.S. behavior to enhance security cooperation that is consistent with realism.

PAGE 60

56 A set of security institutions designed by the United States under these conditions might look very much Hke a cooperative security regime. For instance, the United States can faciUtate transparency by providing examples (its own public reporting, for instance) and training ("how-to" instruction on preparation of defense white papers, for instance). The United States can facilitate verification by sponsoring specific mechanisms, such as "open skies" fly-overs, and by participating in third party verification, such as the Military Observer Mission in Ecuador and Peru (MOMEP). In addition, the U.S. traditional guarantee of protection against extra-hemispheric threats may permit lower levels of defense postures. Finally the predominance of the U.S. military model in the hemisphere means that simply by following U.S. standards (protocols, communications, arms, training, doctrine, equipment) which are readily available and essentially cost-free coordination for regional burden sharing is substantially facilitated. Largely, though not exclusively, it is the exercise of U.S. influence over the militaries of the region and the strong presence of the United States in multilateral institutions that is responsible for the emergence of a common "military" language which promotes interoperability a prerequisite for effective burden sharing and collective security action. In the functional language that Keohane (1984) applied to regimes, the United States would be reducing the "transaction costs" which serve as one barrier to cooperation.26 But such a regime would be based on 26 The existence of a common military framework helps solve some "coordination" problems (e.g., coordination of ship passage), but is not enough by itself to forge security cooperation. Coordination solves dilemmas of common aversions (Stein, 1990). Cooperation, practiced to realize common interests, requires policy adjustments to accommodate partners (Keohane, 1984).

PAGE 61

57 and responsive to U.S. security priorities. Krasner (1991), in his study of power and institutions, provides some examples of how those priorities might be expressed in institutional arrangements: (1) determination of who plays; (2) establishment of the rules of the game; and (3) manipulation of payoffs through tactical linkages. In sum, realism maintains that cooperation is especially difficult in security matters, due to the security dilemma, cheating, shirking and free riding, and relative gains considerations. Security cooperation, unless supported by a large power, is only likely as a short-term instrument to balance threats. When the threat subsides, so will the cooperation. While some cooperation is possible through unilateral bargaining strategies of "defensively positioned states" (Grieco, 1988a, p. 507), the degree of cooperation actually taking place will be limited to small, incremental, reversible steps, in issue areas where survival is not at stake and early gains do not translate into large advantages.27 Institutions matter little in this conceptualization. The institutions "governing" cooperative activities in the hemisphere reflect underlying power realities. They are more likely to be arenas of conflict than instnunents to enhance cooperation. States may be involved in "talks," but are more likely to view them as an instrument of power politics 27 See Stein (1990) and Jervis (1999) for a more expansive view of the possibilities of cooperation under anarchy. In his attempt to bridge the gap between realists and neoliberal institutionalists, Jervis includes many cooperative strategies within what he terms "defensive" realism. Institutions, even effective ones, are also in his view consistent with some forms of realism because they are created when states see them as instruments to reap benefits of cooperation. Stein adopts a similar approach, although he is not sanguine about the possibilities of cooperation in security matters. In any case, I do not employ such a broad version of realism, partly because it makes development of contending hypotheses impractical.

PAGE 62

58 than good faith bargaining. In Krasner's view, institutions are "crucial means by which winners win and losers lose" (Moe, as cited in Krasner, 1997, p. 361). Liberal institutionalism, as we will see, is much more sanguine about the possibilities of cooperation. Institutions can be designed to overcome incentives to cheat and address distributive gains problems. Institutionalist Theory The theoretical framework of institutionalism focuses on how multinational institutions help states cooperate. The approach has a long pedigree, which derives from the liberal belief that progress is possible in human affairs, partly through the design of social institutions.28 Classical idealists believe that international organizations, norms, and values are important. If institutions are properly constructed, peace is the "natural" state of affairs. Institutionalism benefited from the work of the functionalists of the 1960s developed by authors as disparate as Mitrany (1966), Haas (1964), Deutsch (1957) and Pentland (1973) who sought to generalize the experience of European integration, hypothesizing that increased cross-border flows would build a web of cooperation due to spill-over and spin-offs from the core economic cooperation, creating a matrix of peace that war could not sever. ^ I use "liberalism" here in the sense that Keohane (1989) applies the term: a school of thought that "stresses the role of human-created institutions in affecting how aggregations of individuals make collective decisions (Keohane, 1989, p. 10)."

