Contending Theories of International Conflict A Levels-of-Analysis Approach JACK S. LEVY A glance through the table of contents of this volume reveals -how dramatically our perceptions of international conflict have changed since the end of the Cold War. Conventional concerns about the balance of power, alliances, arms races, deterrence, and superpower confrontation have given way to new concerns about ethnonationalism, religious militancy, environmental degrada- tion, resource scarcity, preventive diplomacy, peacekeeping, humanitarian intervention, and small-state conflict. The perception that the end of the Cold War has changed “all the answers and all the questions”! goes too far, however, and the theme of change must be tempered by that of continuity. Many of the factors that play a central role in con- temporary international ‘conflicts would have been quite familiar to Thucydides over two millennia ago, nd many ofthe “nontra~ I” causes of war discussed in this vol- paths that occupy a central place in the mainstream conflict literature. My aim in this essay is co put many of the factors discussed in the individual chap- ters in this volume into a larger theoretical context, which I organize around a “levels of-analysis” framework. I select some of the leading thearies of international conflict, identify their key variables, specify the causal paths by which they affect decisions for war and peace, highlight some of the key inter~ action effects among variables at different levels of analysis, and suggest which of these theories are likely to be most useful for understanding global conflict as we enter a new millennium. I include both traditional models of international war and new theories about ethnonationalism, population move- ‘ments, and environmental change. I aim to provide an overview of contending theories rather than detailed evidence, examples, or analytical critiques. I begin by discussing : 3 Scanned with CamScannerthe levels-of-a then conflict a nalysis framework, which I to organize a survey of theories of ind war? THe LEvets-of-Anatysis FRAMEWorK ‘The levels-of-analysis concept was first sys ‘ematized by Kenneth Waltz, who sug- Sested thar the causes of war can be found at the levels of the individual, the nation- State, and the international system. ‘The individual level focuses Primarily on human nature and instinct theories and on individ- ual political leaders and their belief systems, Personalities, and psychological processes The national level inéludes both govern- mental variables (such as the ‘structure of the Political system and the nature of the policy- making process) and societal factors (such ‘as the structure of the economic system, the tole of public pinion, economic and non- economic interest groups, ethnicity and nationalism, and political culture and ideol- ogy). Systemic-level causes include the‘anar- chic structure of the international system, the number of major powers in the system, the distribution of military and economic power among them, patterns of military alliances and international trade, and other factors that constitute the external environ- ment common to all states. Some scholars have modified Waltz’s framework by collapsing the individual and nation-state levels to create a simplified dichotomy of nation (or unit) level and sys- tem level.4 Others have created four or five levels by disaggregating the nation-state level into distinct governmental and societal factors, a practice that I follow here* At the systemic level, it is sometimes useful to distinguish among system characteristics all states, dyadic relationships common to S irs of states, and the external envi- Deas Ons Fh these ment of individual states, although t zon s are often blurred in the litera distinction: ‘ nuniber of Se es ee Jack S. Levy levels in the real world, for these alternative schemes are simply analytic constructions, which should be evaluated in terms of their usefulness in generating theoretically impor- tant and empirically valid hypotheses to help derstand the dynamics of global politics. Scholars generally use the levels-of- lysis concept as a framework for classify- ing independent variables that explain state foreign policy behaviors and international outcomes. This framework helps us ask such questions as whether the causes of war are found primarily at the level of the inter- national system, the nation-state, or the indi- vidual, and how variables from different lev- ls interact in the foreign policy process. But the levels-of-analysis concept is sometimes used in a different way, to refer not to the independent causal variables but instead to the dependent variable—that is, to the type of entity (individual, organization, state, or system) whose behavior is to be explained. In this second sense, the systemic level of analysis refers to explanations af patterns and outcomes in the international system, the dyadic level to explanations of the stra~ tegic interactions between two states, the national level to explanations of state for- ign policy behavior, and the individual level to explanations of the preferences, beliefS, or choices of individuals. In this usage, the state~ ment that democracies go to war less fre quently than do other states is a national- level hypothesis, whereas the statement that democracies rarely ifever fight each other is a dyadic-level hypothesis. The frequent fail- ure of scholars to be explicit about exactly how they are using the levels-of-analysis concept is a source of considerable confu- sion in the field. Te is logically possible, and in fact often desirable, to combine variables from differ ent levels of analysis in causal explanations. Independent variables from one level can be used to explain dependent variables at another level, and variables from several us. Scanned with CamScannerfags nie (CONTENDING THEORIES OF INTERNATIONAL CONFLICT Jevels can be combined in an explanation of a single dependent variable. Systemic or dyadic-level security threats can shape individual-level foreign policy preferences (ah the direction of hard-line policies to deal ith those threats, for example) and also nptional-level structures (toward an increas- ¥g Concentration of political power in state institutions, for example). In addition, national- or individual- Idvel variables can be the Primary causes of systemic- or dyadic-level outcomes. Domes- tic political pressures, bureaucratic conflicts ahd compromises, or the belief systems of individual leaders can be the main causes of spme wars. A logically complete explanation such cases, however, requires the analyst 1 go beyond the individual level to explain yw individual beliefs and preferences get nslated into state decisions and actions, d beyond the state level to explain how the actions of two or more states interact to Itad to war as a dyadic or systemic outcome. fo the extent that most wars generally in- ‘Ive the mutual and interactive decisions of tWwo or more adversaries, an explanation for 1¢ outbreak of war logically requires includ- ing dyadic- or systemic-level variables. sysTeMic-Levet SOURCES OF ATIONAY. CONFLICT ¢ traditional literature on the causes of far has been dominated by the realist para~ digm, a broad systemic-level framework hat incorporates several distinct theories. ‘ach of these theories assumes that the key ors are sovereign states that act rationally 9 advance their security, power, and wealth 4 ‘ning international i ieee tpothe- ~~ 2 Fees see CSHSCCCTCT ECS a ® ————=5 predictions about what those specific out comes are. In the realist world view, assumptions of an inherently conflictual world and uncer- tainties regarding the present and future intentions of the adversary lead political leaders to focus on short-term security needs and on their relative position in the system, to adopt worst-case thinking, to engage in a struggle for power, and to use coercive threats to advance their interests, influence the adversary, and maintain their reputa~ tions. Wars occur not only because some states prefer war to peace; they also occur because of the unintended consequences of actions by those states that prefer peace to war and that are more interested in mini- mizing their losses than in maximizing their gains, but that are willing to take consider- able risks to avoid losses.* Even defensively motivated efforts by states to provide for their own security through armaments, alli- ances, and deterrent threats are often per- ceived as threatening by others (the “security dilemma’), which leads to counteractions and conflict spirals that become difficult to reverse.” ‘The leading realist theory is balance of power theory, which posits the avoidance of hegemony as the primary goal of states and the maintenance of an equilibrium of power in the system as the primary instrumental goal. The theory predicts that states, in par- ticular great powers, will balance against those who constitute the primary threats to their interests and particularly against any state that threatens to secure a hegemonic position. Balance of power theorists argue that the balancing mechanism almost always works successfully to avoid hegemony, either because potential hegemons are deterred by their anticipation of a militiry coalition against them or because they are defeated in war after deterrence fails. There is a division and unresolved debate between “classical realists,” who arguc that stability (defined as Scanned with CamScannera ee - = — Jack S, Levy the absence of a major war) is further sup- Ported by the presence of a “balancer” state, the absence of permanent alliances or a ralnseel alliance system, and especially by a multipolar distribution of power; and eorealists,” who argue that bipolarity is more stable than multipolarity.