A Structural Analysis of International Conflict From a Communication Perspective.pdf

International Interactions ISSN: 0305-0629 (Print) 1547-7444 (Online) Journal homepage: https://www.tandfonline.com/loi/gini20 A Structural Analysis of International Conflict: From a Communication Perspective Jang Hyun Kim & George A. Barnett To cite this article: Jang Hyun Kim & George A. Barnett (2007) A Structural Analysis of International Conflict: From a Communication Perspective, International Interactions, 33:2, 135-165, DOI: 10.1080/03050620701277764 To link to this article: https://doi.org/10.1080/03050620701277764 Published online: 17 Apr 2007. Submit your article to this journal Article views: 594 View related articles Citing articles: 3 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=gini20 International Interactions, 33:135–165, 2007 Copyright © Taylor & Francis Group, LLC ISSN: 0305-0629 DOI: 10.1080/03050620701277764 A Structural Analysis of International Conflict: 1547-7444 Interactions 0305-0629 GINI International Interactions, Vol. 33, No. 2, March 2007: pp. 1–34 From a Communication Perspective JANG HYUN KIM and GEORGE A. BARNETT Communication J. H. Kim and G.and A. Barnett International Conflict Department of Communication, State University of New York at Buffalo, Buffalo, New York, USA The study describes the structure of international conflict with the tools of network analysis to enhance the understanding of multilateral conflict-communication relations and to predict the conflict structure with existing international relations theories (liberal and realist) plus global communication variables. Using data obtained from the Corre- lates of War Project (http://cow2.la.psu.edu/), the structure of interna- tional conflict is described for the period 1993–2001 for 145 nations. The results indicate that this network is very sparse; 42 nations had no conflict, and 36 only one bilateral disagreement. The network is cen- tered about former Yugoslavia (Serbia and Montenegro), Russia, the United States, Iraq, and China. Most conflicts are regional. The paper also evaluates both the liberal, expanded liberal (with communica- tion variables included), and realist (including Huntington’s Clash of Civilizations Theory) perspectives as predictors of conflict. The results indicate that communication variables substantially enhance explan- atory power of a predictive model, but the effects of the communica- tion variables are inconsistent. A multiple regression model including history of colonialism and prior conflict, physical proximity and contiguity, whether or not a nation is a democracy, and the com- munication variables—international telecommunication, freight, and exports—accounted for 30.0% of the variance in the structure of international conflict and each variable was significantly related to conflict. The need for further research is discussed. KEYWORDS communications, conflict structure, social network analysis The earlier version of this paper was presented to the International Communication Association, New York, NY, May 26–30, 2005. Address correspondence to Jang Hyun Kim, Department of Communication, State Univer- sity of New York at Buffalo, 359 Baldy, Buffalo, NY 14260. E-mail: [email protected] 135 136 J. H. Kim and G. A. Barnett Since the time of Immanuel Kant and Adam Smith scholars have exam- ined the causes for conflict between nations. Theories about the causes of war generally fall into two perspectives, the liberal and real- ist. The liberal perspective suggests that interdependencies between nations, typically trade, act as deterrents to war. The realist point of view suggests that the power relations both within and among nations are the best predictors of conflict. 1 Past international relations research indicates that while both of these paradigms significantly predict con- flict, neither provides particularly good predictions in current global circumstances. This study points out two major problems in conventional conflict studies and suggests a new approach. The first problem is that past stud- ies didn’t examine the structure of conflict, although they argued that they have been doing so. This study presents network analysis as a rem- edy to the problem. Network analysis is a set of research procedures for identifying structures in systems based on the relations among the sys- tem’s components rather than individual attributes. Existing monadic and dyadic approaches have failed to describe the multilateral nature of international conflict. For this study, the “network” is the pattern of bilat- eral conflicts among the nation-states that compose the global system. The other salient problem is that existing studies, especially those from the liberal perspective, have failed to include communication variables to predict international conflict. They focused only on trade, which is just one aspect of international interaction. This study argues that adopt- ing international communication variables can enhance the prediction of conflict between nations. Specifically, it expands the liberal definition of the interdependency between nations beyond trade to include the com- munication and cultural relations. This approach enables us to grasp the multiple effects of interdependency, departing from the trade-only posi- tion in liberal scholarship.2 This study makes more contributions to the field. One, it examines conflict between nations in the post-Cold War era. The end of the Cold War constitutes a critical moment in the history of global conflict and communi- cation. Two, it explores, in depth, the internal conditions within nations that lead to the initiation of conflict, such as cultural identity and radical political regime change. With the structural description of international conflict, this study con- structs a model that predicts conflict with elements from both the liberal and realist perspectives along with recent research from international and cross- cultural communication. The model may be applied to enduring conflicts including North and South Korea, and can be disseminated to policymakers, scholars, the media, and the general public to help in the understanding of international conflicts and to encourage those behaviors to prevent future conflicts. Communication and International Conflict 137 A REVIEW OF CURRENT APPROACHES TO INTERNATIONAL CONFLICT AND COMMUNICATION A structural approach is necessary to analyze international conflict (Hart, 1974). However, its utility has been overlooked. “Structural,” in this context, stands for a newer approach than existing monadic or dyadic analysis.3 Bremer (2000) argues that designating an enemy state is not based on a ran- dom selection. Thus, a dyadic approach is useful to predict who will fight whom. However, a one-to-one relationship (dyads) does not denote “struc- ture.” If nations A and B, B and C, C and A are in conflict with each other, the triangular structure is not considered in existing one-to-one approach. Only network analysis can see the role of each nation in multilateral complexity of international conflict by different measures like centrality and density. It should be noted that even the “neorealists” who emphasize the structure of the inter- national system look at only one-to-one relationships. International relations are composed of numerous interactions in which nations realize their intention to influence other states. Also, nongovernmental exchanges such as tourism and telecommunication affect the international system (Hamelink, 1994; Barnett and Lee, 2002). Hart (1974) defines the structural approach as a way to analyze international influence that relies on the relations between dyads. For Hart, influence indicates “a relationship between a pair of social units” (1974, p. 142), and conflict may be predicted by these influences. A structural analysis can be used to quantify the influences of each social unit and to show that a slight change in dyadic influence can lead to a greater effect on the interna- tional political regime (Hart, 1974). This study employs social network analysis (SNA), a method of finding structures in systems based on the relations among its components, as an alternative to conventional research. Network analysis is required because modern military conflicts are dis- putes among networked agents. This is why Ronfeldt and Arquilla (2001) described such trends as “netwars” (2001, p. 311) among networks. With the diffusion of communication technology, state as well as nonstate actors are equipped with computer networks. First, networked states are confronting potential threats from their state-level competitors who are defended by robust-to-nuclear networks such as Internet. Second, they also have to con- front nonstate actors such as nongovernmental organizations and violent extremists. Netwars are coexisting with the post-Cold War era, which is the time full of conflicts between nationalism and globalism. This paper first addresses two contrasting perspectives in interna- tional relations, the liberal and neorealist, and describes the relationship between communication and these orientations. Classical liberals have stressed mutual cooperation and interdependence as a structural trait of international relations. In spite of the anarchic nature of the international community, progress toward less conflict within the global society is possible through interdependence (Kegley and Wittkopf, 2006). Adam 138 J. H. Kim and G. A. Barnett Smith and his followers stressed the peacemaking role of economic rela- tions (Oneal and Russett, 1997). Smith’s trade-centered position has con- tributed to a tradition that focuses on international transactions as a solution for international conflict. Kant (1795) enumerated three sources of international peace: republican constitutions (regime), international organizations/law, and trade. Communication can elaborate the liberal position. Liberals have focused only on trade, international organizations, and democratic regimes. They have overlooked the effect of broader international communication such as transnational information flows, transportation, and mail exchanges. International communication helps to create trust among nations that leads to collective identity and solidarity, and the reduction of mutual mispercep- tions (Harrison, 2000). This benefit is important because it clarifies the ques- tion that the liberals didn’t solve, namely, why international conflict may be prevented by communication. The decrease of uncertainty in the interpreta- tion of the actions of nations through communication is another rationale to adopt communication variables (Harrison, 2000). Deutsch (1966) also postu- lated that the analysis of