PAGE 63

59 In the 1980s, a particular form of institution international regimes became the subject of most the conceptualizing and theory building associated with the institutionalist framework. Hasenclever, et al. (1997) offer a detailed analysis of the principal concepts employed in institutionalist theory, beginning their conceptual review with the consensus definition of "international regime" proposed in 1982 in a special issue of International Organization devoted to theoretical discussion of the concept. In the introduction to that voliune, Stephen Krasner defines regimes as set of "implicit or explicit principles, norms, rules and decision-making procedures around which actors' expectations converge in a given area of international relations." Principles are "beliefs of fact, causation, and rectitude." Essentially, they offer a coherent set of theoretical statements about how the world works in a given area of international relations. Norms are general standards of behavior. Rules are "specific prescriptions or proscriptions for action." Decision-making procedures are "prevailing practices for making and implementing collective choice." Various regimes have been catalogued in terms of their constituent parts. The interstate system itself can be considered as a regime that operates on the principle of sovereignty, a norm of non-intervention and rules/procedures established in the UN charter (Krasner, 1985). Cohen (1983) outlines an international monetary system that operates on the principle of a stable international currency, the norm that states should receive adequate but not unlimited funds for balance of payment problems and rules/decision making procedures as outlined in the charter of the International Monetary Fund. Keohane, in After Heeemonv (1984), defines the international liberal trade regime as one

PAGE 64

60 based on a set of neoclassical economic principles that demonstrate that global utility is maximized by the free flow of goods ... a basic norm that tariff and nontariff barriers should be reduced and ultimately eliminated ... specific rules and decision-making procedures ... spelled out in the General Agreement on Tariffs and Trade. 29 (p.4) While most of regime-based analysis belongs to the field of international political economy, the approach has a strong tradition in environmental studies and international law. Theoretical work in the area of security regimes is much less common, partly in recognition of the power of realism to predict interstate security behavior, although Keohane and Martin (1995), among others, express the hope that regime analysis will "gradually invade security studies" (p. 44).^^ Security regimes have been "discovered" in the area of strategic nuclear weapons reduction, the "European military order," prevention of nuclear war, and nuclear non-proliferation.^^ Efforts to categorize the theoretical insights developed within the institutionalist paradigm are legion. Haggard and Simmons, working specifically with the concept of international regimes, organize variants by their theoretical component. From their The application of institutionalist theory to explain international trade behavior is perhaps more complete than in any other area of interstate relations. The scholarly debate on the topic emerged in the context of the declining U.S. economic hegemony in the 1970s. Institutionalists argued that what they saw to be an enduring practice of fi-ee trade despite the U.S. decline was linked to the presence of institutions and norms. Critics disagreed both with the institutionalist characterization of trading practices as largely free and with the claim that such behavior was in any case due to the impact of institutions. See Keohane ( 1 984). 30 Axelrod and Keohane (1986) advocate a "single framework" for militarysecurity and political-economic issues (Axelrod and Keohane, 1986, p. 227). See Jervis (1983) for the classic explanation of "why security is different." See Mueller (1993), Tate (1990) and Brzoska (1992).

PAGE 65

61 analysis, four basic forms emerge: structural, which links regime creation and maintenance to hegemonic power (Gilpin, 1987); functional, which claims that regimes play a facilitating role that reduces the transaction costs associated with formulating agreements (Keohane, 1984); game theoretic, where regimes help structure the context in which states make strategic choices in pursuit of their preferences (Stein, 1990); and cognitive, which is strongly related to the conditioning role that norms/ideas have on behavior (Ruggie, 1983 and Haas, 1964). Institutionalism, as currently employed (Keohane, 1 989), contains strong traces of this prior work. Its principal characteristic is its view that institutions have independent causal significance rather than simply reflecting power relationships. ^ 2 Thus, hegemonic stability theorists, who see institutions simply as creations of powerful states, are not liberal institutionalists. However, the functional notion that institutions can help states reduce costs associated with forging cooperative agreements, as well as the game theoretic concept that institutions affect "pay-offs" for cooperation and defection, are central to what Keohane (1989, 166) calls the "rationalistic" study of institutionalism, commonly referred to a "neoliberal institutionalism." Cognitive regime theorists have contributed to what Keohane refers to as "reflective" institutionalism, which is distinct from neoliberalism in that it operates below the system level, where unit-level processes (such as the formation of preferences) are important variables. As Jervis (1999) notes, it is not enough to show that cooperation and the presence of institutions are correlated, as they most certainly are (see work of Russett, Oneal, and Davis, 1998). Institutionalism must demonstrate causality.