° Balance of Power politics often generates a “checker- board” pattern of alliances, based on the old realpolitik idea that “the enemy of my enemy is my friend,” An important alternative to balance of power theory is “power transition theory,” a form of hegemonic theory that shares realist assumptions but that emphasizes the existence of order within a nominally anar- chic system.” Hegemons commonly arise and use their strength to create a set of polit ical and economic structures and norms of behavior that enhance the stability of the system at the same time that these struc tures and norms advance the security of the hhegemon. Differential rates of growth lead to the rise and fall of hegemons, however, and the,probability of a major war is great~ ‘est at the point when the declining leader i being overtaken by the rising challenger. Either the challenger initiates war to bring ies benefits from the system into line with its rising military pover or the declining pains a ive wat” to block ee while the chance is still available. Hegemonic theory includes the dyadic-level “power preponderance” hypo thesis, widely confirmed in the empirical lt Na i east likely when one state archical equilibrium ‘theory each suggest that the erosion or collapse of hierarchical authority within an empire or bloc system will increase the probability of violent con- flict between political entities or communal groups in that system, especially if they have a tradition of rivalfy. In fact, the very exer- cise of imperial authority may contribute to postimperial conflicts among its constituent units. Imperial authorities going back to Caesar have exploited divisions or historic rivalries among weaker units within the empire in order to facilitate their own impe- tial control, and these rivalries can escalate to war once the imperial power is no longer able or willing to control such conflicts.!® Another theory in which national growth is an important variable is Choucri and North's “lateral pressure theory.” Growing population-and advancing technology gen— erate increasing demands for resources that cannot be satisfied by the state’s domestic resource endowments or by existing levels of foreign trade. These demands generate “lateral pressure” for access to external raw materials and marke's and often for political control through colonial expansion to guar- antee such access. Expansions by several states intersect, leading to pressures for states to defend their expanding interests through increased military expenditures and alliances, reciprocal actions by adversaries, conflict spirals, and an increasing probability of war? Although lateral pressure theory has not received as much attention as balance of power theory or power transition theory, its emphasis on population pressures and resource scarcities increases its relevance for era. ae vanable that afects exch of these theories is military technology, particularly the presence or absence Since ‘nuclear weay Dee ee ca reece continental delivery systems, an¢ biliz- able retaliatory forces was ¢ ma)ot ne ing factor during the Cold War be Scanned with CamScanner7 > y rey. CONTENDING THEORIES OF INTERNATIONAL ConFuCT it Beery raised the expected costs of War for all states, the stronger as well 2s the weaker, the initiator as well as the target. By facilitating direct “countervalue” strikes against enemy populations indepen- dent of the military outcome on the battle- field, the new technology allowed both weak: and strong nuclear powers to inflict unatceptable damage against the other. As Chirles De Gaulle noted, after a nuclear war|both sides would have “neither powers, nor laws, nor cities, nor cultures, nor cradles, nor|tombs.”!® This prospect of “mutual assured destruc made it virtually inconceivable that state could conclude that it would be » blurred the distinction between win- and losers, and strongly reinforced In addition, although mispercep- driven many wars in the past by erro- ty convincing political leaders that war d bring military victory with tolerable the certainty of destruction associated for such misperceptions on the part of the nutlear powers in the contemporary era.”? Aq US. President Reagan and USSR eral Secretary Gorbachev affirmed, “A lear war cannot be won and must never 720 ee of power hypotheses about bal- ing against primary security threats, ‘transition hypotheses about the con- ences of uneven growth, and lateral pres- hypotheses about resource demands and cersecting interests were developed with experi | Ser some key propositions of certain realist the- ories need to be modified for regional sys~ tems, which are nested within the larger global system and which are significantly influenced by the distribution of power and influence in that broader system.” ‘The impact of systemic structures on state behavior is not always obvious, however, for those structures may exert their influence by establishing “permissive conditions” for cer- tain types of behavior as well as by creating the conditions that more directly encourage behavior? Many analyses of the 1990-1991 Persian Gulf War, for example, neglect the causal impact of the end of the Cold War and the collapse of bipolarity, which created conditions that facilitated the American and allied responses. Had the Iraqi invasion of Kuwait occurred a few years earlier, the risk of Soviet intervention on behalf of its Iraqi ally would have made it much less likely that an Arab coalition would have formed against Iraq os that the United States would have risked military intervention. Although it is likely that conflicts involv ing economic interests, religion, and ethnic identity will play an increasingly important role in world politics now that the Cold War is over, hypotheses about the effects of these variables usually assume American global dominance and the absence of serious mil tary threats to the leading powers in the sys- tem. A change in the current distribution of power in the system (the rise of China or the resurgence of Russia, for example) would give rise to a rather different set of political dynamics on the global and regional level. Any analysis of the possible impact of ethno~ nationalism, international migrations, and environmental change must acknowledge the larger systemic structures within which these variables influence governmental poli- ies and external responses. ‘Another systemic-level theory, one that from a liberal rather than realist par and that also includes a substantial Scanned with CamScannera ae eS fac (5 Cov domestic component, is the liberal economic theory of war. The basic argument, which goes back to Smith and Ricardo, is that free trade within an international market econ- omy is the best guarantor of peace. A num- ber of theoretical arguments are advanced to support this proposition, One is the idea that trade promotes prosperity, following the law of comparative advantage, and that prosper- 8 promotes peace, in part because “people are too busy growing rich to have time for war.” Tariffs, quotas, and any other restric~ Hons on the natural operation of market mechanisms have the opposite effect. A related argument emphasizes the inter- vening role of interdependence rather than Prosperity: Because trade increases inter- dependence and creates mutual vulnerabil- ities to any interference with free trade, the fear-of the loss of the welfare gains from— trade deters political leaders from initiating militarized conflict. This assumes, of course, that trade is more efficient than military co- excion in promoting state wealth. As the foundations of wealth and power have shifted from territory to industrialization and now to knowledge-based forms of production, the value of territorial conquest has diminished, at least for the advanced industrial states. At the same time, the costs of territorial con- quest have increased, with growing cultural self-assertiveness and nationalist opposition to political control, the incompatibility of political control with the liberal foundations of postindustrial economic innovation and productivity, the growth of antiwar attitudes in the West, the increasingly powerful mili- tary capabilities of many medium-size states, and the development of nuclear weapons. Liberals conclude, therefore, that trade is economically efficient and that it promotes i states c they frequently promote war. One trading partner may be tempted to engage in eco- nomic coercion to exploit the adversary’s vulnetabilities and influence its behavior relating tg security as well as economic issues, which can lead to retaliatory actions, conflict spirals, and war. These tendencies are reinforced by demands for protectionist pressures from domestic economic groups that are especially vulnerable to external developments, particularly in bad economic times. Whether the incentives for the gains from trade dominate the incentives for coer- cion or protection based on economic asym- metrics, and whether the latter escalate to trade wars and militarized conflicts, is an empirical question that analysts have only recently begun to analyze systematically, and the results are rather inconclusive.?