The Korean Journal of Defense Analysis Vol. 29, No. 3, September 2017, 429－454

Picking a Fight: Democracies, Liberal Constraints, and Selective Conflict Initiation∗1 Johann Park∗∗2 Incheon National University, Incheon, Republic of Korea

Research shows that autocracies as well as democracies have reasons to avoid strong enemies. What, then, make democracies distinctive from autocracies in conflict selection? A critical difference may lie in the normative constraints democracies experience. Focusing on the roles of public sentiment and liberalism, we argue that democrats bearing high levels of accountability are less likely to choose armed conflict without justifiable causes. We assess the roles of three liberal factors in regard to the characteristics of target countries: respect of human rights, democratic representation, and economic interdependence. Material factors, such as relative military capability and geographic constraints are also considered. The results show that both autocracies and democracies tend to attack easier foes, but democracies avoid attacking countries that respect human rights and are economically interdependent. Additionally, unlike autocracies, democracies do not view other democracies as attractive targets of military attacks. Keywords: democracy, norms, conflict selection, conflict initiation, constraints

* ‌This study has been significantly modified and developed from the last substantive chapter of the author’s Ph.D. dissertation, “Exploring the Second Wave of Democratic Peace Critiques: Commonality, Territoriality, and Selectivity” (Michigan State University, May 2011). Various earlier versions have been presented at the 2013 Annual Meeting of International Studies Association, the 2015 Annual Meeting of the Midwest Political Science Association, the 2015 Regional Conference of the International Studies Association–South, Mississippi State University (May 2011) and Korea University (May 2015). I would like to thank Valentina Bali, Cristina Bodea, Michael Colaresi, Petra Hendrickson, Patrick James, Bueno de Mesquita, and Allan Stam for their invaluable comments. ** E-mail: [email protected] ISSN 1016-3271 print, ISSN 1941-4641 online © 2017 Korea Institute for Defense Analyses http://www.kida.re.kr/kjda

430

Johann Park

Introduction This paper examines how and why democracies select armed conflict distinctively from autocracies. Despite the enormous scholarly efforts made to confirm and explain the inter-democratic peacefulness and even its law-like status,1 it is also found that armed conflicts between two autocracies are infrequent.2 Additionally, democracies are not necessarily more likely to shy away from taking the first step toward dispute militarization than autocracies.3 What, then, really makes democracies and autocracies different in conflict behavior? Who have democracies fought with and why? By examining selective conflict initiations by democracies, this study can answer these important but neglected puzzles. We argue a critical difference lies in how democracies selectively initiate interstate conflict because of a unique kind of constraint they experience that is less material, but more normative. We identify three liberal factors in regard to potential targets of military attacks: respect for human rights, political representation, and commercial hospitality. We test how these liberal factors affect democracies differently than autocracies in the decision to initiate foreign attacks. This study also has critical implications for other important but underspecified phenomena like the conflict proneness of mixed dyads (i.e., democracy vs. autocracy), roughly dubbed the “cats-and-dogs” relationship, and the various foreign policy successes associated with democracies, dubbed “democratic victory.” Although underspecified, selective conflict initiation by democracies per se is hardly a new idea in the literature. The well-known “selection effects” argument links deliberate choice to war victory: democracies tend to win wars because they pick easy fights. In effect, democratic selectivity is an agreed–upon explanation for democratic victory among the institutional variants of the democratic peace theory: structural,4 institutional,5 and informational.6 Democratic selectivity, however, is short on direct evidence. Existing research has been too outcome–oriented, having accrued evidence by only looking at the outcome stage of war and conflict. As realist critic Desch points out, evidence must come from the initiation stage in that the selection effects argument presupposes deliberation throughout the decision–making process from peacetime to the initial stage of militarization.7 Indeed, little has been done to examine whether and how democracies selectively initiate interstate conflict. Additionally, extant research has been silent on specific metrics democracies use in their conflict choice. At best, only a scant discussion has occurred about relative material capability, which is believed to most directly affect war outcomes. Our present study fills this gap in the literature in two important ways. First, we offer a direct test for the selection effects argument by looking at the initiation stage of militarized disputes from peacetime without selecting the dependent variable. Second, we identify and measure specific material selection criteria that have been loosely defined in the literature. However, our greater contribution is that we reframe democratic selectivity in terms of liberal constraints beyond material costs. Of course, democracies should have reasons to choose easy fights because undesirable foreign policy outcomes can result in harsh public backlashes against decision-makers. Victory matters, but so too

Democracies and Conflict Initiation

431

does relative power estimation. Autocracies, however, seem to have no less necessity to avoid strong military powers as war outcomes more severely hurt the fate of their political leaders.8 Careful estimation of relative power alone does not seem to account well for democracies’ conflict selection distinctive from that of autocracies. Leaders in democracies, commonly assumed to bear greater public constraints and political accountability, need to make sure, or at least make it appear, that their policies conform to the public’s general preferences. Of foreign policy issues, security issues such as military conflict and war are found to be the most salient.9 When a democracy launches a military attack, its aims should make sense to the public. Gelpi, et al. show that the public assessment of war not only depends on the prospect of victory but also on the rightness of war aims.10 Roughly put, the United States, the world’s most capable and representative democracy, is not likely to invade much weaker neighbors like Canada and Mexico because the U.S. public will hardly perceive it as being right. In fact, the United States has mobilized significant human and material resources for various difficult and expensive wars against illiberal regimes in two World Wars, the Korean War, the Vietnam War, and the recent wars in Iraq and Afghanistan. What, then, makes democracies different from autocracies in conflict selection? It appears that democracies often claim some sort of liberal cause when they launch military forces. Truthful or not, these liberal causes seem to be used to justify motivations for violent military actions. A remark, like the one President Obama made soon after he authorized military action against Libya under Muammar Qaddafi (March 28, 2011), seems to be typical when democracies launch a military attack: In the face of the world’s condemnation, Qaddafi chose to escalate his attacks, launching a military campaign against the Libyan people…. Confronted by this brutal repression and a looming humanitarian crisis, I ordered warships into the Mediterranean…. In this effort, the United States has not acted alone…. joined by…. nations like the United Kingdom, France, Canada, Denmark, Norway, Italy, Spain, Greece, and Turkey... It’s true that America cannot use our military whenever repression occurs. …given the costs and risks of intervention, we must always measure our interests against the need for action. But that cannot be an argument for never acting on behalf of what’s right.