PAGE 66

62 Neoliberal institutionalism, operating at the system level, accepts the neorealist assertions that states are the most important actors of the international system and that they act rationally in their interactions with each other. However, the assumption that the distribution of capabilities is the most important system-level variable is sharply questioned. Neoliberal institutionalism holds that "deeply embedded institutions are as fimdamental to world politics as are the power resources of units" (Keohane, 1989, pp. 89). Institutions change the context in which states operate (a game theoretic notion) by facilitating communications so that agreements can be made (a fimctional notion), by providing information to reassure states that agreements are being respected, by institutionalizing norms so that expectations converge around rule-based behaviors, and by linking compliance to other areas of concern to states so that the costs of defection are raised. These institutionalized settings facilitate cooperation in two primary ways. First, they reduce uncertainty by providing information that can be considered in a state's decision making process. Second, they reduce the transaction costs of entering into negotiations by providing an arena for communication. Institutionalists hold that even in the absence of changes in the international distribution of power or the preferences of actors, information can "profoundly affect international political behavior" (Keohane, 1986, p. 19). In the area of security cooperation, information is especially valuable because it avoids worst-case analysis (Keohane and Martin, 1995; Jervis, 1986). In fact, Keohane (1984) sees the provision of high quality information as a main function of these institutions.

PAGE 67

63 If agreements are to be made on the basis of this information, however, its relevance and reliabiUty are critical. Information about the resources and formal negotiating positions of potential partners is not sufficient. What is needed is information about their "internal evaluations of the situation, their intentions, the intensity of their preferences" (Keohane, 1989, p. 118). That information must be verifiable, through formal inspections, visits, exchanges, data cross-checking, or other mechanisms (Nye, 1991; Oye, 1986; Anthony, 1993). Institutions can provide this service by levying reporting requirements on members and establishing rules for verification. Institutions also help verification through "explicit codification of norms," making it easy to evaluate the degree of compliance (Oye, 1986, pp. 16-17). Since these mechanisms taken together greatly facilitate the operation of reciprocity, they are critical to agreements regarding reductions in defense postures and burden sharing. In addition to providing information, institutions facilitate the making of agreements in a variety of ways. In the first place, they provide a forum, or "constructed focal poinf for negotiation (Keohane and Martin, 1 995, p. 45), which increases opportunities to negotiate (Nye, 1991). This allows bargaining to proceed in an iterative fashion, promoting the tit-for-tat pattern of cooperation. Furthermore, institutions can enhance communication by standardizing signals through the adoption of standard formats and reporting calendars. These steps make it easier for states to reach agreements regarding cooperative security. They also create the expectation that states will comply with security rules. This is because institutions formalize standards for behavior and create the means to monitor compliance and evaluate state reputations (Hasenclever, et al., 1997). As long as

PAGE 68

64 states "discount the value of agreements on the basis of past performance" (Keohane, 1984, p. 105), a reputation for compliance is valuable due to the prospect future gains.33 The costs of defection are raised because of the existence of institutions, just as the transaction costs of cooperation are lowered. Institutions that can be effective in this sense must have the administrative capacity to support negotiations and collect, analyze and report data. These are essentially performance criteria, which can be observed and measured in security institutions. Institutions rarely operate in a vacuum. They may overlap issue areas with other institutions (Keohane, 1 986), creating multiple links between states that connect, for instance, rules on cooperative security with regional mechanisms to promote respect for human rights, encourage economic development and strengthen democratic governance. In the broadest sense, the effect of these issue linkages is to create a dense web of constraints that fortifies rules and makes rule-based behavior more likely. More specifically, linkage is important to institutionalist theory in several concrete ways. First, linkage allows for more iteration, increasing the opportunities for "tit-for-tat" reciprocation (Oye, 1986). Second, linkage may erode hierarchy, by allowing small states to link security issues to other areas (Keohane and Nye, 1989). Third, linkages increase the benefits of cooperation and the costs of defection. They "allow for ^^ Institutions can also reward cooperative behavior through provision of leadership positions (president of the OAS Committee on Hemispheric Security, for example) and sponsorship of security conferences.