® ~ These economic relationships are likely to become increasingly important sources of international conflicts in the future, particu- larly for developing states that face serious economic scarcities, but also for the advanced industrial states. As military security threats to these states have diminished significantly with the end of the Cold War, economic competition among them has become more and more salient, particularly as some states have come to define economic primacy as a vital national interest27 Te is conceivable that economic rivalries, fueled by parochial domestic interests and by hard-line publics sympathetic to the appeals of economic nationalism, could escalate into strategic rivalries and then possibly into crises or military confrontations. Although. the militarization of commercial rivalries was an important path to war between great powers in earlier. times, particularly in the seventeenth century, this pattern is much less likely to repeat itself in the future, at Jeast among the leading industrial states. ‘The increasing destructiveness of military technology has weakened the link between economic strength and military power, and Scanned with CamScannert (>) thk decline of mercantilist economic philo- \* sophies and the growth of complex inter~ ddpendence further undermine the tradi- tignal belief that military force might be a Pa _ useful instrument of state policy to advance the economic interests of society. Still, con > teinporary economic rivalries are sufficiently intense and the consequences of their possi- ble militarization sufficiently profound that tifis issue remains a vital area for future rasearch.2* The sources of state behavior and inter- nptional patterns in balance of power theory, power transition theory, lateral pressure the- of, and liberal economic theories of trade and war are found primarily at the system y _Idvel, although domestic demographic and efonomic variables clearly shape the differ- chtial rates of national growth that deter- ine the distribution of power in the system, ahd liberal theories do recognize the domes- tic factors that affect the terms of trade bree states, the ideologies that help ape state interests, and parochial domestic io that constrain state behavior. With the exception of some die-hard realists, how- ever, scholars have increasingly recognized at systemic structures fail to explain a sig jficant amount of the variance in the out eak or expansion of international conflict. is conclusion has led to increasing chal- lenges to realism and other systemic-level theories from theoretical perspectives linked o the societal, bureaucratic- organizational, .d individual levels. The argument is not bnly that these other levels of analysis will become increasingly important in the farure, but also that their influence orsinternational ehavior in the past has been seriously BD CONTENDING THEORIES OF INTERNATIONAL CONFLICT — tional-conflict after decades of neglec=. by political scientists, although at this point we have collections of hypotheses rather than well-developed and integrated theories. While interest in Marxist-Leninist theories has waned, scholars have given enormous attention to the relationship between demo- cracy and war and to the ways in which political leaders resort to external military force as a means of bolstering their domes- tic political support. In addition, many of the newly emphasized causes of war— including ethnonationalism, environmental scarcity and degradation, and population movements—either fit into the societal level of analysis or include that level as a major step in the hypothesized causal chain lead ing to war.? Marxist-Leninist theory focuses on the domestic economic structure of capitalist societies and posits that the inequitable dis- tribution of wealth generates undercon. sumption, inadequate domestic investment, and stagnant economies. This leads to expan sionist and imperialist foreign policies to secure external markets for surplus products, external investment opportunities for sur- plus capital, outlets for surplus population, and access to raw materials at stable prices, each of which can result in colonial wars and an escalation in the conflict between capitalist states °° Capitalist states also require high military spending to stabilize and stimulate the economy, and this can lead to arms races, conflict spirals, and wars? ‘There have been numerous critiques of the logical consistency and empirical valid- ity of the Mandist-Leninist theory of for- cign policy and imperialism, and these have Jed to alternative theories based on the inter ests of great powers (regardless of economic system) or of parochial military clites.? One common criticism focuses on economic determinism and the absence of political variables to explain the linkages between underlying social and economic conditions Scanned with CamScannerto —__ ee eet a and the state policies that result, One the- ee ofimperial expansion that better inte Sere ou economic variables is Pee ae tion formation and fs Snyder observes that states often expand yond the point at which thei, ial eee cir imperial is can be Supported by available Tesources. He rejects atte sou ejects attempts to explain this in terms of the interests ofa single class or elite and focuses instead on a coalition of clites, each of which prefers a different form of limited expansion, military buildup, or economic autarky. These groups create lop. rolled coalitions that secure power at the cost of a more expansionist foreign policy than is desired by any single group or that can be supported by available resources. The coalitions reinforce their positions of power and rationalize their policies by propagating self-serving strategic myths that masquer~ ade as lessons of history. These dynazaics of overexpansion are most likely to arise in cartelized political systems and least likely to occur in democratic systems, where diffuse interests and the absence of information monopolies work against strategic myth- making. Snyder applies his model to the great powers in the industrial era (the coali- tion of iron and rye in Germany in the late nineteenth century, for example). The the- ory can also be applied to other states in other contexts, however, and it continues to be quite relevant for the future. Individuals and groups support expan- sionist policies both because it is in their parochial interests to do so and because of the “rally ‘round the flag” effect created by symbolic politics and the images of the enemy fostered by political pare some conditions this tempts oi = Jack S. Levy ago, for-example, Bodin argued that “the best way of preserving a state, and guaranteeing it against sedition, rebellion, and civil war is to...find an enemy against whom [the sub jects] can make common cause.” This hypothesis is theoretically grounded in social identity theory and the in-group/out- group hypothesis, which suggests that con flict with an out-group increases the cohe- sion of a well-defined in-group. Although the diversionary theory of war appears to fit a number of historical cases, internal political insecurity does not always lead to external scapegoating, which raises the question of the conditions under which this is most likely to occur. Recent studies have begun to specify theoretical models that link domestic politics and the external use of force and to identify empirically the internal conditions under which the use of force is most likely. Among these conditions are low to moderate levels of domestic polit- ical support and legitimacy, poor economic performance, and the perception that a diplomatic or military victory is feasible with minimal costs. Another possible vari- able is regime type,and some have suggested that because of their electoral accountability, democratic leaders may have greater incen- tives for scapegoating than do authoritarian leaders. The incentives for scapegoating must be balanced against its potential costs, however, and recent evidence suggests that democratic leaders who initiate wars, partic~ ularly losing wars, are thrown out of power much more frequently than are nondemno- cratic leaders. In fact, one can fact find many examples of démocracies engaging in the diversionary use of force against other states. What we do not find, however, is external scapegoating by one democracy against another democ- racy that escalates to war. A striking finding, of recent research, consistent with Kant’s “pacific union” among democracies, is that democracies rarely if ever go to war with Scanned with CamScanner nmefees other.37 Elsewhere, I describe this | absence of war between democracies as coming “as close as anything we have to an empirical law in international relations.” Jt is contrary to realist theory, which posits | that the effect of systemic structures does | not vary with regime type and implies that democratic dyads will go to war as often (proportionately) as any other pairs of states. Itis true that democracies occasionally | getinvolved in crises and use limited amounts | of force against each other, fight imperial | wars, and once in war adopt a crusading spirit and fight particularly destructive wars, as Kant predicted. The evidence suggests | that democracies tend to get involved in wars as frequently as do nondemocratic states. But they rarely if ever fight each other, and there are few (if any) unambiguous cases of actual wars between democracies. It has also been demonstrated, at least for the period since 1945, that the absence of war between democracies is not the spurious result of economic or geopolitical factors correlated with democracy.” Scholars have proposed two types of models to explain the dual finding that | democracies engage in frequent wars but | that they rarely, if ever, fight each other. In | the cultural or normative model, the norms of peaceful conflict resolution that have evolved within democratic political cultures are extended to relations between demo- cratic states. In the structural or institu- tional model, checks and balances, the dis- persion of power, and the need for public debate make it more difficult for democratic states to use force against cach other. In both models, there arc fewer constraints on en ke ° SHEKERETUE “a UV OU COUTTS cers | CONTENDING THEORIES OF INTERNATIONAL CONFLICT provides additional