This statement suggests that justice matters to democracies’ decision to use force against other countries. Does this type of normative claim carry any truthfulness or is it merely cheap talk without any causal significance? One could argue that democracies make normative claims for mere justification of their military actions. Even if true, however, this skepticism does not nullify the relationship between norms and conflict initiation. The fact—countries want to justify their military attacks—per se attests to the importance of normative causes as “it is literally an attempt to connect one’s actions to standards of justice.”11 Democracies seem to avoid conflicts that are not only hard to win but also hard to justify in terms of liberal causes. Certain kinds of countries seem to constitute harder military targets for democracies. We explore a set of liberal factors regarding target characteristics that distinctively constrain democracies from launching

432

Johann Park

military attacks against certain international targets. Therefore, we theorize democratic selectivity in terms of liberal constraints, theorizing the roles of public sentiment and liberalism that uniquely form democracies’ selection criteria for conflict initiation. We specify a causal mechanism about how political accountability and public consent leads to selective conflict initiations that are expected to generate not just low material but also low normative costs. Armed conflicts generate not just material costs, but also moral costs, and the people, rather than the leaders, tend to be more sensitive to these costs.12 Highly constrained by public sentiment and assessment, leaders in democracies want to minimize associated costs when they decide to launch military force. Liberal norms shape normative assessments of war in democratic societies. Thus, democracies are more likely to initiate armed force against countries that can be portrayed as unjust and illegitimate by the principles of liberal democracy. We assess the roles of three liberal factors in regard to the characteristics of target countries—(1) respect of human rights, (2) democratic representation, and (3) economic interdependence—as they appear to be the most frequently claimed liberal causes by democracies. Also, we employ specific material selection factors, such as relative capability, geographic proximities, and topographical features that make military attacks difficult. We test the statistical interactions of challengers’ regime types and targets’ (il)liberal characteristics in affecting conflict initiation. This allows us to test not only whether, but also how, democracies are selective at the initiation stage of conflict, avoiding the inferential complications caused by the selection bias that can occur when one looks at the outcome stage to get the evidence that presupposes initial selectivity.13

Democratic Selectivity and Democratic Victory All three main institutional explanations in the democratic peace literature agree that careful selection is vital to the victorious war records of democracies noted below. For example, the structural model argues that democracies are constrained from engaging in risky, costly, and unpopular military adventures.14 Regular and competitive elections empower citizens and make leaders susceptible to public appraisal. Democracies initiate quick, low-cost, and winnable wars to avoid domestic backlash. The selectorate theory suggests that democrats with large winning coalitions are more sensitive to the successes and failures of foreign policy than autocrats.15 Once engaged in wars, democracies fight harder for victory, but they try harder to evade strong enemies before fighting than autocracies. The information perspective stresses that democrats have inherent difficulties in concealing and misrepresenting preferences, resolve, and strengths against foreign opponents due to time–consistency pressure,16 free media,17 and effective opposition parties.18 Constrained from bluffing, democracies should be more selective about making threats, but this selectivity increases the effectiveness of the threats they make.19 In sum, democratic accountability creates a strong impetus for leaders to launch armed forces only when a quick and easy victory is expected. Various findings support the selection effects argument. As compared to autocracies,

Democracies and Conflict Initiation

433

democracies tend to win wars, fight shorter wars, and suffer fewer battle and civilian casualties, and these positive effects are even stronger when democracies are initiators.20 When democracies initiate militarized threats, they meet fewer violent responses from target countries.21 Even the inconvenient history of democratic colonial imperialism is proposed as evidence of the selection effects argument: “democracies engage in colonial wars and imperial wars against much weaker adversaries.”22 This ample evidence, however, has not muted criticism. Critics suggest that proponents have failed to consider the constraints imposed upon many autocrats to engage in risky wars,23 incorporate the possibility that democratic politics do not prevent democracies from fighting unnecessary, expensive, and long wars,24 and provide direct evidence with proper research designs.25 These caveats pertain to previous research’s outcome–oriented approach, focused on the victorious war records of democracies. Theoretical efforts have been made ex post to catch up with the interesting phenomena related to democracy, considered the most desirable, or least vicious, form of political systems. While understandable, the fascination and resulting outcome–oriented approach leave two critical problems for confirming democratic selectivity: (1) the lack of direct evidence and (2) underspecified definition. First, whereas choosing weak foes has to do with conflict initiation, much evidence has been accrued in terms of desirable outcomes from the samples of dyads already involved in wars. Looking at war samples, in this simple example, is likely to underspecify democratic selectivity. Critics like Desch ring true: if “democracy makes states better in making winnable wars before they start,” evidence should come from “the peacetime decision-making process in the first stage, not from the second, wartime stage.”26 Indeed, proponents have made analytical leaps in linking outcome–stage results to initiation–stage decision-making or vice versa. Desch, however, does not follow up his good idea in testing the selection effects argument. By focusing on the war stage, not on the peacetime decision-making process, his process–tracing study is unable to look at “the dogs that did not bark.” Imagine a world where democracies select fights based on low expected costs. Many states challenged by democracies may have backed down short of war. If Desch finds only “ten clear cases” of war initiations by democracies since 1815, such rarity may suggest that democracies have been deliberate in initiating armed forces. 27 Looking at the initiation of lower level armed disputes such as militarized interstate disputes (MIDs) can provide a direct test for democratic selectivity. By definition, while all wars are a subset of MIDs, a war emerges from a pre-stage MID in most cases. Second, previous research underspecifies the meaning of democratic selectivity. It overuses ambiguous adjectives like “winnable,” “weak,” “easy,” or “tough” to describe the kinds of military targets that democracies prefer, instead of specifying selection criteria for conflict initiation. An exception exists. Anderson and Souva define selectivity as the tendency to target enemies with inferior military capability by interacting political accountability with relative capability to explain conflict initiation using all directed dyad years of 1970–2000.28 It finds that as winning coalition size (W), measured by Bueno de Mesquita et al.,29 increases, the effect of military advantage, measured by the

434

Johann Park

Correlates of War Project’s Composite Index of National Capability (CINC), increases. Anderson and Souva offer a direct test improving upon previous research. However, they never use a democracy measure while the other two institutional explanations strongly argue for democratic institutions as the main source of accountability and selectivity. Although W is intended to measure accountability, Clark and Stone question its validity vis-à-vis the commonly employed Polity index.30 Furthermore, democratic selectivity is not just confined to confirming military superiority before initiating armed forces. According to the accountability–based explanations, democrats are constrained from making foreign policies against the public’s generic preferences. Although important, a high expectation of victory is rarely enough for public consent. Before its initiation and ultimate victory, war aims must make sense to the public, which Gelpi et al. show depends on the rightness of war as well as the benefits the victory would bring about.31 As we argue below, democrats face not just material constraints but also normative constraints regarding conflict initiation, and they make normative assessments and justification based upon liberal norms. This is the very aspect of democratic selectivity that extant studies have failed to capture. In summary, both proponents and critics have relied upon war outcomes with a subsample of actual war participants. This outcome–oriented approach has led to indirect statistical tests of democratic selectivity, unreliable selection of cases, complications of measurements regarding war participants and outcomes, and theoretical inattention to factors underpinning a selective conflict decision.