international communication would reveal the glo- bal political structures because it is related to the way nations generate and select information based on their cultures and interests. The liberal perspective suggests many variables that predict interna- tional conflict. Dyads composed of democratic nations (joint democracy) have been shown to have fewer disputes (Maoz and Abdolali, 1989; Bueno de Mesquita and Lalman, 1992; Lake, 1992; Starr, 1992; Russett, 1993; Ray, 1995; Rummel, 1997; Weart, 1998), even controlling for such variables as contiguity, alliance ties, economic development, and balance of power (Oneal and Russett, 1997). However, Mansfield and Snyder (1995, 1996) stipulate that “the democratic peace” implies that there is a higher chance of peaceful relations mainly among mature democracies, but that the process of democratization is vulnerable to international conflict. A nation’s degree of democratization has been suggested as a predictor of international con- flict. Further, Oneal and Russett (1997) found that regime difference between nations is a stronger predictor of conflict than an absolute mea- surement of a nation’s degree of democratization. On the border between liberals and realists, Keohane and Nye (1998) argued that transparency of democratic regimes is positively associated with information revolution and telecommunication technology influences domestic/international politics. Economic interdependence, measured as trade, is inversely related to the average number of wars (Mansfield, 1994; Oneal and Russett, 1997). Trade can decrease the chance of war by enhancing the range of decisions in a cri- sis (Morrow, 1999). Oneal and Russett (1997) argued transnational trade rela- tions imply that individual and national level communication channels exist and act as potential inhibitors of international conflict. Also, Mansfield and Pollins stated: Communication and International Conflict 139... economic intercourse increases contact and promotes communica- tion between private actors in different countries as well as between governments. Increased contact and communication, in turn, are expected to foster cooperative political relations (2001, p. 836). Similar studies have shown that the economic importance of a nation’s trade measured by trade/gross domestic product, trade-to-GDP ratio or by dyadic trade/total trade is inversely related to interstate conflict, especially when additional measures of the expense of ending trade are considered (Benson, 2004; Mansfield and Pollins, 2001; Polachek, 1980; Polachek, 1992; Polachek, Robst, and Chang, 1999). Gartzke and Li (2003) suggest that economic globalization enables states and markets to communicate. States integrated into the global economy don’t have sufficient power to control economic agents such as transnational corpora- tions at their disposal, and are less likely to declare war without the agents’ con- sent. Also, they suggest that asymmetric integration in the global economy does not increase the danger of military disputes. They attribute these results to posi- tive political externalities of the networked global economy. Polachek, Robst, and Chang (1999) found that for countries where trade gains are pursued, for- eign aid and contiguity decreases conflict. However, Toffler and Toffler (1993) argued that interdependence does not make the world more peaceful. For instance, nations most dependent on trade (U.S. and Japan) need relations to sustain their economies. That is, in spite of their international status, they are the least free nations to benefit from trade. Unlimited economic competition is not a substitute for military conflict and may be a prelude to a war (Barbieri, 1996; Mansfield and Pollins, 2001). Goenner (2004) also argues that trade’s pacifying effect is not supported by any of Bayesian Model Averaging (BMA) models. Some scholars have argued that Intergovernmental Organizations (IGOs) have a pacifying effect on international conflict (Oneal, Russett, and Berbaum, 2003; Russett, Oneal, and Cox, 2000). They represent forums for communication among member nations (Kim and Barnett, 2000). In contrast, Hafner-Burton and Montgomery (2006) used network analysis of MID (military interstate disputes: Jones, Bremer, and Singer, 1996) and studied IGO membership by dividing nations into structurally equivalent clusters. They found that the chances of a nation’s involvement in conflict are in proportion to the number of other states in same structurally equivalent cluster. As the distribution of power in the interna- tional system is reflected in IGO membership (Kim and Barnett, 2000), the num- ber of states with the same status can be interpreted as more competition among nations with similar capabilities. Also, they found that the nations with large pres- tige disparities and in-group preference have less chances of involvement in inter- national conflict. Their research also indicates the utility of network analysis as a method for studying the political impact of intergovernmental organizations. International aid as a mode of international communication has two aspects. First, it facilitates economic and cooperative relations between 140 J. H. Kim and G. A. Barnett nations (Lumsdaine, 1993). This implies its peacemaking role. Second, it is a means to realize political strategic interests of the donor country. This study examines the effect of international aid on global conflict. Neorealists provide another perspective on the structure of interna- tional conflict. They focus attention on the anarchic nature of the interna- tional system, assuming that all nations compete in an attempt to attain a better relative position with more power (Waltz, 1959). Thus, variables such as the parity of power between nations have been used to predict interna- tional conflict (Lemke and Werner, 1996; Lemke and Reed, 2001). Regard- less of a nation’s characteristics, the struggle-based international structure determines the behaviors of individual nations. Nations form alliances and seek hegemony to get ahead in the international system (Waltz, 1959). Geographical proximity is positively linked to the frequency of interna- tional conflict (Benson, 2004; Goertz and Diehl, 1992; Kocs, 1995; Vasquez, 1993). Whether this is due to enhanced opportunities for conflict or to increased intentions to fight each other is not clear (Bremer, 2000). Since proximity is a major predictor of conflict, research may present spurious associations if it is not controlled. Alliances of nations result in fewer conflicts between members than those between members and nonmembers (Bremer, 1992). Like individuals, nations construct communities (alliances) based on shared norms and inter- ests. Alliances tend to dampen international conflict. Neorealists focus on the major powers’ roles in international conflicts. These nations are more likely to be involved in conflicts, either as bullies or enforcers of the international order. They tend to fight with other strong nations rather than weak ones (Geller, 1993). Weak nations tend to fight more with weak nations. Also, the major powers interfere in many issues as a community leader and thus are more conflict-prone than minor nations (Small and Singer, 1982). Huntington (1993a, 1993b, 1996) argues that cultural factors have become the key cause of international conflict since the end of the Cold War. The Cold War suppressed the vent of nonnegotiable identities, a central component of culture, because the ideological powers (U.S. and USSR) regulated other nations. In the post-Cold War era the major cause of interstate conflict is not ideological. Rather, Huntington (1993a) argues that the source of conflict is “cultural.” International conflicts break out across the frontiers of different civilizations. Civilizations are cultural entities that can be conceptualized by objective elements such as history, religion, and language and people’s self- identification (Huntington, 1996). The different civilizations include Western, Confucian, Japanese, Islamic, Hindu, Orthodox, Latin American, and African. Cultural characteristics and differences are not apt to change or to be easily compromised in spite of globalization. People tend to separate based more on cultural homogeneity than on physical distance. Henderson and Tucker (2001, p. 318) label Huntington’s clash of civilizations claim as “cultural realism.” Communication and International Conflict 141 Huntington’s request to examine the cultural factors in international interactions is supported by several studies. Barnett (2002) reports evidence of regional civilizations. He found that international telecommunication flows during the 1990s constituted a single integrated network of nations centered about North America and Western Europe. In addition, communi- cation among culturally homogeneous regions became stronger since the mid-1990s and this trend has slowed centralization of global telecommuni- cation. Barnett and Choi (1995) found that the language(s) spoken as well as geographical location determine the network’s structure. Further, Barnett, Salisbury, Kim, and Langhorne (1999) concluded nations cluster into cul- tural (religious, linguistic, and geographical) groupings, from their analysis of global monetary, telecommunications, and trade networks. Müller (1999) argues that Huntington’s theory cannot be supported. He argues that Huntington simply substituted civilization for ideology and overly focused on the arms trade between Asian and Islamic countries, ignored that the U.S. exports ten times more weapons to the Middle East, and that only small number of worldwide violent incidents occurred between civilizations. However, Huntington argues that nations from the same civilization tend to cooperate in international conflicts against nonmembers (“kin-country syn- drome,” 1993a, p. 35), and to restrain competitors within the same civilization. Others object to Huntington’s contentions that the clash of civilizations will replace ideological rivalries as the major cause of conflict (Russett et al., 2000; Chiozza, 2002). Russett et al.’s (2000) results were drawn from 1950–1992 data. Henderson and Tucker (2001) found that the clash of civilization claim doesn’t explain conflict in the short post-Cold War period (1989–1992). How- ever, both data sets encapsulate an inappropriate length of time to character- ize the post-Cold War period. Finally, political transitions and other historical factors may co-occur with struggles between different ethnic groups and international conflict (Goertz and Diehl, 1993). Internal conditions of nations such as the transition from dictatorship to democracy are reported to lead to ethno-political conflict (Gurr, 1994). This research postulates that interdependence as a result of interna- tional communication reduces conflict between nations. This is an exten- sion of the liberal position. Interdependence increases the expected cost of initiating war against connected nations. Thus, it is assumed to have a dampening effect on conflict. Further, communication can contribute to resolve conflict peacefully (Galtung and Jacobsen, 2000). This study also tests Huntington’s claim as an extension of intercultural/international com- munication discourse (Hamelink, 1994; Carey, 1989). International commu- nication is the exchange of information reflecting cultural values (Carey, 1989). Thus, international communication represents an important factor in the structure and prediction of international conflict. As pointed out above, another rationale for examining international communication is that previous research that has examined international 142 J. H. Kim and G. A. Barnett interaction as a source of conflict prevention or resolution has relied only on economic interdependence. In addition to the exchange of goods, interna- tional communication involves people and information (Hamelink, 1994; Barnett and Lee, 2001). The trade-only approach has reduced the quality of the explanation of international conflict based on interdependence. Bremer (2000, p. 26) suspects that “socially unconnected” nations tend to have more conflict than those with ties. Nations with different value systems may have more diffi- culty communicating and understanding, resulting in a greater chance of con- flict. This research examines transnational flows as predictors of international conflict. The study describes the networked structure of international conflict during the post-Cold War era, and determines the role of international commu- nication and other factors in predicting conflict between nations. METHOD This study performs a structural analysis of global conflicts using network analysis and develops a model to predict international conflicts. The theo- retical model is composed of liberal, expanded liberal (communication), and realist variables. It stipulates that international conflict is a function of democratization, economic interdependence, geographical proximity, politi- cal transition, alliances, major power membership, civilization, IGO mem- bership, and international communication. Network analysis is a set of research procedures for identifying struc- tures in systems based on the relations among its components rather than individual attributes (Rogers and Kincaid, 1981; Wasserman and Faust, 1994; Barnett and Park, 2005). The method may be applied to describe the pat- terns of communication and conflict among nation-states (Barnett and Lee, 2001). This research examines the multilateral communication/conflict struc- ture with the degree and frequency of dyadic relations. Network analysis has been used to investigate the relations among nations, in general (Snyder and Kick, 1979; Kim and Barnett, 2000) and various forms of international communication including telecommunications, (Barnett and Choi, 1995; Barnett, 2002), the Internet (Barnett, Chon, and Rosen, 2001; Barnett and Park, 2005), print journalism (Kim and Barnett, 1996), student flows (Barnett & Wu, 1995; Chen & Barnett, 2000), film (Chon, Barnett, and Choi, 2003), air traffic and mail (Barnett and Choi, 1995; Barnett, et al., 2001). The structural relations among the countries in the international conflict network were described using the following indicators suggested by NEGOPY (Richards, 1995): connectedness, centrality, integrativeness, and system density. Connectedness is a node’s number of links (bilateral conflicts). Centrality is the mean number of links required to reach all other nodes (countries) in a network, such that the lower the value the more central the node. It is the average minimal distance from one node to all Communication and International Conflict 143 others in a group. Centrality is the degree of involvement of international actors in global networks (Kim and Barnett, 2000). Integrativeness (Integration) is the proportion of nodes that are connected to one another. Density is the total number of links divided by the number of possible links (N*(N-1)/2). Since NEGOPY’s measure of centrality only considers links, not the strength of relations (conflict) among countries, Bonacich’s eigenvector centrality measure presented by UCINET was also employed. Bonacich’s measure is calculated as the eigenvector of the largest positive eigenvalue of the net- work (Bonacich, 1972). Each of these measures indicates the state of the global conflict system at a single point in time. Multidimentional scaling (MDS) is employed to describe the overall structure of the international conflict. MDS is a set of mathematical tech- niques to analyze data sets based on the similarities of relationship between entities. MDS illustrates nations as points in space, such that the more simi- lar they are to each other, the closer they are together. In this study, the metric multidimensional procedure from UCINET 6 (Borgatti, Everett, and Freeman, 2002) was employed. Quadratic assignment procedure (QAP) correlation and regression coefficients calculated by UCINET 6 (Borgatti, Everett, and Freeman, 2002) were used to evaluate the strength of the relationship between international conflict and its predictor networks.4 Krackhardt and Porter (1986) described the procedure as comparing two N × N matrices: The procedure has several advantages over traditional linear model hypothesis testing. First it directly tests whether two matrices are similar to each other. The QAP tests take advantage of all the dyadic informa- tion represented in each matrix. That is, QAP compares each dyadic cell in Matrix A with the corresponding cell in Matrix B. The dyad is retained, then, as the appropriate unit of analysis. The second advantage of QAP is that it does not make parametric assumptions about the data. Ordinal, even categorical data can be tested without violating the distri- bution assumptions behind the procedure. (1986, p. 52) Data. Data for this research come from a variety of sources. As a mea- sure of international conflict, the research used the MID (Military Interstate Disputes) data set from the Correlates of War (COW) project (Jones, Bremer, and Singer, 1996). The dataset has been used for numerous conflict studies (See COW webpage for the details.). International conflict and its sub-concept, military interstate disputes (MIDs), are defined as follows (Jones et al., 1996): (international) Conflict refers to a sharp disagreement or collision in interests between two or more actors, while a crisis and ultimately war are more serious and intense episodes of militarized interstate disputes that have escalated. (1996, p. 168) 144 J. H. Kim and G. A. Barnett The term “militarized interstate dispute (MID)” refers to united historical cases in which the threat, display, or use of military force short of war by one member state is explicitly directed toward the government, offi- cial representatives, official forces, property, or territory of another state. (1996, p. 168) In the operational definition, Jones et al. (1996) broaden the concept of international conflict to include multilevel collisions that are not as intense as full-scale war. This broader inclusion is necessary because since the end of the Cold War the number of full-scale wars has decreased. However, minor militarized disputes among nations have increased. Sarkees, Wayman, and Singer (2003) reported only one interstate war between 1990 and 1997. There were a total of 31 wars including interstate, intrastate, extra-state wars, and 24 civil wars. The period covered by MID is 1993–2001. Each case contains the nation that initiates the hostility and the target, the nation of the aggressive action. Hostility is scaled from 1 to 5. One (1) represents “no militarized action.” Two (2) indicates “threat to use force,” which includes “threat to use blockade, to occupy territory, to declare war, to use chemical, biologi- cal, and radiological (CBR) weapons, and threat to join war.” Three (3) indi- cates “display use of force,” including “nuclear alert, mobilization, fortifying border, and border violation.” Four (4) is “use of force,” which encompasses “blockade, occupation of territory, seizure, attack, declaration of war, and use of CBR weapons.” Five (5) denotes “beginning or joining of interstate war” (Ghosn and Bennett, 2003; Ghosn, Palmer, Bremer, 2004). The use of only post-Cold War data allows for the control of the change in the influences of economic exchanges (Mansfield and Pollins, 2001). That is, economic relations are stable over this relatively short period of time. The relations among 145 countries are examined to describe the international conflict network based on MID data. The large number of countries will prevent bias due to strategic censoring in which systematic distortion of the data results from the examination of a limited set of conve- nient cases (Lemke and Reed, 2001). Degree of Democratization data were acquired from Freedom House’s “Freedom in the World 2000–2001” (Karatnycky, 2001). To dichotomize regime types, a nation is coded “free” or “not free.” “Partly free” was included in the latter category. Degree of democratization was measured as the sum of the political rights and civil rights indices. Trade data were obtained from the International Monetary Fund’s Direction of Trade Statistics (IMF, 1998). It contains data on the monetary value (USD) of trade (imports & exports) between trade dyads (IMF, 1993). Each country’s export data is denoted in f.o.b. (free on board) price, and the imports are denoted in c.i.f. (cost, insurance, freight) price (IMF, 1993; 1998). This study has analyzed the export and import data for 1997 (the data Communication and International Conflict 145 was also used for Barnett et al., 1999). Trade disparity is calculated as the difference between imports and exports for each pair of nations. Data on the GDP for each nation also comes from IMF. The proximity (distance) data between nations was acquired from the United States Department of Agriculture’s webpage. It includes 221 nations’ capitals and their distances from each other (Byers, 1997). The data are also used to determine geographical contiguity. Political transition (instability) is based on whether there was a drastic change in political regime such as the removal of a dictator or the establishment of a legal institution such as a new constitution between 1993 and 2001. Alliance data were obtained from “Formal Alliances v.3.03,” a section of COW (Gibler and Sarkees, 2004). It contains formal