incentives for democra~. tic regimes to use force against authoritarian regimes to eliminate their violent tendencies. The democratic peace has important policy implications, for it suggests that by promating the development of democracies around the world, the United States can contribute to the elimination of war as well as to the establishment of liberal institutions and political freedom. One recent study questions this prescription, however, argu- ing that although well-established derno- cratic dyads are peaceful, the process of transition to democracy is a particularly war-prone time for states and that democ- Fatizing states occasionally go to war against each other.” Although the empirical validity of this study has been challenged,*? its theoretical argument is certainly plausible: The democ- ratization process brings new social groups with widely divergent interests into the political process at = time when the state lacks the institutional capacity to integrate conflicting interests and respond to popular demands, which creates cnormous social conflict. This is exacerbated if democrati- zation is coupled with the introduction of market forces into nonmarket economies, which leads to popular pressures for state protection against the pain of economic adjustment. Elites competing for mass polit- ical support are tempted to make nationalist appeals and engage in external scapegoating in order to bolster their internal support. This scapegoating is particularly appealing to those elites whose interests are threatened by the democratization process and who be- lieve that the symbol of an external enemy might allow political leaders to reverse that process and further centralize political power. Given the intensity of the emotions associ- ated with ethnic loyalties, which are exacer- Dated by the process of democratization, ethnic rivals make particularly good targets Scanned with CamScannerCCSCHDODOEEdE ie cot Although scholars began to devote;more attention to the sources of ethnonatjorialism. during the 1980s, they said relatively little about its effects on international politics and on its possible linkages to international war in particular. With the explosion of eth- nic conflicts after the end of the Cold War, scholars are now beginning to cxamine how different types of ethnic and communal conflict can lead to intranational and inter- national warfare and the conditions under which each is most likely to occur. In fact, some suggest that the primary sources of conflict in the fature will be cultural, rather than ideological or geopolitical, and that a “clash of civilizations” between ethnically and religiously defined peoples will be the pri- mary source of international conflict in the future. Cultural self-awareness is increasing, and conflicts over culturally defined identi ~ ties are more difficult to resolve through compromise than are conflicts over tangible strategic or economic interests. ‘Although “ancient hatreds” explanations for cthnonational conflicts—in which the collapse of authoritarian rule in Eastern Europe, the former Soviet Union, and else- where rernoved the “lid” on ancient rivalries and allowed peoples to attempt to settle long-suppressed grievances—are still popu- lar among journalists, many scholars have been skeptical. The “ancient hatreds” hypo- thesis fails to explain why violent ethnic conflicts have broken out among some eth- nic communities but not others, when those violent conflicts occur, and how intensely they are fought.‘ Concern has now shifted toward identifying the conditions that ex- _ plain the variation in the incidence and the eee amen onesie iene. o — Jack S. Levy with such threats, so that nationalism is most likely to arise when state structures have weakened or collapsed entirely, as in the case of Eastern Europe and the former Soviet Union.*® A number of other factors also affect the security of various national groups and therefore the intensity of nation alism and the likelihood of ethnic conflict. One factor is ethnic demography and geography. The greater the congruence be- tween state territorial borders and commu~ nal boundaries, the fewer the number of stateless nations seeking statchood, the greater the legitimacy of existing states, and the lower the likelihood of ethnic conflict. ‘This is facilitated by large, compact ethnic groupings, for the larger the size of an ethnic community, the greater its security against outside threats. Dense ethnic intermingling creates potentially unstable_ multi-ethnic states and leaves many ethnic minorities outside of the boundaries of their nation states. This increases the likelihood of seces- sionist wars by captive nations seeking to withdraw from the territory of a larger state and create their own state, and of irredentist wars by ethnic groups in one state seeking. to retrieve ethnically kindred people and their territory from another state.” ‘The likelihood of the internationalization and expansion of wars of secession and na- tional unification increases if they threaten to alter the balance of power at the regional or global level. This creates incentives for external states to intervene to ensure the fur- ther weakening of a potential enemy, block the further disintegration of an important ally, or exploit the window of opportunity to make gains while its enemy is divided. Thus systemic- and societal-level variables can combine to make ethnonational conflict particularly destabilizing for the interna-. tional system. In addition to these structural consider political leaders may deliberately -nationalist fervor. They may invoke Scanned with CamScanner Ana a a aeSSB" € B DIG , | the symbols of the nation and engage in | ethnic-bashing scapegoating because it facil- | itates the mobilization of mass armies, adds . co the military power of the state, and helps to justify the economic sacrifices necessary to support costly defense efforts.° Or, they may be more concerned with domestic chal- lenges to their leadership positions and gamble that scapegoating will help rally the public and bolster their domestic support.5! In cither case, ethnic scapegoating and the | use of self-glorifying and othes-depreciating myths further intensify nationalist passions | Chauvinist mythmaking is fueled further by | past legacies of suffering at the hands of others, because memories of past genocides, murders, and population expulsions acquire great symbolic value, create incentives for revenge, and block opportunities for conflict resohation.*? Once political leaders appeal t | the symbols of nationalism and use nation. alist myths to help sell policies domesticall they can become entrapped by their ow! thetoric—entrapped politically by over! selling their policies to the public and creat ing new domestic coalitions in support o| those policies, and entrapped psychologi cally by coming to internalize the beliefs that they advocate. Once established, enemy! images are highly resistant to change, which farther undercuts the possibilities of com- promise with external adversaries. As Mor- genthau notes with respect to the rise of ideologies in the twentieth century: “Com- promise, the virtue of the old diplomacy, becomes the treason for the new.” Ethnonational conflict may also contrib- ute to international migration, which has | CONTENDING THEORIES OF INTERNATIONAL ConrticT — conflict within and between states through 13 injnorities to migrate in search of security or to join their national. homelands. Large~ scale population movements may be the deliberate aim of governmental policy as well as the unintended consequence of com- munal conflict, as demonstrated by the recent history of ethnic cleansing in the Balkans. Governments sometimes adopt forced emigration as a strategy of achieving cultural homogeneity or the dominance of one ethnic community over another, climi- nating political dissidents, colonizing arcas beyond borders, scapegoating against a pros~ perous bur unpopular ethnic minority, or destabilizing or influencing another state.°> There are also economic and environ- mental sources of international migration. Substantial differentials in income and employment opportunities create economic incentives for people to migrate in search of economic security. In the contemporary era, the collapse of authoritarian regimes has reraoved some of the political barriers to population movements across borders, and ‘expanding global communication and trans- portation networks have created new oppor- tunities for international migration. In addi- tion, environmental degradation, droughts, floods, and famines generate large numbers of “environmental refugees.” Population movements can contribute to a number of different causal paths.