Reframing Democratic Selectivity: Conflict Initiation and Normative Constraints Leaders may launch armed force for private interests as well as national interests. In either case, they want victory to fulfill war aims. This does not differentiate democrats from autocrats. In fact, many autocrats have lost power due to war loss and faced more severe consequences.32 What makes them different is the justification mechanism unique to democrats. Autocrats can use repression, but democrats must persuade people for consent. Although made in a different context, Abraham Lincoln’s statement rings true: “[P]ublic sentiment is everything, with public sentiment, nothing can fail, without it nothing can succeed.” Although not reflected in their analysis, Reiter and Stam make a similar point regarding war initiation: “voters…punish leaders for failing to heed popular sentiments at the time of [policy making] rather than for particular failures or successes.”33 Leader accountability Democracy Public sentiment

Figure 1. Selective Conflict Initiation by Democracies

(Normative) Constraints

Selective Conflict by Iiberal standards

Democracies and Conflict Initiation

435

Regarding conflict initiation, we theorize democratic conflict selectivity focusing on liberal constraints. We argue that people empowered in democracies care about the rightness of military conflict, that their leaders bearing high levels of accountability avoid using force against hard-to-justify targets, and that both leaders’ and people’s normative assessments are anchored in liberal principles. Hence, foreign countries with illiberal domestic and international practices will constitute more acceptable targets of military attack for democracies. We proceed with the common assumption that democracy empowers people and holds leaders accountable. Unlike autocrats, democrats must earn a majority or plurality of the public votes for power while interplaying with prominent political rivals, actively monitoring free media, vociferous independent experts, directly–affected stakeholders, and enthusiastic foreign policy zealots. These all work together to put greater constraints on democrats than autocrats from making policy decisions against the general preferences of the people. We posit that concurrent public sentiment at the time of decision-making as well as ex post public evaluation in elections affect democracies’ foreign policy choices. As Reiter and Stam usefully point out, “voters and leaders alike tend to focus on the matter at hand.”34 The evidence shows that the general public in democracies consistently follows important foreign policy issues and takes them as a useful cue in assessing leaders’ competence.35 Of foreign policy issues, security issues have been regarded as the most salient.36 In time of crises, leaders are eager for public support and mobilization to fight effectively against foreign threats while people are willing to project charisma and capabilities onto their leaders even at the expense of civil liberties.37 Yet, these studies show that as crises linger, people withdraw the support and trust they have extended toward leaders.38 There is another reason that leaders want contemporaneous approval for foreign policy choices. Aggressive foreign policy is subject to the suspicion that leaders exploit external threats for private ends to stabilize political power, divert public attention from domestic difficulties, and profit the military–industry complex. Recent studies show that democrats have strategic reasons to avoid initiating military conflict when the timing and context are suspicious and thus public backlashes and electoral repercussions are likely.39 In addition, having concurrent consent can be an insurance against unsuccessful policy outcomes and unfavorable ex post public evaluation. In evaluating unsuccessful outcomes, voters may consider the necessity, rightness, and difficulties of the unsuccessful policy that they approved. By contrast, however successful their policies are, if leaders fail to heed popular sentiments, they will lose people’s hearts and minds, putting their political fate in danger. Democratic institutions provide ample sources of concurrent consent, such as effective opposition parties,40 free media,41 legislative constraints,42 active political participation and civic cultures.43 Leaders should make sure that their proposed policies are congruent with general public sentiment. Where does public sentiment lean toward regarding conflict initiation? Initiating conflicts that are costly and difficult should be unpopular. Yet, this is only a partial story of democratic selectivity. In fact, the world’s most capable democracy, the United

436

Johann Park

States, has mobilized significant human and material resources for various difficult and expensive wars such as two World Wars, the Korean War, the Vietnam War, and the recent wars in Iraq and Afghanistan. We argue that voters consider the legitimacy of military conflict, not just its unconditional probability of victory. Voters are both material and normative beings. Using force abroad incurs not just material but also moral costs that can be enormous when it erupts into a greater scale of hostilities. The cross-national evidence suggests that people are generally more sensitive to these costs than political leaders and military elites.44 Hence, leaders in democracies have strong incentives to minimize the moral costs associated with foreign military attacks. When initiating armed forces, democrats need to convince their constituents of its legitimacy with normative causes. Doing so can additionally help justify the incurring material costs.45 Yet, without justifiable causes, democrats would not attack even much weaker countries. For instance, despite the overwhelming chance of victory, France is not likely to invade its much weaker neighbors like Luxemburg and Belgium because it is hardly seen as right by French citizens. Though directly related to the public–consent models, normative constraints are the very aspect of selective conflict–making that previous research is mostly silent on.46 We think that both leaders’ and people’s normative assessments for conflict initiation are anchored in the principles of liberal democracy. The competing International Relations paradigms agree that foreign policy success is more likely when “power [is] exercised in accordance with well-defined social conventions.”47 Liberal democracy is the ideology that defines how politics are conducted and assessed in democracies. Liberal ideology recognizes people’s rights, freedom, and interests, and it affects how they understand and respond to proposed policies and outcomes. How politicians compete for power and how they govern are also socialized by liberal norms.48 To persuade people for consent and support, democrats “must live up to the expectations of their own ideology,” liberal democracy.49 According to liberal ideology, universal values such as peace and justice seem inseparable and both can be achieved by promoting liberal values. As President Eisenhower put it, “peace and justice are two sides of the same coin.” Justice is believed to prevail in liberal democracies characterized by political rights, economic freedom, separation of powers, checks and balances, accountability, and representation. For liberal democracies, it is hardly justifiable to attack foreign countries whose policies and conducts are in line with the principles of liberal democracies. Violent conflict may not be desirable or virtuous on its own, but if it could be justified, it could only be done against countries with illiberal government practices. In sum, when democracies initiate conflict, other things being equal, democracies will choose enemies against which using force is more justifiable vis-à-vis liberal standards, and countries with illiberal domestic and international practices will constitute more susceptible targets. One might argue that democracies use force against weak countries and then make up some rhetorical justifications to conceal their true motivation. But this possibility does not nullify the effects of normative constraints unique to democracies. Even realist thinkers like E. H. Carr acknowledge the roles of morality and justification: although not universally, but rather relatively, defined, they are necessary for “the consent of the