alliance pacts includ- ing defense and nonaggression treaties, and ententes from 1816–2000. Only pacts in effect in during the sampled time period are included. Data on the history of prior conflict, peace years and past colony was also obtained from COW (See the details from Small and Singer, 1982; Correlates of War Project, 2004.) The past conflict data covers from 1914, the beginning of World War I. The major powers variable is based on the criterion of permanent mem- bership in the United Nations Security Council. The permanent members are: the United States, the United Kingdom, China, Russia, and France. This dichotomization manifests the roles and positions of the major powers in international interactions. The civilization data come from Barnett (2002). The study conducted a data analysis of the international telecommunication network and the results are consistent with Huntington (1996). IGO joint membership data were modified from “Intergovernmental Organizations (v.2.1)”, a section of CoW project (Pevehouse, Nordstrom, and Warnke, 2004). Only alliances in effect in during the sampled time period were included. Also gathered from CoW is its Composite Capabilities Index data that measures a nation’s capability for war (Singer, Bremer, and Struckey, 1972; Small and Singer, 1982; Jones et al., 1996; Sarkees et al., 2003). Telecommunication data come from TeleGeography (Staple, 1998, 2001). They include the total minutes of international telephone calls among nations. To determine the direction of the causal relation between conflict and telecommunication, data from different points in time are examined. These data are the major telecommunication data set in this field (Engelbrecht, 2001) and have been shown to be reliable and theoretically valid (Barnett, 2002). Data on international air traffic (passenger and freight) and mail were obtained from the Digest of On-Flight Origin and Destination (OFOD) (International Civil Aviation Organization, September, 1998). The original data includes the amount of international air traffic among 1,780 cities, mea- sured as the number of passengers and the tonnage of mail. 146 J. H. Kim and G. A. Barnett International aid data were obtained from International Development Statistics (IDS), an online database constructed by OECD (OECD, 2005). It provides detailed information on aid activities, such as volume, sectors, countries, IGOs, and descriptions. These data also cover 1993–2001. RESULTS Between 1993 and 2001, 103 of the 145 (71.0%) nations were involved in 362 international dyadic conflicts. This network is very sparse. Its density is only.034. Only 3.4% of total dyads were in conflict. The result is similar to Sarkees et al. (2003) who found little international conflict in the post-Cold War period. Table 1 presents the centralities, network role, and integration for the individual countries calculated by NEGOPY (Richards, 1995), as well as the Bonacich measures of centrality calculated by UCINET 6 (Borgatti et al., 2002). Group members, in the result, means that they have at least two links with other members in the same group. The international conflict network was composed of two groups. Group 1 members had 52 nations, and group 2 was composed of 3 African nations: Angola, Zimbabwe, and Namibia. Forty two nations were isolates without conflict (Type 1 isolates) and 22 nations had only one conflict (Type 2 isolates). Centralities show that the former Yugoslavia (Serbia and Montenegro) was at the center of global conflict in the 1990s. The United States, the other NATO nations, and Russia were also involved in global conflict: Other cen- tral nations are Azerbaijan, Belarus, and Iraq. A graphical representation (MDS) of the global conflict network is pre- sented in Figure 1. It is readily apparent that Yugoslavia is at the center of this network. Yugoslavia is a locus of conflict surrounded by (in conflict with) the NATO nations. The intertwined conflict relations of the Eastern European nations present a complicated political context including ideolo- gies (past communist/ liberal regime) and religions (Islamic/non-Islamic). In addition to the Balkans conflict, there are conflicts between China and its Asian neighbors. It is hard to assert whether this is an extension of the Cold War in Asia, but it is clear that there are tensions between the nations around China. The increase of China’s military capability and the restriction by the U.S. and other Asian countries illustrate persistent tensions. The conflict structure also encompasses the regional conflicts such as China–Japan and Korea–Japan, reflecting a former colony relationship (historical factor) and proximity. The centralities and MDS also indicate that the United States is near the center of global conflict structure. It is involved in diverse conflicts including Yugoslavia and the NATO nations, Africa and the Middle East, South America, Africa, and Asia. Structurally, Figure 1 illustrates that international conflicts are grouped by civilization. Some conflicts are found between civilizations, among them India and Pakistan, and Israel with its Arab neighbors. The Communication and International Conflict 147 TABLE 1 Structural Indices of International Conflict Network (1993–2001) Role Centrality Mean Standard No. Nation Links (Group) Distance Distance Bonacich Integration 1 Albania 3 1 2.61 −0.20 14.27 667 2 Algeria 0 T1 0.00 0.00 3 Andorra 0 T1 0.00 0.00 4 Angola 3 2 0.00 0.00 5 Azerbaijan 5 1 2.37 −0.53 22.34 400 6 Argentina 2 1 2.67 −0.12 11.43 0 7 Australia 2 1 3.27 0.74 1.65 0 8 Austria 2 1 2.45 −0.42 6.27 0 9 Bahrain 1 T2 1.78 0.00 10 Bangladesh 1 T2 0.00 0.00 11 Belgium 2 1 2.49 −0.36 13.17 1000 12 Bolivia 0 T1 0.00 0.00 13 Bosnia 0 T1 0.00 0.00 14 Brazil 0 T1 0.00 0.00 15 Brunei Dar 0 T1 0.00 0.00 16 Bulgaria 2 1 5.20 0.00 17 Belarus 1 T2 1.02 0.00 18 Canada 5 1 2.25 −0.69 26.75 600 19 Cape Verde 0 T1 0.00 0.00 20 Sri Lanka 0 T1 0.00 0.00 21 Chile 1 T2 1.93 0.00 22 China 9 1 2.25 −0.69 15.77 194 23 Taiwan 4 1 3.18 0.60 4.80 667 24 Columbia 2 1 3.61 1.20 0.22 0 25 Costa Rica 1 T2 0.01 0.00 26 Croatia 3 1 2.59 −0.23 16.11 667 27 Cuba 1 T2 2.99 0.00 28 Cyprus 1 T2 7.74 0.00 29 Czech Rep 1 T2 9.04 0.00 30 Denmark 2 1 2.49 −0.36 13.17 1000 31 Dominic Rep 1 T2 1.10 0.00 32 Ecuador 1 Dy 0.00 0.00 33 El Salvador 2 1 5.47 3.82 0.00 1000 34 Estonia 1 T2 5.16 0.00 35 Faroe Is 0 T1 0.00 0.00 36 Fiji 0 T1 0.00 0.00 37 Finland 1 T2 5.16 0.00 38 France 5 1 2.35 −0.56 17.17 300 39 Germany 3 1 2.41 −0.47 14.95 667 40 Greece 4 1 2.49 −0.36 19.20 500 41 Greenland 0 T1 0.00 0.00 42 Guatemala 0 T1 0.00 0.00 43 Haiti 7 1 2.69 −0.09 12.56 67 44 HongKong 0 T1 0.00 0.00 45 Hungary 2 1 2.69 −0.09 10.16 1000 46 Iceland 1 T2 9.04 0.00 47 India 2 Tr 0.00 0.00 48 Indonesia 3 1 3.18 0.60 1.80 0 49 Iran 6 1 2.53 −0.31 13.07 333 50 Iraq 18 1 2.12 −0.89 25.47 121 (Continued) 148 J. H. Kim and G. A. Barnett TABLE 1 (Continued) Role Centrality Mean Standard No. Nation Links (Group) Distance Distance Bonacich Integration 51 Ireland 1 T2 5.16 0.00 52 Israel 3 1 3.06 0.43 2.80 0 53 Italy 4 1 2.39 −0.50 16.68 500 54 Jamaica 0 T1 0.00 0.00 55 Japan 4 1 2.55 −0.28 8.49 500 56 Kazakhstan 1 T2 0.45 0.00 57 Jordan 2 1 2.57 −0.25 6.94 0 58 South Korea 2 1 3.22 0.65 2.73 1000 59 Kuwait 1 1 3.56 0.00 60 Lebanon 1 1 0.39 0.00 61 Latvia 1 1 4.13 0.00 62 Libya 1 1 1.79 0.00 63 Lithuania 3 1 2.49 −0.36 14.53 1000 64 Luxembourg 1 1 9.04 0.00 65 Macau 0 T1 0.00 0.00 66 Malaysia 0 T1 0.00 0.00 67 Maldives 0 T1 0.00 0.00 68 Mexico 0 T1 0.00 0.00 69 Moldova 1 1 4.13 0.00 70 Morocco 1 1 5.16 0.00 71 Mozambique 1 Dy 0.00 0.00 72 Oman 1 1 1.78 0.00 73 Netherlands 4 1 2.37 −0.53 17.33 333 74 Neth. Ant. 0 T1 0.00 0.00 75 New Zealand 1 1 0.25 0.00 76 Norway 3 1 2.43 −0.45 17.44 333 77 Pakistan 1 T2 0.00 0.00 78 Panama 0 T1 0.00 0.00 79 Paraguay 0 T1 0.00 0.00 80 Peru 1 Dy 0.00 0.00 81 Philippines 3 1 3.20 0.63 4.05 1000 82 Poland 2 1 2.49 −0.36 15.49 1000 83 Portugal 2 1 2.49 −0.36 13.17 1000 84 Guinea Bis 0 T1 0.00 0.00 85 Puerto Rico 0 T1 0.00 0.00 86 Qatar 1 T2 0.26 0.00 87 Romania 2 1 2.69 −0.09 16.22 1000 88 Russian Fed 24 1 1.75 −1.41 59.12 130 89 Sao Tome 0 T1 0.00 0.00 90 Saudi Arab 3 Tr 3.65 0.00 91 Senegal 0 T1 0.00 0.00 92 Singapore 0 T1 0.00 0.00 93 Slovakia 1 T2 5.16 0.00 94 Vietnam 4 1 3.20 0.63 4.37 1000 95 Slovenia 2 1 2.63 −0.17 6.33 1000 96 S. Africa 1 Dy 0.00 0.00 97 Spain 3 1 2.45 −0.42 16.44 1000 98 Sudan 2 1 2.86 0.16 3.53 0 99 Sweden 1 1 5.16 0.00 100 Switzerland 1 1 5.16 0.00 (Continued) Communication and International Conflict 149 TABLE 1 (Continued) Role Centrality Mean Standard No. Nation Links (Group) Distance Distance Bonacich Integration 101 Syria 3 1 2.78 0.05 6.95 0 102 Thailand 2 Tr 0.66 0.00 103 UAE 2 1 2.57 −0.25 6.94 0 104 Tunisia 0 T1 0.00 0.00 105 Turkey 8 1 2.29 −0.64 34.09 286 106 Ukraine 3 1 2.47 −0.39 10.43 667 107 Macedonia 2 1 2.69 −0.09 16.20 1000 108 Egypt 3 1 3.00 0.35 3.08 333 109 UK 5 1 2.37 −0.53 22.12 333 110 US 12 1 1.92 −1.16 34.25 156 111 Uruguay 0 T1 0.00 0.00 112 Venezuela 4 1 2.75 −0.01 2.46 0 113 Samoa 0 T1 0.00 0.00 114 Yemen 1 T2 0.51 0.00 115 Yugoslavia 41 1 1.73 −1.44 73.94 64 116 Bahamas 0 T1 0.00 0.00 117 TurksCaicos 0 T1 0.00 0.00 118 Cayman Is 0 T1 0.00 0.00 119 TrinidadTob 1 1 0.22 0.00 120 Barbados 0 T1 0.00 0.00 121 Bermuda 0 T1 0.00 0.00 122 Armenia 3 1 3.02 0.38 7.56 667 123 Uzbekistan 2 Tr 0.00 0.00 124 Georgia 3 1 2.43 −0.45 19.11 1000 125 Kyrgyzstan 1 T2 0.00 0.00 126 Turkmenista 0 T1 0.00 0.00 127 Tajikistan 1 T2 0.00 0.00 128 Guyana 2 Tr 0.26 0.00 129 Nicaragua 4 1 4.51 2.47 0.04 333 130 Honduras 2 1 5.47 3.82 0.01 1000 131 Fr. Polyne. 0 T1 0.00 0.00 132 New Caled. 0 T1 0.00 0.00 133 Ghana 0 T1 0.00 0.00 134 Nigeria 0 T1 0.00 0.00 135 Zimbabwe 2 2 0.00 1000.00 137 Suriname 1 T2 0.03 0.00 138 Myanmar 1 T2 0.17 0.00 139 Western 3 T1 0.00 0.00 140 Tanzania 0 T1 0.00 0.00 141 Namibia 3 2 0.00 1000.00 142 Botswana 1 2 0.00 0.00 143 Swaziland 1 Dy 0.00 0.00 144 Lesotho 1 Dy 0.00 0.00 145 Zambia 1 2 0.00 0.00 146 Malawi 0 T1 0.00 0.00 Group Row mean = 2.75. Standard Deviation = 0.713. N = 145. Note. 1. Links: Bidirectional links for each dyad are considered as one. 2. No.136 has no corresponding nation. 3. Role: “1” shows that the node is a member of “group 1.” “2” shows “group 2.” “T1” shows nodes without conflict. “T2” shows nodes with only one conflict. “Dy” shows an isolated “dyad”: Two nodes are linked only to each other. “Tr” shows “tree”: If one or more isolates are linked to a node with a single link to a group, that node is a Tree. 4. Bonacich Centralities are reported only for group 1 members. 