°& Migration can: (1) put added strain on scarce resources in the host (receiving) country, particularly in large urban areas; (2) change land distribution, economic relations, and the balance of political power among ethnic, religious, or other social groups; (3) under ‘mine state capacity to create markets and, other institutions that facilitate adaptation mental change; (4) generate al ed threat to the host country’s cul- ity; (5) trigger a social backlash ous people in response to per- its tO economic security or social Scanned with CamScanner\ identity from migrants; and (6) generally increase commutial conflict, political insta~ bility, and the likelihood of civil strife. J Migrations may also contribute to inter- ( pational conflict by serving as a focal point | for relations between home and host coun Ses. Host countries that cannot easily assimilate the new immigrants or deal with the consequent economic problems and social instability may attempt to influence the home government to stop or slow the flow of reflgces or to climinate the condi. Sons that gave rise to them. If cooperative efforts fail, host countries may resort to coercive threats to achieve these goals, and in the last resort to military intervention to block the flow of refugees. Hard-line actions against the migrants’ home country can also Serve as a useful scapegoating strategy for political leaders who want to bolster their internal political support. In addition, efforts by home governments to stop unwanted emigrations can contribute to conflict with the host state57 Environmental change is one of several sources of large-scale population move- ments, but it can also contribute to inter- national conflict through other causal paths; in fact, the implications of environ- mental change for national security and {ternational conflict have recently attracted considerable attention in the literature. A fine neo-Malthusian perspective suggests that rapidly growing and increasingly urbanized populations competing for scarce resources—and the depletion of those re~ sources by land degradation, desertifica- tion, deforestation, rising sea levels, and pollution—will generate economic and social problems, environmental refugees, instabili erious domestic TI sation is n , an ‘The combination of population growth, uneven resource distribution, and the envi- ronmental degradation of scarce resources leads to scarcities that can contribute to violent conflict through a number of dif- _ ferent causal paths. The most direct path is through “simple-scarcity” conflicts or “resource wars” between states.“ This hypo- thesis is intuitively plausible and is consis- tent with mercantilist theories of economic Power, with Choucri and North's lateral pressure theory, and with Marxist-Leninist theories of imperialism. It also derives em- Pirical support from a number of individual historical cases, including the 1979 “Soccer War” between El Salvador and Honduras, the 1941 Pacific War between Japan and the United States, the 1990-1991 Persian Gulf War, and the seventeenth-century Anglo- Dutch “herring wars,” among-others. It is true that a number of factors com- bine to reduce the likelihood of resource wars in the future. The growing inter- dependence of the world economy makes it easier for states to satisfy their external needs for resources without political or ter~ ritorial control over those resources, and changing technology has increased the range of domestic substitutes for key natural resources, Also, as noted before, the strong incur growing military, diplomatic, and domestic political costs if they forcefully expropriate resources from the weak. These factors may be less compelling for nonwest- em states that do not share the liberal inter- nationalist view, and far more research is necessary before we can conclude that resource wars are an unlikely path to war in the future. Although Homer-Dixon may go too far in concluding that the resource war scenario has been relatively rate in the past, he may be correct that the primary impact of envi- lronmental scarcity"is through its social ts.© Scarcity often leads to declining [standards of living, increasing perceptions of Scanned with CamScanner iCONTENDING THEORIES OF INTERNATIONAL ConsticT Pea 4 zero-sum game among different social groups, and attempts by these groups to pass the costs of economic decline onto others, This can result in increasing class conflict and social discontent, challenges to the legit- imacy of the regime, pressures on demo- cratic political institutions and free-market systems, and an increasing probability of civil strife or external scapegoating. BUREAUCRATIC-ORGANIZATIONAL ‘SOURCES OF INTERNATIONAL CONFLICT The bureaucratic politics and organizational processes models of forcign policy decision making retain the rial aeomption of realist theory but reject WS-WATary acbor Seer a ee Cig aeaiea oe actors indifferent bureaucratic roles within the executive branch of the governinent havedferent interes and Giferent dope of political power. The policy preferences of each of these actors are often influenced more by the i onga- nization than by the national interests of the state, and foreign policymaking is funda- mentally a political process that involves pulling, hauling, and political compromise among these key actors. Bureaucratic politics and organizational processes can contribute to war through a variety of causal paths. In spite of the popu lar argument that politically powerful mili tary organizations shift the internal balance of power toward more hawkish policies, most of the evidence suggests that military ‘organizations in most states are generally no more hawkish than their civilian counter- parts in decisions for war but are more hawk- ish with respect to the escalation of war.‘ is the military on decisions ct.M etal eee erik eae, eee ) ® ato severity of external threats and the necessity 6f enhanced military preparedness. This often leads to arms buildups that go beyond what is required by real security concerns, and these can trigger arms races, conflict spirals, and war. The dangers of escalation are increased by the standard operating pro- cedures of military organizations, which increase the rigidity of policy and deprive Political leaders of the flexibility they need to manage a crisis in a way that avoids a war while preserving vital interests.” In addition, military organizations gen- erally have a preference for offensive doc~ trines, which help rationalize larger military budgets, enhance military morale and pres tige, and facilitate seizing the initiative, structuring the battle, and hence reducing uncertainty. Offensive doctrines contribute to war by increasing incentives for territorial conquest, preemptive strikes, and preventive war by fueling arms races and conflict spirals because they increase the threat to adversary security, and by increasing the destructive ness of war. Although these hypotheses are quite plausible, the central concept of an offensive advantage is extraordinarily diffi- cult to define conceptually, and the magni- tude of its impact has never been established by systemic empirical analysis. Moreover, whatever the impact of this factor in previ- ‘ous eras, that impact has probably declined since World War Il as a result of the nuclear revolution, the evolution of postindustrial economies, and the growth of democracy and an open press.” InpiyiD JAL-Levet Sources oF INTERNATIONAL CONFLICT ~ Individual-level theories trace international conflict to individual political leaders, the content of their beliefs, the psychological processes through which they acquire in- formation and make decisions, and their yalitics and emotional states. Unlike Scanned with Cam Gos ScannerHe fac S. Levy theories at other levels, these theories sume that key decision makers often vary ta their preferences for foreign policy goals, their images of the adversary, or their beliefs about the optimum strategies to achieve their goals and meet those threats.” As a” result, different decision makers with differ- ut belief systems or “operational codes” about world politics will respond differently under similar situations, so that individuals sake a difference in state foreign policy behavior”! These beliefs arise from differ- ences in political socialization, personality, education, formative experiences, the les- sens people learn from historical experience, and a host of other variables.”2 The beliefs of political leaders cannot alone explain foreign policy behavior, how- ever, for beliefs have little impact unless those holding beliefs are in a political posi “tion to implement them. Theories of foreign policy and international conflict that recog- nize the importance of individual beliefs ‘must incorporate not only the beliefs them- selves, but also an explanation of how the policy preferences of different individuals in different roles and positions of power get aggregated into foreign policy decisions for the state. This facilitates the construc- tion of richer and more descriptively accu- rate theories of international conflict, but these theories are more complex, less elegant or parsimonious, more demanding in terms of the types of data that are necessary to test them, and less powerful in terms of their generalizability across different states in dif- ferent situations at different times. Forcign policy behavior can be Rie etterecsirestion of the adversary’s re- solve by an aggressive state may lead it to by their prior beliefs as cuch as by the objective evidence, that they utilize heuris- tics or cognitive shortcuts in lieu of more normatively rational decision rules, and that these heuristics can produce some rather serious discrepancies between the percep- tions of individuals and the “real world.” I focus here on the consequences of misper~ ceptions for international conflict behavior, but not on the psychological processes that generate these misperceptions. Misperception-based explanations are often seen as necessary causes of wars, on the assumption that if both sides had correctly assessed the adversary’s intentions and the likely outcome of the war, they could have agrecd to a settlement commensurate with that outcome while avoiding the costs of fighting.”