Democracies and Conflict Initiation

437

governed” as political “order cannot be based on power alone.50 The fact that leaders often attempt to justify their policy decisions in light of shared values and expectations held by relevant audiences attest to the importance of norms and justifications. Our contribution is to specify the normative constraints that circumscribe possible military targets for democracies in a certain boundary. Democracies will choose foreign governments that are easy to denounce as well as defeat vis-à-vis liberal principles. Our reframed democratic selectivity can address an empirical inconsistency against the selection effects argument. Studies find that autocrats, rather than democrats, tend to lose their office and suffer harsh punishment for unsuccessful international conflicts.51 Democrats, bearing greater public constraints, initiate conflict only when they can earn public consent or justify its aims. Sometimes, democrats cannot help going to war with inevitable or justifiable causes despite high expected costs. Sometimes, the public pushes leaders toward military adventurism or sometimes, leaders find it easy to justify their hawkish agenda. When war is inevitable, or once war aims are justified, unfavorable outcomes will be less critical to the political fate of democratic leaders. Therefore, our theory suggests that democrats may have opted for some risky and difficult conflicts because those were unavoidable or at least justifiable.

Hypotheses Although extant research on democratic selectivity has ignored the roles of liberal norms, they are prime forces for foreign policy deliberation in our reframed theorization. It suggests that military attacks against foreign governments that are seen as right and just by liberal standards will be exceptionally unpopular for democratic societies. Therefore, foreign governments with illiberal practices constitute more susceptible targets for potential democratic aggressors. Informed by the public opinion research cited below, we focus on the three characteristics of target countries in relation to liberal norms such as respect for human rights, political representation, and commercial hospitality. Two precautions must be made before presenting the hypotheses and rationales. First, though in line with liberal principles, we do not consider wars of existential defense as uniquely justifiable to democracies. When the very existence of a state is in danger, autocrats should be no less motivated to wage war for self-defense. Since democratic uniqueness, not a litany of all justifications, is our main interest, we do not include the right to self-defense as one of the normative factors. Second, with all of our discussions and examples that follow, we do not suggest that military operations by democracies are always based on good causes or follow national interests. True, underlying motivations of leaders are unobservable, and they can starkly differ from their publicized objectives. The bottom line is that as compared to autocrats, democrats have more difficulty engaging in military actions without plausible public explanations. For democrats, there must be certain types of states against which military attack is more or less legitimate than others.

438

Johann Park

Respect for Human Rights Democracies should be reluctant to attack domestically well-behaved foreign governments because such an action is hardly endorsable to their citizens. Rather, they will try to pick foreign governments that treat their people badly, repressing and denying their basic rights. The inhumane practices of foreign countries do not necessarily free democratic challengers from their normative constraints. The normative costs and justifications, however, should be lower and less difficult when attacking such governments than when attacking those with good human rights records. Violent oppression of its people by a government is directly counter to the principles of liberal democracy. These principles are grounded on the fundamental postulation that all human beings deserve to pursue lives of dignity and integrity free from arbitrary arrest, detention, and torture, regardless of one’s religious, political, ethnic, and socioeconomic origins. Democratic citizens socialized in this normative context should have a greater abhorrence of human rights abuse abroad than their autocratic counterparts. Citizens in autocracies, to some degree, would abhor anti-humanitarian acts and consider abusive foreign countries as cruel and unjust. However, they lack influence over rulers and policies as autocratic rulers isolate them from the prevailing political processes. Therefore, public sentiment on human rights issues are not likely to translate into actual foreign policy in autocracies. In sum, leaders in democracies will find that targeting countries fraught with human rights abuses incurs relatively few normative costs, lessening the level of difficulty in persuading their citizens about its rightness. Conversely, it will be extraordinarily difficult to justify assaults against foreign governments with respectable human rights records. Therefore, we hypothesize: H1: Unlike its autocratic counterpart, a democratic potential initiator is increasingly less likely to attack a potential foreign target as the latter’s respect for human rights increases.

Target Democracy As the democratic peace literature has confirmed, the regime type of the targets is of importance to democratic challengers. Both traditional normative and structural explanations suggest that democracies are unwilling to attack one another. 52 Democracies, though existing in various forms (e.g., parliamentary or presidential, and proportional or majoritarian), share ideological values that emphasize individual rights, freedom, fairness, and people’s leverage over rulers. Therefore, in democratic societies, with other things being equal, a foreign country is seen to be politically correct when it is a democracy, more so than when it is an autocracy. Attacking a fellow democracy, therefore, would be hardly justifiable. As Owen states, democratic citizens tend to believe foreign democracies are “reasonable, predictable, and trustworthy, because they are governed by their citizen’s true interests….”53 According to Doyle, regimes that rest on the public consent presume each other to be “just and therefore deserving of

Democracies and Conflict Initiation

439

accommodation.”54 Furthermore, a formal model by McGillivray and Smith finds that due to the difference in the extent of accountability imposed upon leaders, democracies are trustworthy and cooperative while autocracies tend to violate international norms.55 In sum, the consensual and representing natures of democracy are the key references to democratic citizens’ perceptions of justice about foreign governments. H2: Unlike its autocratic counterpart, a democratic challenger is increasingly less likely to attack a potential target as the latter’s level of democracy increases.