PAGE 69

65 more effective retaliation against cheaters and create scope for mutually-beneficial exchanges" (Keohane and Martin, 1995, pp. 49-50). This last point touches on the globalization literature,^^ which resurrects some of the early thinking regarding the beneficial impact of trade (Viner, 1944). Oneal et al. (1996) note that Kant, in his treatise Perpetual Peace, asserted that "economic interdependence reinforces constitutional constraints and liberal norms by creating transnational ties that encourage accommodation rather than conflict" (Oneal et al, 1 996, p. 12). This proposition is contrary to the realist view that interdependence creates vulnerability and heightens concerns over distribution of gains, thereby increasing the propensity for conflict (Waltz, 1979). Realists see this as especially valid in cases of asymmetrical interdependence, such as Latin America (Hirschman, 1945; Sanderson, 1992). 35 Empirical tests of the "perpetual peace" proposition have failed to sway proponents of either side of the debate, in part due to disagreements over testing Barbieri and Schneider (1999) note that "the term 'globalization' has largely superceded the term 'economic interdependence' to describe the rapidly growing links between nations, economics, and societies" (Barbieri and Schneider, 1999, p. 387). For two recent critiques of the liberal argument that free trade leads to peace, see Rowe (1999) and Ripsman and Blanchard (1996/7). 35 Mares (1997) argues that economic globalization will not change the nature of security relations in the hemisphere: "economic growth ... in the absence of a stable and credible balance of power" is likely to lead to "increased conflict, rather than the first steps toward integration or a pluralistic security community" (Mares, 1997, p. 215).

PAGE 70

66 methodology.^^ Even if agreement were possible on the methodology, however, results showing simple correlation are not convincing, since causal connections are not clear. Gowa (1989) finds for instance that because of the security externalities associated with open borders, free trade is more likely between allies than adversaries. It her analysis, the security environment determines trade relations, not vice versa. Even if institutions are able to craft linkages between security and trade that expand potential gains, the struggle over distributional gains still needs to be addressed. Despite claims that institutionalism has ignored relative gains (Grieco, 1988a, Mearsheimer, 1994/5), Keohane and Martin (1995) assert that situations where relative gains considerations are important offer an especially productive area for institutions: Liberal theory argues that institutions provide valuable information, and information about the distribution of gains from cooperation may be especially valuable if the relative-gains logic is correct. Institutions can facilitate cooperation by helping to settle distributional conflicts and by assuring states that gains are evenly divided over time, for example by disclosing information about the military expenditures and capacities of alliance members. (Keohane and Martin, 1995, pp. 45-46) Thus, the mechanism of reciprocity, facilitation for negotiation of side agreements, and the presence of linkages all help institutions deal with distributional problems. Institutions do not always have a strong impact on behavior. According to Hasenclever, et al. (1997), institutions differ in terms of their robustness (the extent to which institutions resist change when the interests of the most powerftil member diverge from the norms) and effectiveness (the extent to which actual behavior can be See the recent debate in Journal of Peace Research (vol. 36, no 4), especially Barbieri and Schneider (1999) and Oneal and Russett (1999). Also, see Morrow (1993).

PAGE 71

67 characterized as rule-based). But the important point is that while institutions may reflect the distribution of power and the interests of their members, they have an independent impact on behavior and mute the impact of power over outcomes (Keohane and Nye, 1989). Cooperation between states, acting as unitary, rational actors attempting to maximize interest in an anarchic, self-help system, is possible: (1) because states are interested in attaining benefits of reduced tension and enhanced security cooperation and (2) because institutions help them overcome the dilemmas of defection free riding and cheating and help them negotiate solutions to the distributive problems that relative gains present. How might this work in situations of asymmetry? The kind of cooperation described under realism in which a large power establishes institutions as tools to achieve its own ends is a possibility admitted by neoliberalism. In fact, where mutuality of interests is high, it may be difficult to find significant differences in the maimer in which these kinds of institutions evolve. But the politics of security cooperation imder asymmetry are likely to be more complex. Keohane and Nye (1989) provide a set of potential outcomes that differ sharply from realism. While these were constructed for a specific context what the authors called "complex interdependence" 1 argue that the effects they predict are evident in cases where a large state practices a policy of reassurance (as described by Mastanduno) that opens negotiating room for smaller states. The United States may opt for a multilateral approach as a cost-effective solution that simultaneously reassures its security partners and engenders the cooperation it seeks. However, such a strategy is likely to produce perverse outcomes over time. This is because the institutional forum it promotes restores symmetry by offering smaller states