* The concept of misperception is _extremely-difficult to-defitre analytically or measure empirically, however, and there are a plethora of types of misperceptions, but the ones most likely to have a major impact on the processes leading to war are misper- ceptions of the capabilities and intentions of adversaries and third states.” Exaggeration of the hostility of the adversary’s intentions is particularly impor- tant. In the short term it can induce one to take counteractions (in the extreme case, a preemptive strike) that trigger a conflict spiral and unnecessary war, and in the long term they can lead to an arms race or system of alliances and counteralliances. Under- estimation of the adversary’s hostility by a status quo state can contribute to war by undercutting the need to build up military capabilities in the long term or to demon- strate resolve in the short term, either of which undermines deterrence. Alternatively, € more coercive military threats in the tation that the adversary will back which results in a conflict spiral. of the adversary’s intentions Scanned with CamScannerCONTENDING THEORIES OF INTERNATIONAL CONFLICT —— may derive liom secondary misperceptions of the adversary’s value structure, its defini- sien f its vital interests, its definition of the sinuation, its expectations about the future, and the domestic or bureaucratic constraints on its freedom of action. Misperceptions of adversary capabilities can also be critical. The underestimation of adversary capabilities relative to one’s own generates military overconfidence and the common belief that a rapid military victory involving minimal costs is quite likely. The overestimation of adversary capabilities may lead one to overreact and initiate an arms buildup that is followed by an arms race and conflict spiral. Alternatively, it can lead to excessive passivity that undermines deter- rence. Misperceptions of the intentions and capabilities of third states can have a similar effect. The most common tendency is to exaggerate the likelihood than one’s poten- tial friends will intervene on one's behalf and the likel:hood that one’s potential ene- mies will stay neutral, and cither of these beliefs reinforces military overconfidence. Misperceptions of third-state capabilities, and hence the impact of third states on the course of the war should they choose to intervene, have similar effects. There is little reason to believe that the role of individual belief systems or psycho- logical processes has been significantly the end of the Cold War. We might hypo- authoritarian regimes has somewhat de- creased the importance of the beliefs or psy- single decision maker by Sevve geuwd influenced by the nuclear revolution or by thesize, however, that the recent decline of » ® 17 the situation. In addition, the greater the complexity, the greater the variation in beliefs, images, and perceptions across indi- viduals, and thus the greater the importance of the processes through which preferences are aggregated at the small-group and orga- nizational level Conctusion I have reviewed some of the leading theo- ries of international conflict, specified key variables and the causal paths that lead to war or peace, identified some of the inter- action effects between variables at different levels of analysis, and commented on the relevance of various theories as we move into the next century. It is more difficult to reach definitive conclusions regarding the likely causal importance of different levels of analysis for international conflict in the coming decades. First of all, assessing the relative impact of different variables at dif- ferent levels is an empirical as well as.a the- oretical question, and such a task is beyond the scope of this chapter. A more important conceptual problem is that theories of inter- national conflict have increasingly begun to incorporate variables from several different levels of analysis, so that an evaluation of the validity of a particular theory is not nec~ scarily congruent with the evaluation of the importance of a particular ievel of analy- sis. In addition, although the levels-of- analysis framework serves as a useful orga- nizing framework, we must not apply it so rigidly that it distracts attention from the important task of understanding how vari- ables at different levels of analysis interact in the processes contributing to interna- tional conflict. We must also recognize that the “impor- tance” of various levels of analysis is itself a ‘somewhat ambiguous concept. Theories rmltiple purposes, and variables at dif- levels of analysis may be more useful Scanned with CamScanner €eum see ood és’ &|18 — for some theoretical purposes than for others. The trade-off between the analytic power, generalizability, and predictive value of parsimonious theories and the descriptive richness of more complex theories is parti _ Ularly salient. If we want a general theory that can provide maximum explanatory power actoss different temporal and spatial contexts and. generate predictions about unobserved events, then variables based on individual-level beliefs and personalities or governmental-level decision processes are unlikely to be very helpful, because they are so difficult to operationalize and measure for a large number of cases. On the other hand, if we want a theory that can guide a more detailed interpretation of a small number of key historical cases, decision making variables (in conjunction with oth- ers) are likely to be extremely useful. A few more substantive comments are in order. Although systemic distributions of power will continue to be important in pro- viding the context within which regional systems and dyadic rivalries opcrate, and although balancing against primary threats and adjusting to changing power differen- tials through alliances and armaments will continue to be central themes in interna- tional relations, realist theories are probably too limited theoretically and too tied to the great-power experience of the past to provide an adequate explanation of international conflict over the next several decades. The locus of conflict has clearly begun to shift away from the great powers and away from the West. Although realist theories help to explain the origins of this shift, they need to be broadened if they are to explain the dynamics of conflict in a changing world. In terms of levels of analysis, far more Jac S. Leve of state leaders and the constraints on their actions. Ethnonational and other identity- based conflicts will be particularly impor tant in influencing the political agendas of states and in contributing to domestic instability and international tensions. The importance of economic variables is also likely to increase, for several reasons. The collapse of the Soviet Union has signifi- cantly reduced a major source of military threat to the leading economic powers in the system, and this, along with the declin- ing utility of military power, has increased the importance of economic competition among advanced industrial states.” In addition, resource scarcities—exacer- bated by environmental degradation and resulting population migrations —will likely threaten the economic welfare of many de- veloping societies, while processes of democ~ ratization will increase popular expectations of a minimum level of economic prosperity and also the influence of parochial eco- nomic groups. The combination will put enormous pressure on domestic political systems, many of which have only recently begun to develop the institutions to aggre- gate demands from different groups in society. In addition, the erosion of the order and certainty of the Cold War will increase internal conflicts over foreign policy goals and the means to achieve them and will therefore increase the importance of indi- vidual belief systems and the political pro- cesses through which individual policy pref- erences are aggregated into state forcign policy decisions. Much of this is speculation,-of course. Scholars had enough trouble predicting the end of the Cold War, which some now see in retrospect as overdetermined. Forecasting the future in a more complex and chaotic is an even more daunting task. But makes it all the more important that r attempts to understand world politics by well-developed theoretical Scanned with CamScanner a mmgD Typ x » » » » » 5 CONTENDING THEORIES OF INTERNATIONAL CONFLICT. ————————————_ 19 _ frameworks that help to illuminate and structure this coniplexity. Notes 1 thank Lori Gronich for helpfal comments on an earlier draft of this chapter. 1. Charles W. Kegley, Jr, “The Neoidealist Moment in International Studies: Realist ‘Myths and the New International Studies,” International Studies Quarterly 37, no. 2 (June 1993): 141. 2. This survey builds on my longer review essay, “The Causes of War: A Review of Theories and Evidence,” in Bebavior, Society, and Nuclear War, vol. 1, ed. Philip E. Tetlock, Jo L. Husbands, Robert Jervis, Paul C. Stern, and Charles Tilly (New York: Oxford University Press, 1989), pp. 209-333. 3. Kenneth N.Waltz, Man, the State, and War (New York: Columbia University Press, 1959). 4. J. David Singer, “The Levels-of- Analysis Problem ir, International Relations,” in Inter- national Politics and Foreign Policy, rev. cd., ed. James N. Rosenau (New York, 1969), pp. 20-29; and Kenneth N. Waltz, of International Politics (Reading, Mass.: Addison-Wesley, 1979).. 5. Robert Jervis, Perception and Mispercep- tion in International Politics (Princeton, NJ: Princeton University Press, 1976), chap. 1; and James N. Rosenau, The Scientific Study of Forcign Policy (New York: Nichols, 1980), chap. 6. 6. This does not necessarily mean that these variables have a greater causal influence than do individual or domestic variables. 7. Waltz, Theory of International Politics; and Robert O. Keohane, ed. Neoreafism and Its Critics (New York: Columbia University Press, 1986). 10. Hans J. Morgenthau, Politics Among Nations, 4th ed. (New York: Knopf, 1967); Edward V. Gulick, Europe's Classical Balance of Power (Ithaca, N.Y.: Cornell University Press, 1955); Waltz, Theory of International Politics, chap. 6;John Mearsheimer, “Back to the Future: Instability in Europe After the Cold War,” Inter- national Security 15, no. 1 (Summer-1990): 5-56; and Levy, “Causes of War,” pp. 228-242. 