Commercial Hospitality A long-held liberal view states that trade promotes international cooperation. Although some question this proposition,56 evidence has been accrued on it.57 We think that the use of force against trade partners is counter to public sentiment in democracies for two reasons. The first reason is that military conflict reduces trade ex ante when expected58 and ex post when unexpected,59 which, in turn, may be detrimental to the political survival of democratic leaders. In mainstream economics theory, trade is thought to promote economic growth.60 Also, trade is presumed to increase the overall welfare of a society.61 Hostile actions ruin economic benefits that many domestic constituents may have enjoyed from good economic relations with the targeted state. Thus, maintaining good relations with important trading partners is of greater importance for democratic leaders than autocratic leaders because the former’s political survival is a function of providing broader policy successes.62 The second reason is more normative and relates to transnational ties and commercial hospitality between trading partners. These all facilitate mutual understanding and respect between two societies.63 The increased familiarity and understanding help distinct societal preferences converge on a broader cosmopolitan identity and friendship.64 Attacking a country that the challenger’s citizens consider familiar and friendly will be hardly acceptable. Citizens should be unwilling to endorse bellicosity toward commercially–friendly countries. The findings in Kleinberg and Fordham based on a forty-seven-country survey suggest that individuals with positive views on international trade and investment hold positive views on trading partners.65 Certainly, not all citizens in a democracy support economic liberalization, but we believe the general preference level is higher for those in democracies than autocracies. Also, attitudes on trade are likely to be endogenous to the level of interdependence (and vice versa) as suggested above. Therefore, democratic leaders, whose political lives depend on the median voters’ approval, and who have reasons to heed the general people’s hearts and minds, are highly motivated to maintain peaceful relations with the nation’s important economic friends. By contrast, economic interdependence may not be a pacifying force for autocratic challengers. Although some nondemocratic countries might respond by avoiding conflict initiation against trading partners, they, in general, bear lower political accountability

440

Johann Park

and fewer public constraints, the workhorses for interdependence to promoting interstate peace.66 Barbieri argues that economic interdependence heightens interstate conflict.67 This is in line with the realist argument that due to international anarchy, the reliance on foreign economies increases vulnerabilities that can be exploited by enemies in times of crisis and war.68 As nations become increasingly dependent on foreign markets, they become desperate to extend the scope of their control to the source of vulnerability. For realists, “interdependence...lead[s] to greater [security] competition, not to cooperation.”69 This vulnerability scenario may be confined to nondemocracies that may have many reasons to inflate foreign threats, offsetting the overall benefits of integration.70 H3: Unlike its autocratic counterparts, a democratic challenger is increasingly less likely to attack a potential target as their level of economic interdependence increases.

How plausible are our hypotheses that emphasize the roles of the three liberal factors? The findings accrued in the public opinion literature corroborate our argument that the public is concerned with not only the costs of military mission but also its aims and motivations, and thus decision-makers need to provide agreeable or legitimate causes.71 For instance, Russett and Nincic found that support for military assistance to other nations varies by geographic distance and the level of economic interdependence.72 Providing aid to neighbors and commercial friends is seen as the right thing to do by the U.S. public, not taking into account security and economic benefits. Jentleson and Britton revealed that humanitarian intervention (providing emergency relief to “people suffering from famine or other gross and widespread humanitarian disasters”) and foreign policy restraint (coercing enemies taking aggressive actions against the United States or its interests) tends to generate high levels of support.73 Analyzing the most extended data on all U.S. public opinion polls from 1981 to 2005, Eichenberg reproduces this finding and further shows that initial baseline support for any military intervention depends on its principal objectives.74 Humanitarian intervention and foreign policy restraint are found to be among the most popular. These military actions appear to be legitimate to the U.S. public because using force as a means to save others’ lives aside from defending its own nation against encroachments by other states is in line with the principles of liberal democracy and international law. Feaver and Gelpi provide complementary evidence that the general public, unlike those in or related to the military, is willing to support military intervention even with high human costs when its mission concerns human rights abuses or democratic stabilization.75 Regarding military casualties, Gelpi et al. (2009) find that expectations of military mission success and beliefs about the rightness of entering the conflict are the two most prominent factors that mitigate the U.S. public’s casualty sensitivity, with the former carrying more weight.76 Yet, in their analysis of the 2004 presidential election, perceptions of the rightness of the Iraq War rather than assessments of its success were found to have a stronger effect on the voting decisions.

Democracies and Conflict Initiation

441

Research Design We examine all the directed dyad years from 1981–2000 in the international system for which all the measures are jointly available. The directed dyad analysis differentiates two states in a dyad into a challenger (side A) and a target (side B). We made the basic template of our dataset from EUGene.77 Our dependent variable is initiation, coded 1 when the country on the challenger side initiated a militarized dispute against the country on the target side, drawing upon the Militarized Interstate Disputes (MIDs) data set (version 3.1). We exclude ongoing dispute–dyad years unless a new initiation occurred in a given dyad year. To account for the possible endogeneity between our dependent and independent variables, we employ a one-year lag structure. Our primary independent variables are Human Rights, Target Democracy, and Interdependence. We measure the human rights practices of target countries, Human Rights, using the CIRI project’s physical integrity rights index referring to “freedom from extrajudicial killing, disappearance, torture, and political imprisonment.”78 This index ranges from 0 (no respect) to 8 (complete respect).79 Target Democracy is measured from the Polity IV dataset.80 Polity2 score is used to facilitate the time series analyses of this study. This score ranges from 10 (a fully democratic dyad) to –10 (a fully autocratic dyad). We add 10 to this score to facilitate the interpretation, resulting in an index of a 0–20 scale. Following the weak link assumption, we measure Interdependence as the minimum of trade dependence levels [(export + import)/GDP] of two states in a given dyad year.81 We use Gleditsch’s expanded trade and GDP dataset 4.1.82 We also provide a test for the importance of material constraints for democratic selectivity by identifying specific factors such as relative capability, mountainous terrain, and tropical topography. Relative Capability is a natural logarithm of the ratio of the target’s capability to the challenger’s capability based on the COW CINC index (v.3.02). According to the selection effects argument, this variable is expected to have a larger negative impact on conflict initiation for democratic challengers than autocratic challengers. Countries’ rough or inaccessible topographic features like mountainous terrain and tropical swamps or jungles make them difficult to attack. We employ two measures, Mountain and Tropics. We use Fearon and Laitin’s mountain variable measured as “the proportion of the country that is mountainous.”83 We measure Tropics as the proportion of the country area in tropical weather from Gallup et al.’s data.84 In line with the selection effects argument, we expect these two measures to have a greater initiation–restraining effect for democratic initiators than nondemocratic ones. We control for other factors found to be important for conflict initiation in the literature, such as geographic proximities, major power status, alliance, rivalries and past conflict. We include a dummy for Major Power coded 1 for a dyad that includes at least one major power state, and 0 otherwise, using the COW state system list (v.2008.1). We measure alliance from the COW alliance data, coded 1 for any dyad with defense pact, neutrality, or entente, and 0 otherwise. We employ two proximity measures for geographic influences, Contiguity and Distance. Contiguity is coded 1