PAGE 72

68 an opportunity to voice preferences, build coalitions and reinforce negotiating positions through linkages.37 Small states can leverage these advantages to engage in balancing behavior that would not be possible in the absence of the institution. Furthermore, in order to build support for its agenda, the United States may need to provide side benefits and security outlets which ultimately strengthen the bargaining position of the smaller states. While an equilibrium may be theoretically possible, a more likely outcome is an eventual retreat from multilateralism, leaving the institution void of resources and effectiveness. Mutuality of interests declines and the shadow of the future shortens as smaller states come to expect a return to unilateralism on the part of the United States. The determinants of U.S. security behavior are largely beyond the reach of hemispheric institutions. The primacy of global interests and the impact of the domestic agenda are clearly more important to the formulation of U.S. security policy than regional concerns. When global concerns dominate, support for regional cooperative security and respect for regional rules take a backseat to other strategic priorities. When domestic concerns dominate, U.S. offers of assistance (arms, training, resources, information) can be subject to intrusive end-use agreements which fall outside of the regime. The certification process which oversees provision of counterdrug assistance and the Leahy amendment which makes military assistance contingent on human rights performance are two examples. 37 Jervis (1999) claims that institutions, even if created by the system's most powerful states, can erode power by giving "voice, legitimacy and forms of influence to weak or new actors" (p. 61). Institutions could restore some symmetry to U.S.-Latin American relations. This finding can also be derived from Krasner (1985), although his argument is a realist one.

PAGE 73

69 The ever-present threat of unilateral U.S. action and the knowledge that institutional constraints are insufficient to prevent such action undermines confidence that other states in the hemisphere have in hemispheric institutions. Eventually they will decay. But along the way, an agenda of security issues and related rules that is substantially different from U.S. power interests is possible. By contrast, realism rejects the possibility that institutions can have significant effects and denies them any independent impact on state behavior. Even if institutions facilitate agreements and make information available, that does not mean agreements will actually be consummated or respected when interests clash with regime norms. In the ultimate instance, regimes cannot guarantee survival. In a self-help system, what matters is a state's relative position. Created in the interests of their most powerful members, institutions will not constrain behavior and will be abandoned when no longer convenient. What happens when we move from the system level and look at preference formation within member states? This is the focus of the "reflective" version of institutionalism, reviewed only briefly here, since we will not return to it until the conclusion of the study when we consider the limitations of systemic approaches. Keohane (1989) offers this summary of the reflective approach: "Institutions do not merely reflect the preferences and power of the units constituting them; the institutions themselves shape those preferences and that power. Institutions are therefore constitutive of actors as well as vice versa" (Keohane, 1989, p. 162). "Reflective" approaches encompass "constructivist theory" (Checkel, 1998), "critical theory" (Mearsheimer, 1994/5) and "grotian theory" (Krasner, 1983). These labels in turn lump together a wide

PAGE 74

70 variety of approaches. Wendt (1995), for instance, identifies "postmodernists (Ashely, Walker), constructivists (Adler, Kratochwal, Ruggie ... Katzenstein), neo-Marxists (Cox, Gill), feminists (Petersen, Sylvester) and others" (Wendt, 1995, pp. 71-72) as belonging to critical international relations theory. The approaches are united by a belief that structures are sociological constructs that shape actors' identities and interests, rather than just behavior. Hasenclever, et al. (1997) usefully divide the reflective approach into weak and strong cognitivists. Weak cognitivists believe that institutions can change preferences through learning gleaned from the experience of cooperation. Strong cognitivists believe that the institutions themselves are deeply embedded constructs that form actors' identities.3^ Both of these versions of institutionalism delve into the ideas that statesmen use to form policies regarding security competition and cooperation. New ideas can cause normative change and lead to corresponding changes in behavior (Goldstein and Keohane, 1993). Thus, the penetration of new ideas pertaining to appropriate roles for the armed forces under democratic regimes could give rise to strong proscriptions on use of force and restrictions on military activity. See March and Olson (1998) for a recent review of this approach. 39 This argument is tangentially related to the debate over the effect of democracy on interstate conflict. The observation that democratic dyads tend not to fight one another has been thoroughly studied, although there is still debate as to whether common interests or common polities explain the phenomenon (see Chan, 1984, and Farber and Gowa, 1995, for opposing views). Within the camp of those that believe democracy makes a difference, there is disagreement over whether democratic norms of peacefiil resolution or democratic mechanisms of leader accountability explain the phenomenon (Starr, 1997). Recent work on democratization shows that it may positively correlate to conflict (Mansfield and Snyder, 1995).