11. Stephen M. Walt, Tbe Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1987). ‘Thus Churchill commented, “If Hitler invaded Hell, I would make at least a favorable reference to the Devil in the House of Commons.” Winston Churchill, The Second World War, vol. 3: The Grand Alliance (Boston: Houghton Mifflin, 1950), p. 370. 12. A.F.K. Organski and Jacek Kugler, Te War Ledger (Chicago: University of Chicago Press, 1980), chap. 1 and Robert Gilpin, War and Change in World Politics (New York: Cam- bridge University Press, 1981). 13. The Israeli strike against the Iraqi nuclear reactor in 1981 is the classic preventive strike. Jack S. Levy, “Declining Power and the Preventive Motivation for War,” World Politics « 40, no. 1 (October 1987): 82-107. 14. Daniel S. Geller, “Power Differentials and War in Rival Dyads,” International Studies Quarterly 37, no. 2 (June 1993): 173-193. The logic is that under preponderance the strong are satisfied and do not have the incentives for war, and the weak, though dissatisfied, lack the capa- bility for war. Note that the stabilizing effects of power preponderance at the dyadic level do not necessarily imply that imbalances of power at the systemic level are stabilizing. 15. Manus I. Midlarsky, The Onset of World War (Boston: Unwin Hyman, 1988). 16. These phenomena are illustrated in the ‘case of the former Soviet empire and in. Yugo- slavia and are discussed in Mark Katz's essay on. ‘and to Japan before and after World answer the question of why resource ‘to military expansion in the 1930s n trade after World War Il). Nazli Scanned with CamScanner20 C. Choucri and Robert C. North, Nations in Conflict (San Francisco: W. H. Freeman, 1975); and Choucri, North, and Susumu Yamakage, The Challenge of Japan Before World War I and Afier (London: Routledge, 1992). 18. Cited in Robert Jervis, The Meaning of the Nuclear Revolution (Ithaca, N.Y Cornell University Press, 1989). See also Thomas C. Schelling, Arms and Influence (New Haven, Conn.: Yale University Press, 1966). 19. Geofttey Blainey, The Causes of Wan 3rd ed. (New York: Free Press, 1973), chap. 3. 20. New York Times, November, 22, 1985. ‘These arguments are less compelling, however, for new nuclear states with retaliatory forces that are smaller, less redundant, more vulnerable, and backed by less sophisticated command and con- trol systems. Consequently, we cannot necessar~ ily generalize from the stability of superpower deterrence to stability among newer nuclear _ states in an era of proliferation. 21. The usefillness of the balance of power or power transition theories for understanding the power dynamics of regional systems in the future depends in part on whether the restraining influ- ence of the great powers continues to diminish after the end of the Cold War and collapse of bipolar bloc systems, on whether economic sanc~ tions provide an effective tool of restraint in an increasingly globalized and integrated world cconomy, and on whether the great powers, ‘alone or in conjunction with the United Nations, are able to cooperate to establish effective collec tive security and peacekeeping systems. 22. That is, systemic structure may serve as necessary as well as sufficient conditions for Jack S. Irevy ting involved in an enormously destructive gen- eral war. 25. Richard Rosecrance, The Rise of the Trad- ing State (New York: Basic Books, 1986). 26. Soloman W. Polachek, “Conflict and Trade," fournal of Conflict Resolution 24, 00. 1 (March 1980): 55-78; John R. Oneal, Frances ‘Oneal, Zev Maoz, and Bruce Russett, “The Liberal Peace: Interdependence, Democracy, and International Conflict, 1950-86,” Journal of Peace Research 33, no. 1 (February 1996): 11-29; and Katherine Barbieri, “Economic Interde- pendence: A Path to Peace or Source of Inter- state Conflict?” Journal of Peace Research 33, n0. 1 (February 1996): 29-49. 27. Samuel P. Huntington, “Why Inter- national Primacy Matters,” International Security 17 (Spring 1993): 68-83. 28. Pethaps the classic case of the milita- rization of a commercial rivalry involved Eng- land and the Netherlands in the seventeenth century. For a analysis of its historical evolution and theoretical implications, see Jack S. Levy and Salvatore Ali, “From Commercial Com= petition to Strategic, to War: The Rise Of the Anglo-Duh Rivalry, 1609-1652," in The Dynamics of Enduring Rivalries, ed. Paul F. Dich! (Champaign: University of Illinois Press, forthcoming). For a hypothetical account of the militarization of the current Japanese~ ‘American rivalry, one in which the causal dynamics are strikingly similar to those of the Anglo-Dutch rivalry nearly four centuries ago, see George Friedman and Meridith Lebard, The Coming War with Japan (New York: St. Martin's, 1991). 29. Admittedly, these factors do not fit neatly into a levels-of-analysis framework, which implicitly assumes that states are the primary acters and that there is a clear distinction between the domestic and the international tealms. Both the actors whose behavior we want to explain (ethnic and other forms of communal groups) and the independent variables that influ- ‘ence them cut across state lines. Although this does complicate the applicability of a levels- ‘of-analysis framework, to the extent that these ‘groups are politically organized we can still ask ‘whether the factors that influence group behav- Scanned with CamScannerjor derive primarily from factors external to the group, from leaders’ concems for their political support within the group, or from the beliéfsys- cans or idiosyncracies of the leaders themselves. Politically unorganized groups are both more difficult to analyze and generally less infliential international politics. 30. Imperialist policies help shape the'struc- ture of the international capitalist ecohomy, which creates further opportunities for capita- list states and imposes serious constraints on the more dependent states in the world system. Thus there are important connections between the societal and systemic levels of analysis in some neo-Marxist theories. See Immanuel Wallerstein, The Politics of the World Economy (Cambridge, U.K: Cambridge University Press, 1984). 31. V.L. Lenin, Imperialism (New York: Inter- national Publishers, 1939); Bernard Semmel, €d., Marxism and the Science of War (New York: ‘Oxford University Press, 1981). 32. Waltz, Theory of International Politics, pp- 18-29; Joseph A. Schumpeter, Imperialism and Social Classes, trans. Heinz Norden (New York: Augustus M. Kelley, 1951); Anthony Brewer, Marsist Theories of Imperialism (Lon: don: Roaedge and Kegan Paul, 1980). 33. Jack Snyder, Myths of Empire (Ithaca, N-Y.: Comell University Press, 1991). 34, Cited in Jack S. Levy, “The Diversionary ‘Theory of War," in Handbosk of War Studies, ed. Manus I. Midlarsky (Boston: Unwin Hyman, 1989), p.259. 35. These cases include the decisions of both British and Argentine leaders to go to war over the Falklands/Malvinas in 1982, Reagan's deci- sion to invade Grenada, Saddam Hussein's decision to stand firm in the face of U.S. coercive threats, and Serbian policy in Bosnia. ~ 38. Jack S. Levy, “Domestic Politics and War,” Journal of Interdisciplinary History 18, no. 4 (Spring 1988): 662. %. Criteria for war include a military con- flict jnvolving at least 1,000 battle deaths, and critetia for democracy include regular elections, tolerance of opposition parties, and a parliament that/at least shares powers with the executive. Possible exceptions to this “law” might include the ‘American Civil War and the Spanish- American War. Brice Russett, Grasping the Democratic Peace (Princeton, NJ. Princeton University Press, 1993). Note that it is demo- cratic regime type, not similarity of regimes, that makes a difference, for authoritarian regimes often fight each other. 40. Edward D. Mansfield and Jack Snydes, “Democratization and the Danger of War,” Inter national Security 20, no. 1 (Summer 1995): 41. Andrew J. Enterline, “Driving While Democratizing,” International Security 20, no. 4 (Spring 1996): 183-193, 42. Renee de Nevers, “Democratization and Ethnic Conflict,” in Ethnic Conflict and Inter~ national Security, ed. Michael E. Brown (Princeton, NJ.: Princeton University Press, 1993), pp. 61-78. 43. An ethnic community is “a named hu- man population with a myth of common ances- try, shared memories, and cultural elements; a link with a historic territory or homeland; and a measure of solidarity.” Donald Horowitz, Ethnic Groups in Conflict (Berkeley: University of Cali- fornia Press, 1985), pp. 55-92. Nationalism involves the devotion of the primary loyalties of group members to the ethnic or national com- munity and the desire for their own indepen- dent state. Stephen Van Evera, “Hypotheses on Nationalism and Was,” International Security 18, no. 4) (Spring 1994): 6; Ted Robert Gurr, Minorities at Risk (Washington, D.C.: United. States Institute of Peace Press, 1993); and Manus I. Midlarsky, ed., The Internationalization. of Communal Strife (London: Routledge, 1992). _ 44. Samuel P. Huntington, “The Clash of ,” Foreign Affairs 72, no.3 (Sommer 1993): 22-49, Huntington argues that the clash tions will replace the clash of princes Scanned with CamScannerthat began with the Treaty of Westphalia in 1648, the clash of nations that began with the French Revolution 1789-1792, and the clash of ideologies that began after World War I and the Russian Revolution. This provocative argument underestimates the importance of nationalism ” and other sources of division within broadly defined Islamic, Confucian, and western civiliza~ tions, but it highlights the importance of previ- ‘ously neglected variables. For a critique see David Little's essay on religious militancy (chapter 5 in this volume) and The Clash of Civilizations? The Debate: A Foreign Affairs Reader (New York: Foreign Affairs, 1993). 