442

Johann Park

when two states are contiguous or separated by less than 150 miles of water according to the COW contiguity data (v.3.1), and 0 otherwise. Distance is the natural logarithm of the great circle distance in miles between two states. Following Rasler and Thompson’s advice, we account for the effect of interstate rivalries in assessing the roles of political institutions for interstate conflict.85 Countries may perceive other countries as friendly and democratic when they need peace, but when they need war, they may define them as hostile and undemocratic.86 We dichotomously measure the rivalry relationship between challenger and target, using Thompson’s perception–based list on strategic interstate rivalries.87 Conflict begets conflict, and peace begets peace. To account for temporal dependence and heteroscedasticity, we use the cubic polynomials of peace years (i.e., t1, t2, and t3) suggested by Carter and Signorino with robust standard errors clustered on dyad.88

Results In this section, we test the two dimensions of democratic selectivity. While Model 1 assesses only the net effects of the selection factors, Model 2 examines their interactive effects conditional upon the regime type of the challengers. We discuss the results for the three liberal factors first and then those for the three material factors. Table 1. MID Initiation by Democracies. Variables Human Rights

Model 1 -.0553*** (.010634)

× Challenger Democracy Target Democracy

.0131*** (.0036)

× Challenger Democracy Interdependence

-5.5954 (5.2233)

× Challenger Democracy Relative Capability

-.0668*** (.0090)

× Challenger Democracy Mountain

-.0014 (.0010)

× Challenger Democracy Tropics × Challenger Democracy

-.1999*** (.0548)

Model 2 .0288* (.0156) -.0085*** (.0014) .0262*** (.0059) -.0012** (.0004) 11.1450*** (4.2645) -1.1532** (.4358) -.0505*** (.0142) -.0017 (.0012) .0002 (.0015) -.0002 (.0001) -.0967 (.0872) -.0107 (.0072)

Democracies and Conflict Initiation Challenger Democracy Contiguity Distance Rivalry Major Power Peace Years Peace Years2 Peace Years3 Constant N Log likelihood

-.0126*** (.0033) .8202*** (.0733) -.1888*** (.0282) .7339*** (.0942) .4971*** (.0553) -.0230*** (.0043) .0003*** (.0001) .0000** (.0000) -1.2085*** (.2318) 371761 -2712.5435

443

.0459*** (.0090) .8891*** (.0725) -.2037*** (.0292) .7375*** (.0960) .5093*** (.0542) -.0235*** (.0044) .0004*** (.0001) .0000*** (.0000) -1.7448*** (.2585) 371761 -2643.8166

Note: Standard errors clustered by dyad are in parentheses. *p<.1, **p<.05, ***p<.01.

Liberal Factors First, Human Rights is significant and negative in Model 1, suggesting that countries, in general, are unwilling to initiate conflict against other countries with a good record of respect for human rights. However, this reluctance to attack human rights–respecting countries does not apply to autocratic challengers according to the interactive analysis presented in Model 2. The main effect term, Human Rights, is positive and significant at the 0.1 level, suggesting that highly autocratic challengers are more likely to initiate an attack against targets with decent human rights conduct than those without. By contrast, the interactive term (Human Rights × Challenger Democracy) is negative and significant at the 0.01 level. Given the net estimate for Human Rights in Model 1 is negative, we can infer that targets’ human rights conduct is conflict–reducing for democratic challengers but conflict–increasing for autocratic challengers. However, examining the marginal effects is necessary for proper interpretation of these non-linear interactive effects.89 We utilize a set of dot plots to examine how the effects of our independent variables on conflict initiation change across all different values of Challenger Democracy. Each dot in Figure 2.A denotes the average marginal effect of Human Rights expressed on the y-axis, with its 90% confidence interval, on the probability of conflict initiation computed over all observations of our independent and control variables while holding Challenger Democracy constant at each individual point on its 21 point (0–20) scale. The plot demonstrates that the conflict–inducing impact of targets’ respect for human rights is limited for fully autocratic challengers whose Challenger

Johann Park

444

0

1

2

3

4

5

6 7

8 9 10 11 12 13 14 15 16 17 18 19 20

Levels of Challenger Democracy

.1

C. Marginal Effects of Interdependecec

0

1

2

3

4

5

6 7

8 9 10 11 12 13 14 15 16 17 18 19 20

Levels of Challenger Democracy

.0001 .0002 -.0003 -.0004

1

2

3

4

5

6 7

8 9 10 11 12 13 14 15 16 17 18 19 20

Levels of Challenger Democracy

0

.00001

.00002

E. Marginal Effects of Mountain

-.00001

Average Marginal Changes in Pr(initiation)

0

-.00002

0

.00005

.0001

.00015

.0002

B. Marginal Effects of Target Democracy

Average Marginal Changes in Pr(initiation)

8 9 10 11 12 13 14 15 16 17 18 19 20

0

1

2

3

4

5

6 7

8 9 10 11 12 13 14 15 16 17 18 19 20

Levels of Challenger Democracy

F. Marginal Effects of Tropics .0005

6 7

Levels of Challenger Democracy

0

5

-.0005

4

-.001

3

D. Marginal Effects of Capability

-.0015

2

Average Marginal Changes in Pr(initiation)

.0004 .0002 0 -.0002 -.0004 -.0006 1

.05

Average Marginal Changes in Pr(initiation)

0

0

Average Marginal Changes in Pr(initiation)

A. Marginal Effects of Human Rights

-.05

Average Marginal Changes in Pr(initiation)

Democracy is 0, the minimum. The marginal effect sign for Human Rights changes into a negative one when Challenger Democracy equals 4 and begins to be significantly negative when it is 6. Further, the conflict–reducing impact of Human Rights increases as Challenger Democracy increases, suggesting higher levels of democracies are more reluctant to attack human rights–respecting countries than any other challenger regime types.

0

1

2

3

4

5

6 7

8 9 10 11 12 13 14 15 16 17 18 19 20

Levels of Challenger Democracy

Figure 2. Interactive marginal effects on the probability of MID initiation, at different levels of Challenger Democracy