150 FIGURE 1 International Conflict Network 1993–2001. The figure is a result of MDS conducted by UCINET 6 (S. P. Borgatti, M. G. Everett, and L. C. Freeman, 2002). The thickness of the line indicates the strength of conflict. See Table 1 for key. Dotted line and names of region are added by the authors. T1 nodes are not reflected in the figure. Communication and International Conflict 151 major powers, the U.S., China, and Russia serve as bridges between these conflicts. Predicting International Conflicts This section of the paper reports the association between the international conflict network of 103 nations with conflicts in 1993–2001 and the struc- tures of regime type, degree of democratization, trade, national capability, proximity, political transition, alliance, major powers, civilization, IGO joint membership, and international communication. The QAP correlations among the variables are shown in Table 2. The first column provides the strength of the correlation of each variable with conflict. Table 2 shows that the correlations of the neorealist variables with con- flict are higher than the liberal ones. Geographical contiguity, past conflict, past colony, and civilization difference are positively correlated with con- flict, with strong significance (p <.005). However, liberal variables such as trade (export, import, and discrepancy of trade) and democracy showed weaker significance (p <.05). Among neorealist variables, geographical contiguity, distance, past conflict, and peace years) are the strongest predictors of international conflict. The effect of alliance membership and peace years deviated from anticipated direction of correlations: allied nations and longer peace years are positively correlated with conflicts. The result of alliance can be interpreted to reflect that political alliance does not guarantee peace among alliance partners since the end of the Cold War. As recent tensions between Korea and Japan who are allied under the U.S. security community shows, other factors such as history may affect the relations. Even if both of nations are major partners of the U.S.’s East Asian policy, they ceaselessly bicker over territory and history. The finding that “peace years” is not nega- tively related to conflict shows that it has low predictive power (Bremer, 2000; Gleditsch and Ward, 2000). The civilization variable’s positive relation- ship to conflict supports Huntington’s argument that nonnegotiable cultural identity is an important factor behind international conflict (Huntington, 1996; see Figure 1). Joint membership in a civilization is positively related to alliance and cooperative variables (trade, telecommunication, and air traf- fics). National capabilities’ positive correlation with conflict shows the trend that indices included in the calculation of national capabilities (i.e., national resource, military budget, military personnel, iron and steel production, dis- posable energy, and population) are positively related factors to joining con- flict. Geographical contiguity is strongly associated with not only conflict but with (expanded) liberal variables, which shows that it is an important factor in conflict, cooperation, and communication. This result can be interpreted further by observing regression results controlling the effect of geographical contiguity. The history of dyadic conflict is a strong predictor of current conflict, which reminds us of the fact that each nation’s perceived injustice TABLE 2 QAP Correlation Coefficients among International Conflict and Its Predictors Conflict MID Neorealism Geog.cont..429† Distance −.156† −.232† Maj.power.111.058° −.022 Nat.capa..077*.031*.064.695† Alliance.105†.229† −.293†.086*.062* † † † Past conf..209.266 −.105.204†.096†.125† Past.co.(0,1).442†.379† −.113†.196†.136†.107†.357† Past colo..194†.162† −.070†.226†.074*.067†.061†.279† Peace yrs.239†.188† −.084†.070*.055†.095†.521†.086†.094† Cul.Real. Civilization.066†.218† −.333†.006† −.042†.452†.081†.050°.011†.068† Liberalist Democ..015* −.094†.028†.019†.027† −.223† −.006†.031†.005†.029* −.407† Rgm.type −.038†.075† −.045†.003†.004†.201†.010† −.032†.000† −.007†.404† −.741† IGO jnt. M.064†.194† −.147†.178°.119*.376†.195†.165†.041†.053°.409† −.240†.181† Pol.instab.038†.041† −.153† −.032† −.065* −.054* −.050† −.029†.001† −.035†.031† −.049† −.003† −.138° Pol.inst(0,1).051†.058† −.217† −.045† −.092† −.077* −.069* −.042†.001† −.048*.020† −.407† −.004† −.196°.704† 152 Trade-ex.069*.195† −.101†.311†.261†.235†.225†.225†.068*.025†.207† −.161†.135†.501† −.127† −.181† Trade-im −.072*.189† −.094*.315†.268†.230†.232†.231†.068*.030†.206† −.158†.134†.503† −.128† −.183†.953† Trade-dis.077*.005† −.058†.253† −.045†.146†.175†.148† −.004† −.035†.110† −.075†.059†.374† −.141† −.207†.657†.688† Comm. Tele ‘97.067*.176† −.064†.154†.134†.155†.113†.100†.057°.057*.113† −.074†.065†.191† −.040† −.057†.614†.599†.284† Tele ‘00.065*.181† −.076†.160†.133°.165†.110†.103†.062*.055°.120† −.076†.066†.193† −.037† −.052*.604†.588†.282†.997† Air mail.035*.041* −.024†.148†.135°.127†.107†.084†.035*.009†.067† −.064†.060†.155° −.045† −.062†.525†.538†.330†.512†.551† Air Freight.041*.017† −.009†.151†.151°.115†.082†.066°.064*.023*.037* −.056†.054†.124° −.049† −.069†.486†.487†.365†.375†.327†.824† Air pasngr..056*.103† −.069†.185†.144°.183†.133†.107†.056*.033*.113† −.092†.081†.235° −.060† −.085†.660†.652†.363†.736†.703†.834†.765† Int.Aid.021† −.008†.012† −.001†.027† −.004† −.006† −.008† −.011† −.007† −.000† −.017†.005† −.002†.014†.024† −.015† −.013† −.020† −.004†.525† −.001† −.002† −.005† Note 1. *: p<.05. °: p<.01. †: p<.005 2. Abbreviated words: Cul.Real:Cultural Realism Comm:Communication variables MID:Military Interstate Disputes. Geog.Cont:Geographical contiguity. Maj.power: Major power Nat.Capa: National Capabilities Past.Conf:Past conflict (History of dyadic conflict) Past.co(0,1):Dichotomized history of dyadic conflicts. Past colo:Past colony Peace yrs:Peace years Democ:Level of Democracy (difference) Rgm.type: Free/Non free nations (dichotomous:1,0 repectively) Pol.inst:Political instability Pol.st(0,1):dichotomized political instability Trade-ex: export Trade-im: import Trade-dis: Trade disparity Tele ‘97: Telecommunication (1997) Tele ‘00: Telecommuni- cation (2000) Air pasngr: Air passenger Int.Aid: International aid. Communication and International Conflict 153 with regard to international relations reflects socially shared knowledge of past conflict: Memory of past conflict and colony is a source of conflict now. Among the liberal and communication variables, only imports and regime type (freedom) are inversely related to conflict. IGO joint membership, politi- cal stability, exports, telecommunication, transnational flows by air traffic are positively related to conflict (Table 2). The absolute degree of democratization is positively related to conflict, which may be supported by democratic nations’ multilateral involvement in Kosovo, Somalia, Rwanda, and the former Yugoslavia during 1993–2001. Several other findings with regard to the liberal variables are: One, like alliance, joint IGO membership didn’t show a pacify- ing effect. This reflects the different atmosphere since the end of the Cold War. International roundtables may be a battlefield to seek for national interest because there is no strong ideological bondage among nations. Two, political instability is positively related to conflict lending support for diversionary con- flict theory. Diversionary theorists argue that national leaders tend to translate internal conflict to aggressive foreign policy (Miller, 1999). Also, internal mili- tary conflict results in multilateral involvement of a group of nations. Three, trade disparity was positively related to conflict. This finding supports the posi- tion that the absolute amount of trade might be less powerful than disparity (Benson, 2004; Mansfield and Pollins 2001; Polachek, 1992; Polachek, Robst, and Chang, 1999). Four, telecommunications for both 1997 and 2000, are pos- itively related to conflict. Five, most cooperative interactions including trade, telecommunication, and air traffic show strong positive associations with each other. This phenomenon is also obvious with IGO joint membership and its relation with other liberal factors. International aid is not significantly corre- lated to conflict. This result may reflect the limits of a data set that overlooked the actions of international aid organizations. Also, it may indicate that the major powers tend to engage in conflict and then offer aid. Thus, in some cases, global conflict involves post-conflict aid as in the case of Iraq. In order to determine the predictive power of the independent variables in combination, a multiple regression analysis was conducted. See Table 3. All the possible combinations of independent variables were examined. The authors present two models representing old (liberal and realist) and new (communication) approaches. Model 1 includes only the liberal and neore- alist (as predictors) and control variables (geographical proximity and dis- tance). It accounts for 21.0% of variance in the conflict network. There are inverse and significant relations between distance, regime type (freedom), export, and conflict. Though export showed positive correlation with con- flict in Table 2, it showed inverse relation with conflict with controlled vari- able such as geographical proximity is considered. Model 2 includes the communication variables (international freight and telecommunication). It explains 30.0% of the variance, a 9.0% increase. Adding communication variables substantially enhanced the predictive power of the model. This model strongly supports this paper’s argument: Trade is not the only indicator 154 J. H. Kim and G. A. Barnett TABLE 3 Multivariate Regression Models for the Prediction of International Conflict Model 1 Model 2 NeoR + Liberal NeoR + Liberal + w/cont. Std.coef.(sig.) Comm w/cont. Std.coef.(sig.) Intercept 0.000(0.923) 0.000(0.293) Neorealism (Cul.Realism) Distance Geog.contig. −0.058(0.010) −0.053(0.018) Major powers 0.405(0.000) 0.290(0.000) Conflict hist. Colony history 0.334(0.000) Peace years 0.110(0.001) 0.117(0.000) Alliance Civilization Liberalism and Communication Freedom(0,1) Democracy. −0.029(0.020) −0.041(0.002) Int.mail 0.047(0.017) Int.freight Int.passenger 0.053(0.019) Telecomm.(‘97) Telecomm.(‘00) 0.023(0.041) Trade – export Trade – import −0.039(0.017) −0.111(0.000) Trade parity IGO jnt.Memb. Int.Aid. R2 0.210 0.300 ΔR2 +0.090 Adj.R2 0.210 0.300 Probability 0.000 0.000 Note 1. N = 103 countries which had international conflict with each other in 1993–2001, 10506 dyads. 