45. Serbs and Croats fought each other very little before this century, for example. 46. Michael E. Brown, “Causes and Impli- cations of Ethnic Conflict,” in Evbnic Conflict and International Security, cd. Michael E. Brown (Princeton, NJJ.: Princeton University Press, 1993), p. the Crisis of the Post-Soviet State,” Survival 35, no. 1 (Spring 1993): 5-6. 47. Barry R. Posen, “The Security Dilemma and Ethnic Conflict,” Surefval35, no. 1.(Spring. 1993): 27-47. 48. Snyder, “Nationalism,” pp. 7-11. 49. This is especially likely if these groups have been oppressed by national majorities or if other ethnic groups are growing in power. Van Evera, “Hypotheses on Nationalism and War,” pp. 10-22; and Donald Horowitz, “Inedentas ‘and Secessions,” in Irredentism and International Politics, ed. Naomi Chazan (Boulder, Colo: ‘Lynne Rienner, 1991), pp.9-22. 50. Barry R. Posen, “Nationalism, the Mass Army, and Military Power,” International Security 18, no. 2 (Fall 1993): 80-124. 51. In the case of the disintegration of Yago~ slavia and the war in Bosnia, for-cxample, it is nation ily national Security 19, n0. 3 ( 331-367. 52, Van Evera, “Hypotheses on Nationalism and War,” pp. 23-30. 53. Hans Morgenthau, Politics Among Na~ tions, Sth ed. (New York: Knopf, 1973), p. 253. ‘On the persistence of images see Janice Gross Stein's essay in this volume (chapter 6). 54. F. Stephen Larrabee, “Down and Out in Warsaw and Budapest: Eastern Europe and East-West Migration,” International Security 16, ‘no. 4 (Spring 1992): 5-33. Note that the framing of environmental issues in terms of “high poli- tics” and national security may also serve a polit- ical strategy. The symbolism of national security increases the prospects that these issues may be put on the policy agenda and that research on these topics might gain funding from the gov- emment or foundations. See Mare A. Levy, “Is the Environment a National Security Issue?” and Jack Snyder, “Nationalism-and—— International Security 20 (Fall 1995): 35-62; and Daniel Deudney, “The Case Against Linking Environmental Degradation and National Security,” Millennium 19, no. 3 (Winter 1990): 469, See also Astri Subrke's essay in this volume (chapter 7). 55. Myron Weiner, “Security, Stability, and International Migration,” International Security 17, no.3 (Winter 1992-93): 91-126. 56. As Subrke argues in chapter 7 of this vol- ume, however, international migration does not always lead to social conflict. Under some condi- tions, migrants are assimilated into the host ‘country, particularly when they provide needed labor and skills and when population movement takes the form of gradual migration (often in response to gradual changes in demography and economic incentives), rather than sudden dis- placement arising from ethnic conflict or envi- ronmental disaster. 57. For example, the construction of the Berlin Wall in an attempt to stop the flood of refugees from East Germany led to the Soviet- American Berlin Crisis in 1961, 58. Paul Kennedy, Preparing for the Twwenty- First Century (New York: Random House, 1993); and Robert Kaplan, “The Coming Anarchy,” Aulantie Monthly, February 1994, pp. 44-76. Scanned with CamScannerOne of the major research projects on these questions concludes that the environmental con cems most likely to lead to international conflict include the depletion and degradation of forests, good agricultural land, fresh ater, and fish stocks, whereas atmospheric changes such as greenhouse-induced global warming and strato- spheric ozone depletion will nat have a majot effect for several decades and then only in con- junction with these other scarcities. Thomas F. Homer-Dixon, “On the Threshold: Environ- mental Changes as Causes of Acute Conflict” and “Environmental Scarcities and Violent Con- Act: Evidence from Cases,” in Global Dangers, ed. Sean M. Lynn-Jones and Steven E. Miller (Cambridge, Mass: MIT Press, 1995), pp. 43-83, 145-147, Another path through which environmental degradation might contribute to international conflict involves pollution across state bound- aries, the possibility of coercive state responses to reduce that pollution, and a conflict spiral that escalates to war. But the magnitude of such con- flicts is likely to be small in relation to those over economic or identity issues. See Deudney,“Envi- ronmental Degradation and National Security,” ; and Suhrke’s essay in this volume (chap- 59. Note that the question of whether envi- ronmental change is a security issue, which has, attracted considerable attention in the literature, is much broader than our more specific question ‘of whether environmental change contributes to international conflict. Environmental change can threaten core values and individual security without involving international conflict. For dis- cussions of environmental change as part of new conecptions of security see Richard ees ity,” International Security 8, Security,” car CONTENDING THEORIES OF INTERNATIONAL CONFLICT 23 ‘that the latter is more peaceful than the former. Karl A. Wittfogel, Oriental Despotism (New Haven, Conn. Yale University Press, 1957);and Manus I. Midlassky, “Environmental Influences on Democracy,” Journal of Conflict Resolution 39, no. 2 (June 1995): 224-262. 61. Homer-Dixon argues that conflicts over nonrenewable resources tend to be more desta- bilizing than disputes over renewable resources and that disputes over river water tend to be par- ticularly serious (“Environmental Scarcities,” pp. 157-159). On the seriousness of conflicts over water, see Peter H. Gleick, “Water and Conflict: Fresh Water Resources and Inter- national Security,” in Global Dangers, ed. Lynn Jones and Miller, pp. 84-117; and Miriam R. Lowi, “Bridging the Divide: Transboundary Resource Disputes and the Case of West Bank Water,” in Global Dangers, pp. 118-143. 62. Deudney, “Environmental Degradation and National Security,” pp. 470-471. 63. Homer-Dixon, “Environmental Searci- tics,” pp. 157-159; and Deudney,“Environmen- tal Degradation and National Security,” p. 471. 64. Graham. Allison, The Eucence of Decision (Boston: Little, Brown, 1971). 65. Richard K. Betts, Soldiers, Statesmen, and Cold War Crises (Cambridge, Mass: Harvard University Press, 1977). Admittedly, there have been few systematic comparative studies of this relationship. 66. Stephen Van Evera, “Primed for Peace: Europe Afier the Cold War,” International Security 15, n0.3 (Winter 1990-91): 18-23. 67.1n addition, the unpredictability of bureaucratic compromises can increase adversary uncertainty regarding one’s intentions, thereby enhancing the misperceptions that contribute ‘to war under certain conditions. For an applica- tion of the organizational process model to the July 1914 crisis, see Jack S. Levy, “Organiza tional Routines and the Causes of War,” Interna~ tional Studies Quarterly 30, n0. 2 (June 1986): 193-222. 68. Stephen Van Evera, “The Cult of the "Offensive and the Origins of the First World War,” International Security 9 (1984): 58-107; Jack S. Levy, “The Offensive/Defensive Scanned with CamScannertT Balance of Military Technology,” Iiternational Studies Quarterly 28, no.2 (June 1984): 219-238, 69. Van Evera, “Primed for Peace,” pp. 12-16, 70. On images of the cnemy, see Ole R, Holst, “Cognitive Dynamics and Images of the Enemy,” in Image and Reality in World Politics, ed. John Farrell and Asa Smith (New York. Columbia University Press, 1967), pp. 16-39: and Richard Ned Lebow, Between Peave and War Baltimore, Md: Johns Hopkins University Press, 1981). 72. On experiential learning se Jack S, Levy, “Learning and Foreign Policy: Sweeping a Com, ceptual Minefield,” International Organization 48, no.2 (Spring 1994): 279-312, see also Janice Gross Stein's essay in this volume (chapter 6). 73D: ic and Amos Tversky, Judgment Under Uncertainty: Heuristion and Biases (Cambridge, UK: Cambridge University Press, 1982); Richard Nisbett and Lee Ross, Human Inference: Strategies and Short- comings of Social Judgment (Englewood Cliffs, NJ.: Prentice-Hall, 1980); and Jervis, Perception and Misperception. 74. James D. Fearon, “Rationalist Explana- tions for War,” International Organization 49, no. 3: 379-414, — JACK S. Levy no. 1 (October 1983): 76-99. See also Robert Jervis, “War and Misperception,” Journal f Interdisciplinary History 18, no.4 (Spring 1988): 675-700. Note that each of these forms of mis- Perceptions can also contribute to peace, although through different causal paths, Bxaggeration of adversary capabilities, for example, may lead a state t0 reject a decision for a preventive or pre~ emptive war that would have been rational with complete information. Note also that mispercep- fons are not necessarily nonrational. Given the extent of complexity and uncertainty int the world, rational information processing some- times leads to incorrect predictions. 76. Liberal economic theories, for example, incorporate the stricture of the global economy, internal economic sources of economic expan- sion and contraction, domestic pressure ‘groups, and economic ideologies. Snyder's coalition model incorporates stages of economic develop ment, coalitions among parochial interests at —____ both the domestic and governmental level, the impact of systemic outcomes on the balance of Power between competing domestic elites, and the psychological impact of self-serving strategic myths. Hypotheses on environmental scarcity” and degradation recognize that the most signifi- cant impact of systemically induced ‘environ- ‘mental change on international conilict ray be through intervening societal-level variables. 77. On the declining utility of military power since the end of the Cold War, see John Lewis Scanned with CamScanner