Democracies and Conflict Initiation

445

Second, the net effect of Target Democracy is positive and significant in Model 1 suggesting that, in general, countries are more likely to attack democracies than autocracies. Put differently, democracies constitute more susceptible targets for foreign military attacks than autocracies. Yet, in Model 2, the significant negative sign for the interaction term (Target Democracy × Challenger Democracy) indicates that the general susceptibility of democratic targets diminishes with Challenger Democracy. Figure 2.B reveals, whereas Target Democracy significantly increases the probability of conflict initiation for most challenger regime types, this conflict–generating impact is never significant for highly democratic challengers. Our interactive result is consistent with the well-known “cats-and-dogs” effect of autocracy–democracy dyads. However, there is an intriguing puzzle to the well-known dyadic democratic peace as our result shows: democracies do not differentiate fellow democracies from autocracies in choosing military targets. Even the marginal impact sign for Target Democracy, although insignificant, remains positive for democratic challengers. How may this anomaly get addressed? We find Schultz’s information model suggestive. The model argues that, in general, democracies make attractive targets for international conflict. High levels of political competition in democracies create opportunities for potential enemies to probe democracies’ true intentions via military means without unwanted escalations. A challenger can issue a military threat against a democracy “that it will carry out in the event it observes domestic dissent but which it will not carry out if it observes unified support” from the targeted democracy.90 This may be why we find that autocratic challengers are more likely to target democracies than other autocracies. Yet, the strategic susceptibility of democracies as targets of international conflict should be offset with democratic challengers who experience difficulty in justifying military attacks against other fellow democracies. Third, the net effect of Interdependence is negative but statistically insignificant in Model 1. This result is inconsistent with the liberal peace thesis that economic interdependence lessens interstate hostility. Yet, the interactive analysis in Model 2 reveals the causal complexity between interdependence and conflict in the noninteractive specification. Both main effect and interaction terms are statistically significant but with different signs, suggesting interdependence has different effects on conflict initiation between democracies and autocracies. As described in Figure 2.C, interdependence decreases the likelihood of conflict initiation for democratic challengers whose Challenger Democracy is higher than 13. This occurs at a greater rate for highly democratic challengers. Interestingly, interdependence has the opposite effect on autocratic challengers. It significantly increases the likelihood of conflict initiation for autocratic challengers whose Challenger Democracy score is lower than 7. This conflict–inducing effect of interdependence is stronger for highly autocratic challengers. In sum, the interactive results described in Figure 2.C demonstrate that interdependence have different meanings for democracies and autocracies, “hospitality” for the former, but “vulnerability” for autocracies.

446

Johann Park

Material Constraints In Model 1, the net effects for all three material factors, Relative Capability, Mountain, and Tropics, are statistically significant in the expected direction. They all decrease the probability that challengers initiate attacks against targets. In other words, countries are unwilling to initiate military force against other countries with superior military power, highly mountainous terrain, and/or extensive tropical areas. In Model 2, the coefficient signs for all three interactive terms (Relative Capability × Challenger Democracy, Mountain × Challenger Democracy, and Tropics × Challenger Democracy) are negative in line with the selection effects argument, suggesting that democracies have a greater reluctance of attacking strong and difficult international enemies than do autocracies. All of the three interactive coefficients, however, are statistically insignificant as all their p values go beyond 0.1, the traditional minimum significance level. Thus, the data does not support the interactive hypotheses for the material factors (H4, H5, and H6). Also, our examination of interactive average marginal effects through Figures 2.D to 2.F only provide limited support for the selection effects argument at best. First, Relative Capability, on average, does not have greater marginal effects for democratic challengers in constraining conflict initiation than for autocratic challengers. The dots for autocratic challengers even tend to be located in greater negative points on the y axis of average marginal effects than those for democratic challengers (Figure 2.D). Second, the marginal effects of Mountain appear to differ across challenger regime types. Mountain is insignificant in cases where Challenger Democracy is lower than 11 whereas beyond that point, it significantly reduces the probability of conflict initiation. This pacifying effect gets stronger as Challenger Democracy increases. Yet, the imagined line that connects through the marginal effect dots for Mountain is not as steep as those for the liberal factors on the x axis of Challenger Democracy. Third, Tropics have significant negative marginal effects not only for democratic challengers but also for autocratic challengers except for highly autocratic challengers whose Challenger Democracy scores range from 0 to 2. Besides, the dots for the negative marginal effects of Tropics do not make a steeper line as Challenger Democracy increases on the x axis. Therefore, it is hard to conclude that targets’ tropical terrain uniquely, or to a greater extent, restrains conflict initiation against democratic challengers.

Conclusion The institutional variants of the democratic peace theory suggest that bluffing in international crises, engaging in unnecessary, long, and costly wars, and losing wars are especially detrimental to the political survival of democratic leaders. These foreign policy failures are thought to reveal the incompetence of leaders, reduce national reputations, and imperil national security. Therefore, democracies have been assumed to be extraordinarily selective in initiating international conflict. Yet, extant studies have

Democracies and Conflict Initiation

447

been too outcome–oriented, only emphasizing materialistic factors as selection criteria that would affect conflict outcomes. Yet, the findings from recent empirical research suggest that autocratic leaders have reason to make deliberate choices of weak foes as they tend to face more severe punishment than democrats from unsuccessful war outcomes.91 Furthermore, there has been only scant empirical examination about whether these material factors are at work for democracies at the initiation stage of conflict, to which the term “selectivity” pertains more. In this present study, we have proposed what makes democracies different from autocracies in terms of conflict selection lies in normative constraints rather than material constraints. The key elements to democratic selectivity are political accountability and relevant constraints that the institutional features of democracy uniquely exert on political leaders and their decision to initiate armed forces abroad. Leaders in democracies must answer to a broad spectrum of constituencies regarding conflict initiation and its outcome. Our modified public consent model sheds new light on democratic selectivity, producing nuanced propositions. Concerned about the matters at hand, voters and leaders interact at the time of foreign policy enactment. Voters consider the rightness or wrongness of impending armed conflict because armed conflict produces normative costs as well as material ones. Leaders need to minimize these costs and justify causes for aggressive foreign policy actions. The institutional checks exerted by vigorous media, powerful opposition parties, and well-established legislative bodies help citizens be attentive to foreign policy issues and monitor the government’s inappropriate foreign policy actions. When it comes to conflict initiation, all these together force democracies to pick fights they can minimize or justify the associated moral costs. Democratic leaders who ignore these unique institutional constraints and engage in unpopular and disagreeable international conflicts will likely provoke domestic political backlash. Drawing upon this argument, we have identified liberal factors as well as material factors in democracies’ calculus of conflict initiation and provided explicit tests for whether democracies are selective in targeting their military attacks. Our statistical findings show that democracies consider if they can easily win their interstate disputes and thus avoid targeting tough enemies with strong material power, mountainous terrain, and/or tropical topography. These material constraints, however, are not unique to democratic challengers. Autocracies also show a similar degree of reluctance of attacking difficult enemies characterized by the three material factors. Rather, the real difference arises from the target characterization in terms of the three liberal elements, respect for human rights, political representation, and economic interdependence. First, unlike autocratic challengers, democratic challengers are less likely to attack foreign governments that respect citizens’ physical integrity than those that do not. Further, target countries’ respect for human rights exerts a greater pacifying influence for more democratic challengers than less democratic challengers. Second, the effect of target democracy on conflict initiation differs between democratic challengers and autocratic challengers. Target democracy significantly increases the probability of conflict initiation for autocratic challengers. Autocracies tend to target democracies rather than other autocracies for military attacks. Contrastingly, target democracy hardly affects