PAGE 75

71 Weak cognitivists hold that institutions contribute to this process both by providing "opportunities for contacts which may contribute to learning" (Nye, 1991, p. 1 57) and through dissemination of new information which changes preference formation (Keohane and Martin, 1994/5). Haas (1964) found that successful integration in Europe was partly due to a learning process experienced by technocrats. Success in one area "spilled over" into others as eHtes expanded the process of economic integration. Adler (1988) looked at the power of ideas and learning by technocratic elites in the area of nuclear power policy in Argentina and Brazil. Hasenclever, et al. (1997) describe an interesting extension of this idea, based on the work of Putnam and Zum. Introduction of domestic institutions creates a two-level game actors within the state apparatus bargain with each other and with counterparts outside the state simultaneously. This relates to the "second image" and "second image reversed" logic that frequently appears in the security literature. For instance. Mares and Bernstein (1999) argue that Argentina's high level of participation in peacekeeping operations is promoted by the government as a way to provide the military with an external mission. Laurance (1996) blames slow progress in arms control in Latin America to the pressure that the military, anxious to protect its budget, exerts on the process. Jervis (1999) notes that arms control agreements favor "doves," strengthening their hand in internal negotiations. There is a vast body of literature related to this theme in Latin America, especially dealing with the two domestic institutions we are most concerned with the military and civilian sectors of government. The literature on corporatism, authoritarianism and transitions and consolidations to democracy are well developed in the field of Latin

PAGE 76

72 American studies. Clearly the more general literature on the relationship of democracy and war, the role of domestic institutions in other areas of international relations (particularly trade) and studies linking military culture and strategic doctrine are also valuable. These issues will be discussed fiirther in the concluding chapter. Returning to the neoliberal institutionalism, we can now readdress the issues posed by the cooperation dilemmas outlined in Table 2-1 We will look to see if institutions facilitate cooperative behavior in each of the behavior clusters without regard to the institutional politics described for asymmetrical situations. Behavior there is driven by the twin dynamics of reassurance and institutional balancing and not necessarily a search for the absolute gains promised by cooperation, thus institutional design may be only partially effective in addressing cooperative dilemmas. Since military transparency can expose free riding, cheating and relative gains problems, it is fundamental to institutional attempts to promote security cooperation. The challenge is to verify military information through various mechanisms, designing reporting formats that can be easily cross-checked and building an extensive network of contacts that produces significant information flows. Institutions can promote reductions in defense postures by addressing distributive gains through safeguards, and trade-offs in types of reductions. But, as we shall see, there are no hemispheric or regional regimes that include these measures as yet. Relative gains are normally not addressed. Will that matter?40 w^i^h regard to burden sharing, again transparency exposes cheating and These kinds of rules, which have been present in the international trade regime for many years, include safeguards (related to balance of payment crises) and exclusions (agriculture, textiles, etc.). See Hasenclever, et al. (1997).

PAGE 77

73 relative gains problems. Free riders may enjoy benefits of enhanced security but are denied benefits of interaction. These points are summarized in the table below. Table 2-2: Cooperative Problems Regime Solution. Security Behavior Absolute Gain Regime Solution Transparency Better security calculation avoids worst case analysis Reliance on reciprocal verification. — importance of reciprocal visits/exchanges. ~ designing formats that can DC easily cross cnecKea. Reduction of Defense Postures More resources available for economic development and other goals Transparency exposes free riding, cheating and relative gains problems. Institutions can address distributive gains through safeguards, and trade-offs in types of reductions. Regional Burden Sharing for Regional Security More efficient allocation of security resources Transparency exposes cheating and relative gains problems. Free riders may security but are denied benefits of interaction. Institutions can promote equitable sharing of costs and benefits, such as rotation of leadership roles, cost sharing for exercises, negotiation of side benefits.

PAGE 78

74 Contending Hypotheses Hemispheric Level At this point, we are ready to summarize contending hypotheses for security cooperation from the two theoretical frameworks discussed above realism and institutionalism. We will formalize hypotheses that address organizational characteristics, rule creation and security behavior in each of the clusters examined. In this section, we focus only on the hemispheric level. Subregional hypotheses are developed in the following section. As we shall see in more detail in Chapters 3 and 4, there is no denying the substantial institution building process in the area of hemispheric security that has been underway since the advent of WWII. Old hemispheric security organizations have been substantially renovated and several new ones have been created during that period. But are they sufficiently resourced by member states to effectively facilitate negotiations and carry out the functions of regime management? Realist thinkers doubt that possibility, except where such an arrangement might serve the interests of the most powerful states. A realist hypothesis would make resourcing contingent on an agenda consistent with the new, post-Cold War security agenda of the United States. Thus institutions should be largely underwritten by the United States and generally supportive of the U.S. security agenda, notwithstanding small deviations that do not threaten core U.S. security concerns. Other countries are likely to participate only in areas of common interests. Institutionalism may not differ markedly from this prediction, since no hemispheric institution or regime could be expected to resist U.S. influence. However, even under conditions of marked power asymmetry, institutions may demonstrate some