2. Among all the models tested, only representative three models are shown. of interdependence in the globalized world. Export is inversely related to international conflict. It includes the combined effects of being a former col- ony, the physical distance between the countries, regime type, whether or not countries shared a common border, their freight flows, and the strength of their prior conflicts. It also indicates that the neorealist variables are more powerful predictors of conflict than the liberal variables and that geographi- cal contiguity and strength of prior conflict are the strongest predictors of global conflicts. In addition, the model shows that communication variables, in addition to trade are valid predictors of international conflict. DISCUSSION The network analysis of international conflict in this study enhanced the understanding of international conflict structure: It showed that most conflicts Communication and International Conflict 155 between 1993 and 2001 were regional, multilateral, and few in number. The MDS shows that states tend to conflict with their geographical neigh- bors. The overall conflict structure shown in Figure 1 is commensurate with geographical proximity (see dotted lines). This trend can be interpreted that groups of culturally homogeneous nations are fighting with members of different groups. In spite of numerous criticisms against “cultural realism,” the structure revealed indicates that it might be impetuous to abandon the heterogeneity of culture as a factor to predict conflict. Also, the locus of conflict (the former Yugoslavia) involved a majority of NATO (the only sub- stantial military alliance in the area). There were group-based involvements in conflicts in the post-Cold War era. The multilateralization of conflict shows that conflict is no longer a one-to-one phenomenon. The same trend also applies to economy and telecommunication. Regional integration deep- ens economic cooperation with roundtables (e.g., NAFTA, EC; Dicken, 2003) and the telecommunication structure has become more bloc-shaped after the Cold War (Barnett, 2002). Finally, 11 dyads had only one counter- part for all their conflicts. Does disappearing international conflict mean that the world is more peaceful than the realist theory suggests? Not necessarily. The frequency and intensity of domestic conflicts (e.g., civil war) should be considered before reaching such a conclusion. Some nations without international con- flict, such as Mexico, suffered internal conflict between government and rebels. Hong Kong became a part of China without any conflict with the UK. And nations may not be the major troublemakers any more. As seen from the September 11 terrorist attack, globalization invited nonstate actors in human violence. This network analysis was limited to what nations do in international system. The regression results showed that the neorealist variables and geogra- phy are superior predictors of international conflict: past conflict, national capability, past colony, contiguity and geographical proximity (Bremer, 2000). However, the liberal variables cannot be ignored. For example, regime type (Free/Not free) confirms the argument that democratized nations are less likely to get involved in conflicts. Rummel (1995) discussed this point and many “democratic peace” theorists argued that a democra- tized regime itself is the strongest pacifier. Given that about half the nations on the globe are democratized (Karatnycky, 2001), the result shows that world peace has a long way to go. The most important finding of this research is that communication vari- ables significantly increase the explanatory power of the liberal conflict pre- diction model. This is a big step forward in the liberal position which has been stagnant for a long time due to its focus on trade as the singular indi- cator of interdependency. Telecommunication (1997 and 2000), exports, IGO joint membership, alliances, air passengers, mail, and freight are all sig- nificantly positively correlated to conflict. Only two variables, exports in the 156 J. H. Kim and G. A. Barnett regression model and imports in the correlations, showed inverse relation to conflict. In other words, notwithstanding statistical significance trade and communication, variables didn’t show a consistent pacifying effect on glo- bal conflicts. One possible interpretation for this result comes from World Systems theory. The central actors in communication networks are the major powers and the structure of global transnational flows are layered sys- tems, even if the structures are not exactly commensurate with the theory.5 In other words, global communication reflects major-power-centered and strat- ified international order more than global conflict. This may be inevitable in the capitalized world (Mosco, 1993). A second explanation is that the comparative evaluation of both the pacifying effect of communication and the feasibility of conflict as a solution to perceived domestic/foreign problems might lead to a nation’s decision in favor of conflict. The specific conditions and limitations that enhance com- munication’s pacifying effect should be the subject of future research. Com- munication as a “reflection” of the existing risk between nations should be simultaneously evaluated with communication as a possible “means” of mutual understanding and economic profit-seeking. Economic actors sensi- tively respond to international relations. Morrow, Siverson, and Tabares (1998) indicated that poor relations between nations strongly inhibit tran- snational trade flows, and this makes the role of trade “indeterminate” (Morrow, 1999, p. 488). In other words, communication sometimes reflects, more than affects, international relations. Morrow et al., (1998) concluded: Politics affects trade flows because economic actors care about political tasks. International conflict disrupts trade, introducing risk. The institu- tions of limited government reassure actors, reducing risk. Democracy by itself is not a panacea for increasing trade. But modern democracies generally have been limited governments (1998, p. 659). This argument discloses different aspects of communication. Morrow’s position may be used to interpret other international communication vari- ables: even if communication can pacify conflict, it simultaneously may decrease when international relations with their counterparts worsen. The participants in international interactions compare costs with bene- fits before joining relationships, and this process leads to varying degrees of interdependence. Decreased autonomy by joining intergovernmental orga- nizations or multilateral roundtables can be considered as a cost, which tends to be compared with benefits from the relationship. The mechanism may make nations evaluate the degree to which communication is either beneficial (e.g., promote mutual understanding, economic profit) or detri- mental (e.g., increase their dependency). The high involvement of major powers in conflict shows that they have greater chances of keeping their costs low; that is, their international relations are less vulnerable to the loss Communication and International Conflict 157 of autonomy. As Jones (1995) points out, the relation that is just a trivial dependence for major powers can be critical for the minor counterpart. Los- ing a communication channel with a major power may be fatal for minor nations. This risk may be a reason for minor nations’ joining in multilateral conflict led by a super power. Extant literature suggests that communication interdependence should be jointly considered with each nation’s relations to major powers and secu- rity relations. Benson (2004) argues that the relation with major powers is consistently a robust predictor of international conflict and that it is strongly related to both security and interdependence. However, she added that interdependence is in proportion to conflict weaker than full-scale war such as threat and the use of force, which implies that nations accept conflict as a feasible solution for national interest as long as the conflict is not as costly as full-scale war. Security relations such as arms trade may be used as well to predict conflict in future investigations (Kinsella, 1998). The results showing that communication variables should be consid- ered combined with trade variables may be supported by Keohane and Nye (1998; p. 1977). They offered a synthesis of liberal and realist perspectives suggesting new communication technology (1998) and multilateral, multi- channel interactions among nations (1977) complicate international relations. They suggested that increasing “complex interdependence” (1977; p. 23) among nations through IGOs, NGOs, telecommunication, trade, and tran- snational companies ironically decreases the sovereignty of individual states and the utility of military means in establishing individual states in the world. The results of this research strongly support the notion that commu- nication has a complex effect on global conflict. New technology also increases the complexity of interdependences by drastically reducing com- munication cost over longer distance (Keohane and Nye, 1998). Information exchanged globally became an origin of power through “free,” “commer- cial,” and “strategic” information like surveillance (1998, pp. 84–85). Chan- nel diversification among nations affects, and simultaneously is affected by, the political geography of the world. For example, telecommunication tech- nology, which is one of the variables measured in this study, can be used by both major powers as well as nonstate actions like “terrorists” or numer- ous NGOs on the web. It prevents a monopoly of information flow. Also, it creates and simultaneously reevaluates the credibility of information sources through open competition. Credibility as a basis of power is getting harder to achieve and this shows a dilemma of globalization. Their position explains why nations are both fighting and cooperating in anarchic compe- tition over the power. The mixed effect of communication can be linked to current IR approaches involving framing, prospect theory, and social construction. From a social constructionist perspective, state’s behavior is affected by decisionmakers’ perception of the status quo. Such a position is supported 158 J. H. Kim and G. A. Barnett by Milliken’s work (2002) that chronologically traced Korean War by analyz- ing contexts, messages, and perceptions of decisionmakers and general publics. Milliken posits that state behavior is significantly influenced by how each party defines situations and prospects, not by simple cost-benefit anal- ysis. Prospect theory suggests that state’s behavior frequently deviates from the axioms and assumptions of “rational choice” theorists because state’s involvement in conflict depends more on “changes on assets” than “net asset levels” (Levy, 2003, p. 216). States sometimes seek “loss aversion” (2003, p. 216) rather than “comparable gains” and such intent leads to reck- less-looking choice to go to war. This position is originated from social psy- chology and its lab-experiment results. According to experimental evidence, leaders set up their decision by “reference points” susceptible to “over- weight losses” (2003, p. 215). To sum up, the content and forms of commu- nication influences the way people “frame” problems from the reference point affected by certainty of loss rather than probabilistic calculation of expected outcome. For example, politicians tend to utilize framing as George W. Bush framed Saddam Hussein as “Hitler” (Mintz and Redd, 2003, p. 193). These approaches suggest us that “framing” is setting context and environment of global conflict. The multivariate regression models in the current study show limited explanatory power (R2 = 0.30), even if the results are statistically significant. This is plausible considering that the conflict network is very sparse and its distribution deviates from normality.6 Over 9,690 of the cells in the matrix are zero. Another possible explanation of the low explained variance is that the motives for conflict might be idiosyncratic or diversionary. Idiosyncratic in this sense means that each international conflict has its own context; for example, chronic disputes over territory, tradition, or natural resources (as seen in many African nations’ conflicts over minerals or water). Another idiosyncratic factor suggested by this study is the political transformation or instability of each nation, which is also a valid predictor of conflict. Diver- sionary motives of national leaders especially when domestic power is at crisis, may lead to the involvement in political conflict (Levy, 1989). This can be examined in future research. In sum, the theory of interdependence expanded with communication variables has shown mixed results: Communication enhances the explana- tory power of the liberal model, but most communication variables are pos- itively related to conflict. This study also suggests that the two paradigms of conflict prediction have only limited explanatory power. This calls for a new endeavor of communication and political scientists to deal with each conflict under multiple considerations including idiosyncratic factors. Because communication can facilitate global peace as well as reflect the risk of international conflict, future research should consider the time order of international events and time-lag of predictors. In addition, as Organski and Kugler (1980) found, power parity may be a predictor of global conflict Communication and International Conflict 159 because nations want to restrict competitors. Parity studies can be expanded to other communication variables. National capability is found positively related to engagement in global conflicts, too. National disputes arise over their shared history — “an unending dialogue between the present and the past” (Carr, 1962, p. 22)—of conflict and imperialism. The finding that geo- graphical contiguity and history are the two major factors of conflict seems obvious. The models evaluating communication predictors produced double- sided results pointing out that communication can be used for conducting computerized war as well as for pacifying conflict. Mosco (1993) pointed out that the modern war system is supported by advanced telecommunica- tion both in conducting warfare with higher efficiency and in seeking pri- vate interest by military-industrial-communication complex. In the era of transnational companies (TNCs), the major actors in global telecommunica- tion and intergovernmental organizations are commensurate with core- nations (Barnett, 2002; Kim and Barnett, 2000). Therefore, this study also calls for the analysis of quality and content of global communication as a subject of future research. Peace can be achieved by mutual understanding as long as its content is not war-prone. Given that communication seeks a sustainable peace, now is the time to analyze further the role of global com- munication in creating international peace. CONTRIBUTORS Jang Hyun Kim (M.A. Yonsei University, 1998) is a doctoral candidate and instructor in the Department of Communication at the State University of New York at Buffalo. George A. Barnett (Ph. D., Michigan State University, 1976) is Professor of Communication at the State University of New York at Buffalo. His research focuses on structural models to model the changes in International commu- nication and their impact on globalization. NOTES 1. See Kegley and Wittkopf (2006), Berridge (1997), and Holsti (1992) for an overview of Interna- tional Relations (IR) approach to international conflict. 2. To elaborate, there are a number of reasons for the failure. One, these paradigms were devel- oped to describe the likelihood of international conflict during the Cold War (1945–1989). Two, scholars have taken too narrow a view of interdependency, viewing the relations among nations only in eco- nomic terms. Three, the internal conditions within nations were not fully examined; rather they focused only on whether or not a nation was a democracy. Four, the analysis of conflict relied on methods that were incapable of fully examining the relations among nations. And the failure of two “main stream” theories, namely liberal and realist, invited alternative approaches focusing on national leaders’ decisionmaking, 160 J. H. Kim and G. A. Barnett low-politics (environmental and economic problems of the world), long-cycle analysis, and feminist arguments. However, neither of them successfully has dominated the disputes over the international conflict. 3. Monadic research focuses on the attributes of each nation or conflict, ignoring the relations between nations. Dyadic research only focuses on individual pairs of nations, ignoring multilateral traits. 4. The number of variables used for correlations coefficients is larger than regression analysis. This is because correlation coefficients are enough indices of the direction and intensity of relationship between variables and regression analysis cannot include all homogeneous variables as independent variables because of the limitation of multicollinearity. This problem is dealt with in the authors’ current research that includes all variables in structural equation models. 5. See a series of discussions about the global communication structures — IGOs (Kim and Barnett, 2000); culture, (Barnett and Choi, 1995); telecommunication (Barnett, 2002); the Internet (Barnett, Chon, and Rosen, 2001; Barnett and Park, 2005); print journalism (Kim and Barnett, 1996); student flows (Barnett & Wu, 1995; Chen & Barnett; 2000); film (Chon, Barnett & Choi, 2003); and air traffic and mail (Barnett & Choi, 1995; Barnett, et al., 2001). 6. Log-transformations of independent and dependent variables to lessen the effect of low vari- ance and normality were conducted, but the result was not significantly different. Also, the replacement of conflict data by the PRIO/Uppsala Armed Conflict Dataset (International Peace Research Institute Oslo, 2005) deepened the low variance and normality problem of the dependent variable. Such attempts might be an excuse for presenting the result of this study without further consideration. The authors do not report all the attempted transformations and analysis in full detail. REFERENCES Barbieri, K. (1996). “Economic interdependence and power. A path to peace or a source of conflict?” Journal of Peace Research, Vol. 33, pp. 29–49. Barnett, G. A. (2002). A Longitudinal Analysis of the International Telecommunication Network: 1978–1999. Paper presented at conference at Beijing Broadcast Institute. Barnett, G. A. and Y. Choi, (1995). “Physical Distance and Language as Determi- nants of the International Telecommunications Network. International Political Science Review, Vol. 16, pp. 249–265. Barnett, G. A. and Y. Wu. (1995). “The International Student Exchange Network: 1970 & 1989.” Higher Education, Vol. 30, pp. 353–368. Barnett, G. A. and M. Lee, (2001). “Issues in Intercultural Research.” In W.B. Gudykunst and B. Moody, eds., Handbook of International and Intercultural Communication, Thousand Oaks, CA: Sage, pp. 275–290. Barnett, G. A. and H. W. Park, (2005). The Structure of International Internet Hyper- links and Bilateral Bandwidth. The Annales des Telecommunications, Vol. 60, No. 9–10, pp. 1115–1132. Barnett, G. A., B. S. Chon, and D. Rosen, (2001). “The Structure of International Internet Flows in Cyberspace. NETCOM, Vol. 15, No. 1–2, pp. 61–80. Barnett, G. A., J. G. T. Salisbury, C. Kim, and A. Langhorne, (1999). Globalization and International Communication: An Examination of Monetary, Telecommuni- cations, and Trade Networks. The Journal of International Communication, Vol. 6, pp. 7–49. Becker, L. B. (1981). Secondary Analysis. In G. H. Stempel, and B. H. Westley, eds., Research Methods in Mass Communication, Englewood Cliffs, NJ: Prentice-Hall. Benson, M. A. (2004). “Dyadic Hostility and the Ties that Bind: State-to-State Versus Secu- rity and Economic Relationships.” Journal of Peace Research, Vol. 41, pp. 659–676. Communication and International Conflict 161 Berridge, G. R. (1997). International Politics: States, Power and Conflict since 1945, 3rd ed., London: Harvester Wheatshaf. Bonacich, P. (1972). “Factoring and Weighting Approaches to Status Scores and Clique Identification.” Journal of Mathematical Sociology, Vol. 2, pp. 113–120. Borgatti, S. P., M. G. Everett, and L. C. Freeman, (2002). UCINET for Windows: Soft- ware for Social Network Analysis, Harvard, MA: Analytic Technologies. Bremer, S. A. (1992). “Dangerous Dyads: Conditions Affecting the Likelihood of Inter- state War, 1816–1965.” Journal of Conflict Resolution, Vol. 36, pp. 309–341. Bremer, S. A.