PAGE 79

75 autonomy. Through issue linkage and the legal principle of formal equality of members, institutions work to reduce asymmetry and promote a more representational agenda that may confound large power interests. Sharper distinctions are possible with regard to the rules that these institutions promote. Grieco (1988a) offers a set a contending hypotheses for international agreements (rules), built around the problem of relative gains. These hypotheses can be fairly easily adopted to our study of patterns of security cooperation in the Americas. In the first instance, Grieco notes that under realism, states will in general eschew cooperative arrangements with binding commitments that present high exit costs. This should be particularly true in the area of security, where relative gains problems are most acute. Thus realism would expect that hemispheric rules regarding transparency, verification, defense postures and burden sharing would be voluntary and non-intrusive. In the absence of reliable verification, the effectiveness of reciprocity will be limited. This contrasts with agreements made between the United States and its security partners, such as the network of mostly bilateral agreements for tracking and interdicting illicit drug production and trafficking that exists between the United States and source and transit countries. As rules become more inclusive, they will lose their specificity. Under institutionalism, theory tells us that states, given the opportunity to negotiate, will tend to insist on binding rules, with strong verification, whenever relative gains may present an obstacle to cooperation, so that the potential absolute gains can be "locked-in." These steps are not coercive; member states are simply binding themselves to reassure partners of their cooperative intent.

PAGE 80

76 We can derive some realist predictions for specific behaviors in transparency/verification, defense postures and burden sharing as well. Transparency in military matters should be limited to information that is already public. Reductions in defense postures should be based on threat perception, not regime rules. And burden sharing, unless sponsored by the United States or another major power, should be extremely tentative. In states that share the kind of cross-border problems that we have dubbed transnational threats, burden sharing should be even more difficult because of the added source of conflict. Institutionalists, by contrast, would expect that relevant and reliable information will emerge from institutional arrangements. Institutions with the administrative capacity to handle the information will add considerably to the degree of transparency in the system. The increase in transparency will allow reductions in defense postures to proceed in a reciprocal manner, where levels might be higher than threats warrant. Under these conditions, transnational security problems should attract cooperative behaviors, since absolute gains to such cooperation are relatively high. Again, transparency will insure that no state benefits unduly from such cooperation. Side negotiations facilitated by institutions can redress distributional problems that might emerge. Finally, with regard to institutional effectiveness, realism would contend that rules will not be effective outside of the U.S. agenda. Patterns of cooperative security behavior should conform to U.S. priorities. Participation will not be inclusive, but rather concentrate on key U.S. security partners. This pattern may attain even in the absence of security institutions, since the United States will pursue its own agenda bilaterally when multilateral approaches do not advance its interests. Finally, fi-om Mastanduno, realism

PAGE 81

77 would predict that the United States is Hkely to act with restraint towards more powerful Latin American states in order to avoid eliciting a balancing reaction. Institutionalism expects to find rule-based behavior not behavior driven by balancing or U.S. influence. Bolstered by expectations of convergence (supported by binding rules and strong verification schemes) and eased by the facilitation that institutions provide (reduced transaction costs), states will tend to follow the rules established in organizations formed for that purpose. However, if acts of reassurance are insufficient to maintain cooperation, then the dynamics of institutional politics will predominate. These contending hypotheses are summarized in Table 2-3 and Table 2-4. Table 2-3: Contending Hypotheses at the Hemispheric Level for Institutions Dependent Variables Realist Hypotheses Institutionalist Hypotheses Security Institutions: Given institutions as "constructed focal points" for negotiations and absolute gains from military transparency, reduced defense postures and burden sharing, under conditions of marked power asymmetry Organizational Features 1 Institutional agendas will not stray significantly from U.S. interests and will be resourced only where interests coincide. Issue linkage will complicate cooperation. 1 Institutional procedures and issue linkage will restore symmetry, providing opportunities for institutional balancing. Agenda will be multilateral, focused on issues where absolute gains are evident. Rule Creation 2. Rules will not be binding and comprehensive, but rather voluntary and partial. 3. Rules will not be supported by verification mechanisms sufficient for reciprocity. 2. Rules will be binding wherever the potential for defection exists. 3. States will insist on strong verification mechanisms so that reciprocity can fimction.

PAGE 82

78 Table 2-4: Contending Hypotheses at the Hemispheric Level for Behavior Dependent Variables Realist Hypotheses Institutionalist Hypotheses Security Behavior: Given institutions with the administrative capacity to carry out the functions of regime management (data collection, analysis and reporting) and rules promoting transparency, defense reductions and burden sharing .... Transparency, Defense Postures, Burden Sharing 4. Transparency measures will not reveal information of value, reporting instead information that is already largely public. 5. Defense reductions are unlikely and in any case will reflect perceived threats. 6. Burden sharing is unlikely unless sponsored by the United States or resulting from balancing and/or bandwagoning. The presence of transnational security problems makes cooperation even less likely because of the added potential for relative gains. 4. Transpeirency will be considerable, providing relevant and reliable information. 5. Defense reductions will take place where transparency provides assurances of reciprocity. 6. Burden sharing will take place where transparency provides assurances of reciprocity. The presence of transnational security problems makes cooperation more likely because of the added potential for absolute gains. Subregional Level What modification of these hypotheses might be necessary at the subregional level? There is good reason to believe that the basic tenets of realism and institutionalism will operate there with little modification. Under realism, the dynamic of security

PAGE 83

79 competition should not be any less salient. If anything, security competition will be more intense at the subregional level where fundamental security dilemmas operate and where threats are more proximate to use Walt's terminology. Under institutionalism, the dynamic of security cooperation should also be more intense, since neighboring states are likely to share compatible security interests especially transnational security problems leading to the possibility of greater absolute gains through cooperation. The fact that both security competition and cooperation are more intense has an impact on subregional institutions if they are to be effective, they must be more binding in nature than their hemispheric counterparts. There are other reasons why two theories should work well at the subregional level. The question of multiple linkages becomes intensified at the subregional level, both because neighboring countries often share transnational concerns over a wide range of issue areas and because subregional institutions are normally nested in their hemispheric counterparts. These factors taken together mean greatly amplified connectivity between states in a given subregion. Of course, subregions may differ in terms of the density of their interrelations, but this factor simply adds another testable hypothesis to a comparative analysis. Grieco (1988a) offers an additional comparative test of the relative explanatory power of realism and institutionalism that fits nicely within the subregional fi-amework. He bases this comparative test on the "number of partners states prefer to include in a cooperative arrangement" (Grieco, 1998a, p. 506). According to his review of the theories, advocates of institutionalism assert that a "smaller number of participants facilitates verification of compliance and sanctioning of cheaters." Thus, Grieco points

PAGE 84

80 out, states that can choose the context of cooperative arrangements will normally prefer to limit membership. Realism, according to Grieco, predicts the opposite, since arrangements with more members provide more partners and hence more opportunities to offset relative gains advantages. For these reasons, realism would expect that cooperative arrangements at the subregional level will be even more difficult to accomplish than at the hemispheric level.'^^ Institutionalism predicts precisely the opposite pattern. In sum, institutionalism expects more institution building and more binding rules at the subregional level than the regional level. Realism predicts the opposite pattern. In any case, both paradigms are operative at this level of analysis. What then about the effectiveness of rules regarding security cooperation? Institutionalism predicts that security behavior will be rule-based, following the logic laid out in the previous section. Transparency will be critical because of its important role in commimication. Reductions in defense postures and burden sharing are likewise important because of the absolute gains they promise. Realism, on the other hand, expects that patterns of security behavior will follow the logic of balancing; cooperation will be limited to informal ententes formed to coimter subregional threats. We can approximate the predicted pattern through balance of threat analysis for each subregion. "Defensive realists" who, according to Jervis (1999), adopt many of the standard "cooperation under anarchy" strategies, would disagree with Grieco's formulation. The security dilemma/collection action literature highlights the difficulties of large alliances, since detecting cheaters and enforcing obligations is difficult with large groups. But Grieco is not focusing here on cheating, but rather on the relative gains problem, which is more intense in smaller groups.

PAGE 85

81 What of U.S. power? The United States is still an important security externality that affects subregional security behavior. But it may be possible to abstract some of that impact from the security equation by adopting a comparative approach that selects subregions differing in terms of their relationship with the United States. In which context