(2000). Who Fights Whom, When, Where and Why? In J. A. Vasquez, ed. What Do We Know About War?, Boston: Rowman & Littlefield, pp. 23–26. Bueno de Mesquita, B. and D. Lalman, (1992). War and Reason: Domestic and International Imperatives, New Haven: Yale University Press. Byers, J. A. (1997). Great Circle Distances Between Capital Cities. Retrieved from http://www.wcrl.ars.usda.gov/cec/java/capitals.htm Carr, E. H. (1962). What Is History?, New York: Knopf. Carey, J. W. (1989). Communication as Culture: Essays on Media and Society, Boston: Unwin Hyman. Chen, T. M. and G. A. Barnett (2000). “Research on International Student Flows from a Macro Perspective: A Network Analysis of 1985, 1989, and 1995.” Higher Education, Vol. 39, pp. 434–453. Chiozza, G. (2002). “Is There a Clash of Civilizations? Evidence from Patterns of International Conflict Involvement, 1946–97.” Journal of Peace Research, Vol. 39, No. 6, pp. 711–734. Chon, B. S., G. A. Barnett, and Y. Choi (2003). Clustering Local Tastes in Global Culture: The Reception Structure of Hollywood Films. Sociological Research Online, Vol. 8, No. 1, Retrieved 8/1/2006 from http://www.socresonline.org.uk/8/1/chon.html. Chon, B. S., J. H. Choi, G. A. Barnett, J. A. Danowski, and S. H. Joo. (2003). A Struc- tural Analysis of Media Convergence: Cross-Industry Mergers and Acquisitions in the Information Industries. Journal of Media Economics, Vol. 16, No. 3, 141–157. Correlates of War Project (2004). Colonial/Dependency Contiguity Data, 1816–2002, Version 3.0. Deutsch, K. W. (1966). Nationalism and Social Communication: An Inquiry into the Foundation of Nationality, Cambridge, MA: MIT press. Dicken, P. (2003). Global Shift: Reshaping the Global Economic Map in the 21st Cen- tury, New York: Guilford Press. Engelbrecht, H. J. (2001). “Statistics for the Information Age.” Information Econom- ics and Policy, Vol. 13, No. 3, pp. 339–349. Galtung, J. and Jacobsen, C. G. (2000). Searching for Peace: The Road to Transcend, London: Pluto Press. Gartzke, E., and Li, Q. (2003). “War, Peace, and the Invisible Hand: Positive Politi- cal Externalities of Economic Globalization.” International Studies Quarterly, Vol. 47, No. 4, pp. 561–586. Geller, D. S.(1993). “Power Differentials and War in Rival Dyads.” International Studies Quarterly, Vol. 37, pp. 173–193. Ghosn, F. and S. Benette, (2003). Codebook for the Dyadic Militarized Interstate Incident Data, Version 3.0. Retrieved December 12, 2003, from http:// com2.la.psu.edu 162 J. H. Kim and G. A. Barnett Ghosn, F., G. Palmer, and S. Bremer, (2004). “The MID3 Data Set, 1993–2001: Pro- cedures, Coding Rules, and Description.” Conflict Management and Peace Sci- ence, Vol. 21, pp. 133–154. Gibler, D. M. and M. Sarkees, (2004). “Measuring Alliances the Correlate of War Formal Interstate Alliance Data Set 1816–2000.” Journal of Peace Research, Vol. 41, No. 2, pp. 211–222. Gleditsch, K. S. and M. D. Ward, (2000). “War and Peace in Space and Time: The Role of Democratization.” International Studies Quarterly, Vol. 44, pp. 1–29. Goenner, C. F. (2004). “Uncertainty of the Liberal Peace.” Journal of Peace Research, Vol. 41, No. 5, pp. 589–605. Goertz, G. and P. F. Diehl (1992). Territorial Changes and International Conflict, London: Routledge. Goertz, G., and P. F. Diehl (1993). “The Initiation and Termination of Enduring Rivalries.” American Journal of Political Science, Vol. 39, pp. 30–52. Greig, J. M. (2002). “The end of geography? Globalization, Communications and Culture in the International System.” Journal of Conflict Resolution, Vol. 46, No. 2, pp. 225–243. Gurr, T. R. (1994). “Peoples Against States: Ethnopolitical Conflict and the Changing World System: 1994 Presidential Address.” International Studies Quarterly, Vol. 38, No. 3, pp. 347–377. Hafner-Burton, E. M. and A. H. Montgomery (2006). Power Positions — Interna- tional Organizations, Social Networks, and Conflict. Journal of Conflict Resolu- tion, Vol. 50, No. 1, pp. 3–27. Hamelink, C. J. (1994). The Politics of World Communication: A Human Rights Per- spective, Thousand Oaks, CA: Sage. Harrison, E. (2000). “Reassessing the Logic of Anarchy: Rationality Versus Reflexivity. Paper presented to International Studies Association. Hart, J. (1974). Structures of Influence and Cooperation-Conflict. International Interactions, Vol. 9, pp. 141–162. Henderson, E. A., and R. A. Tucker (2001). Clear and Present Strangers: The Clash of Civilizations and International Conflict. International Studies Quarterly, Vol. 45, No. 2, pp. 317–338. Holsti, K. J. (1992). International Politics: A Framework for Analysis, New Jersey: Prentice Hall. Howard, M. E. (1978). War and the Liberal Conscience, New Brunswick: Rutgers University Press. Huntington, S. P. (1993a). The Clash of Civilizations? Foreign Affairs, Vol. 72, No. 3, pp. 22–49. Huntington, S. P. (1993b). If Not Civilizations, What? Foreign Affairs, Vol. 72, No. 5, pp. 186–194. Huntington, S. P. (1996). The Clash of Civilizations and the Remaking of World Order, New York: Simon & Shuster. International Civil Aviation Organization (1998, September). On-Flight Origin and Destination, Montreal: ICAO. International Monetary Fund (1993). A Guide to Direction of Trade Statistics, Washington: IMF Statistics Division. Communication and International Conflict 163 International Monetary Fund (1998). Direction of Trade Statistics, University of Manchester. International Peace Research Institute – Oslo. (2005). The PRIO/Uppsala Armed Conflict Dataset, version 3. Jones, R. J. B. (1995). Globalisation and Interdependence in the International Polit- ical Economy: Rhetoric and Reality, London: Pinter Publishers. Jones, D. M., S. A. Bremer, and J. D. Singer (1996). “Militarized Interstate Disputes, 1816–1992: Rationale, Coding Rules, and Empirical Patterns.” Conflict Manage- ment and Peace Science, Vol. 15, No. 2, pp. 163–213. Kant, I. (1795). Perpetual Peace (reprinted in 1957), New York: Liberal Arts Press. Karatnycky, A. (2001). Freedom in the world: The annual survey of political rights and civil liberties 2000–2001, New Brunswick: Transaction. Kegley, C. W. and E. R. Wittkopf (2006). World politics: Trend and Transformation, CA: Thompson Wadsworth. Keohane, R. O. and J. S. Nye (1977) Power and Independence: World Politics in Transition, Boston: Little, Brown and Company. Keohane, R. O. and J. S. Nye (1998). “Power and Interdependence in the Informa- tion Age,” Foreign Affairs, Vol. 77, No. 5, pp. 81–94. Kim, K., and G. A. Barnett (2000). “The Structure of the International Telecommuni- cations Regime in Transition: A Network Analysis of International Organiza- tions.” International Interactions, Vol. 26, pp. 91–127. Kinsella, D. (1998). Arms Transfer Dependence and Foreign Policy Conflict. Journal of Peace Research, Vol. 35(1), pp. 7–23. Kocs, S. (1995). “Territorial Disputes and Interstate War, 1945–1987.” Journal of Pol- itics, Vol. 57, pp. 159–175. Krackhardt, D. and L. Porter (1986). “The Snowball Effect: Turnover Embedded in Communication Networks.” Journal of Applied Psychology, Vol. 71, pp. 50–55. Lake, D. A. (1992). “Powerful Pacifists: Democratic States and War.” The American Political Science Review, Vol. 86, No. 1, pp. 24–37. Lemke, D. and W. Reed, (2001). “War and Rivalry Among Great Powers.” American Journal of Political Science, Vol. 45, pp. 457–469. Lemke, D. and S. Werner, (1996). “Power Parity, Commitment to Change, and War. International Studies Quarterly, Vol. 40, pp. 235–260. Levy, J. (1989). The Diversionary Theory of War. In M. Midlarsky, ed., Handbook of War Studies, Boston: Unwin Hyman, pp. 259–288. Levy, J. S. (2003). “Applications of Prospect Theory to Political Science.” Synthese, Vol. 135, pp. 215–241. Lumsdaine, D. (1993). Moral Vision in International Politics: The Foreign Aid Regime, 1949–1989, Princeton, N.J: Princeton University Press. Mansfield, E. D. (1994). Power, Trade, and War, Princeton: Princeton University Press. Mansfield, E. D. and B. M. Pollins, (2001). “The Study of Interdependence and Con- flict: Recent Advances, Open Questions, and Directions for Future Research.” Journal of Conflict Resolution, Vol. 45, pp. 834–859. Mansfield, E. D., and J. Snyder, (1995). “Democratization and the Danger of War.” International Security, Vol. 20, No. 2, pp. 5–38. 164 J. H. Kim and G. A. Barnett Mansfield, E. D., and J. Snyder, (1996). The Effects of Democratization On war. International Security, Vol. 20, No. 2, pp. 196–207. Maoz, Z., and N. Abdolali, (1989). Regime Types and International Conflict. Journal of Conflict Resolution, Vol. 33, No. 1, pp. 3–35. Maoz, Z. R. D. Kuperman, L. Terris, and I. Talmud (2004). Structural Equivalence and International Conflict, 1816–2000: A Social Networks Analysis of Dyadic Affinities and Conflict. Paper presented at International Network of Social Net- work Analysis Conference, Portoroz, Slovenia, May 12–16, 2004. Miller, R. A. (1999). “Regime Type, Strategic Interaction, and the Diversionary Use of Force.” The Journal of Conflict Resolution, Vol. 43(3), pp. 388–402. Milliken, J. (2002). The Social Construction of the Korean War: Conflict Possibilities, UK: Manchester University Press. Mintz, A., Redd, S. B. (2003). “Framing Effects in International Relations.” Synthese, Vol. 135, pp. 193–213. Morrow, J. D. (1999). How Could Trade Affect Conflict? Journal of Peace Research, Vol. 36, No. 4. pp. 481–489. Morrow, J. D. R. M. Silverson, and T. Tabares, (1998). “The Political Determinants of International Trade: The major powers, 1907–1990.” American Political Science Review, Vol. 92(3), pp. 649–661. Mosco, V. (1993). “Communication and Information Technology for War and Peace. In C. Roach, ed., Communication and Culture in War and Peace, California: Sage. Müller, H. M. (1999). Das Zusammenleben der Kulturen: Ein Gegenentwurf zu Huntington. Frankfurt am Main: Fischer Taschenbuch Verlag. OECD (2005). “International Development Statistics (IDS) Online Databases on Aid and Other Resource Flows.” Oneal, J. R. and B. M. Russett (1997). “The Classical Liberals Were Right: Democ- racy, Interdependence, and Conflict, 1950–1985.” International Studies Quar- terly, Vol. 41(2), pp. 267–293. Oneal, J. R., B. Russett, and M. L. Berbaum (2003). “Causes of Peace: Democracy, Interdependence, and International Organizations, 1885–1992.” International Studies Quarterly, Vol. 47(3), pp. 371–393. Organski, A. F. K. and J. Kugler (1980). The War Ledger, Chicago: University of Chicago Press. Pevehouse, J., T. Nordstrom, and K. Warnke (in press). Inter-Governmental Organi- zations, 1815–2000: A New Correlates of War Dataset. Polachek, S. W. (1980). “Conflict and Trade.” Journal of Conflict Resolution, Vol. 24, pp. 55–78. Polachek, S. W. (1992). “Conflict and Trade: An Economics Approach to Political International Interactions.” In W. Isard and C. H. Anderton, eds., Economics of Arms Reduction and the Peace Process Amsterdam: North-Holland, pp. 89–120. Polachek, S. W., J. Robst, and Y. C. Chang (1999). “Liberalism and Interdepen- dence: Extending the Trade-Conflict Model.” Journal of Peace Research, Vol. 36, No. 4, pp. 405–422. Ray, J. L. (1995). Democracy and International Conflict: An Evaluation of the Dem- ocratic Peace Proposition, Columbia, S.C.: University of South Carolina Press. Richards, W. D. Jr.(1995). The NEGOPY Network Analysis Program, Burnaby, BC: Department of Communication, Simon Fraser University. Communication and International Conflict 165 Rogers, E. M. and D. L. Kincaid, (1981). Communication Networks: T

A Structural Analysis of International Conflict From a Communication Perspective.pdf

Document Details

Tags

Related

Full Transcript

Upgrade to continue