448

Johann Park

conflict initiation for democratic challengers. Unlike autocracies, democracies barely differentiate target countries’ regime type for their military attacks. This finding itself is not quite consistent with the democratic peace. However, democratic challengers, unlike autocratic challengers, at least do not exploit their fellow democracies despite the general attractiveness of democratic targets for military attacks suggested in Schultz’s information model.92 Third, economic interdependence contrastingly affects conflict initiation between democratic challengers and autocratic ones. Democracies tend to avoid attacking countries with commercial ties. Contrastingly, interdependence only increases the probability that autocracies attack other countries probably because they perceive interdependence in the way realists view it as undesirable dependency and vulnerability that impede the self-help interstate system. Taken together, our investigation of democratic selectivity shows that democracies use different metrics than autocracies in choosing interstate conflict. Yet, this difference does not come from material and physical constraints, but rather from more normative and liberal constraints exerted by political legitimacy and commercial relations. We acknowledge that our findings might carry some inconvenient connotations for the real world. That is, foreign governments that are prone to be charged as irrespective of human rights, unrepresentative, and economically remote constitute susceptible military targets for democratic governments with hawkish ambitions. When leaders in democracies pursue aggressive foreign policy for some controversial purposes, illiberal foreign governments can be made scapegoats for military adventurism. These leaders will claim some liberal causes to justify their motivations behind violent military actions. A well-known controversial case is Operation Desert Fox. On December 16, 1998, one night before the House was about to start open floor deliberations on his impeachment, President Clinton spoke to the nation relying on liberal causes: Earlier today, I ordered America’s armed forces to strike military and security targets in Iraq.… The hard fact is that so long as Saddam remains in power, he threatens the well-being of his people, the peace of his region, the security of the world. The best way to end that threat once and for all is with a new Iraqi government—a government ready to live in peace with its neighbors, a government that respects the rights of its people. 

Another well-known controversial case can be found in the way the Bush and Blair administrations justified the lead-up to the invasion of Iraq on March 19, 2003 based on the 9/11 attack. In addition to Iraq’s alleged WMD development and purported link to al-Qaeda, regime change and stopping human rights violations were frequently claimed rationales.93 Afterwards, the continuing failure to find evidence for the WMD and alQaeda claims even shifted the justification emphasis to the liberal objectives.94 Our arguments and findings also have important implication for future research. Although there are many findings supportive of the democratic victory hypothesis, critics have continued to challenge the democracy–success relationship. These critics show that Reiter and Stam’s main finding—that democratic initiators tend to win wars— is not robust in its research design, measurement, and model specifications.95 One

Democracies and Conflict Initiation

449

common flaw found in this debate is the failure to consider the non-material dimension of conflict selection. Democracies may go to war for just and grand causes even when they do not have overwhelming power advantages. When war aims are more justifiable, it may increase the chances of winning the war. Increased justifiability of war may help the mobilization of support and resources that are necessary for successful war conduct. Therefore, including some normative justification factors in the statistical model of conflict outcomes can help reveal the true relationship between democracy and war victory. The justifiability dimension of conflict selection can also help answer questions such as, which kinds of countries are susceptible to foreign attacks in general? The roles of justice in international conflict have been neglected in typical large-n studies. By considering justifiability, we can attempt to link the traditional just war thesis to the scientific study of war. In particular, it is reasonable to argue that governments that treat their people unfairly and unjustly (indicated by human rights abuses and political and economic inequalities) are vulnerable to invasions by foreign forces. From the perspective of potential attackers, countries that are considered unjust domestically and internationally are more susceptible to invasion. First, it would be easier for an attacker to justify its military adventure to both domestic and international societies. Second, unjust countries’ armed forces are likely structured to oppress the internal threat from the dissatisfied at the expense of efficiency in coping with external threats.

Notes 1.

2.

3. 4. 5. 6. 7. 8.

For a recent comprehensive review for the democratic peace literature, see Havard Hegre, “Democracy and Armed Conflict,” Journal of Peace Research 51 (2014): 159–72. For the latest supportive findings, see Johann Park, “Forward to the Future? The Democratic Peace after the Cold War,” Conflict Management and Peace Science 30 (2013): 178–94. Suzanne Werner, “The Effects of Political Similarity on the Onset of Militarized Disputes, 1816–1985,” Political Science Quarterly 53 (2000): 343–74; and Mark Peceny, Caroline Beer, and Shannon Sanchez-Terry, “Dictatorial Peace?” American Political Science Review 96 (2002): 15–26. Park and James even find that democracies are more likely to use militarized means when issues at stake are increasingly salient. See Johann Park and Patrick James, “Democracy, Territory, and Armed Conflict, 1919–1995,” Foreign Policy Analysis 11 (2015): 85–107. Jessica Weeks, “Strongmen and Straw Men: Authoritarian Regimes and the Initiation of International Conflict,” American Political Science Review 106 (2012): 326–47. Dan Reiter and Allan Stam, Democracies at War (Princeton, NJ: Princeton University Press, 2002). Bruce Bueno de Mesquita, Alastair Smith, Randolph Siverson, and James Morrow, The Logic of Political Survival (Cambridge, MA: MIT Press, 2003). Kenneth Schultz, Democracy and Coercive Diplomacy (Cambridge, UK: Cambridge University Press, 2001). Michael Desch, Power and Military Effectiveness: The Fallacy of Democratic Triumphalism (Baltimore, MD: Johns Hopkins University Press, 2008). Alexandre Debs and Hein Goemans, “Regime Type, the Fate of Leaders and War,” American Political Science Review 104 (2010): 430–46

450

Johann Park

Lawrence Jacobs and Benjamin Page, “Who Influences Foreign Policy?” American Political Science Review 99 (2005): 107–23. 10. Christopher Gelpi, Jason Reifler, and Peter Feaver, “Iraq the Vote: Retrospective and Prospective Foreign Policy Judgments on Candidate Choice and Casualty Tolerance,” Political Behavior 29 (2007): 151–74. 11. Martha Finnemore, “Constructing Norms of Humanitarian Intervention,” in The Culture of National Security: Norms and Identity in World Politics, ed. Peter Katzenstein (New York, NY: Columbia University Press, 1996), 153. 12. George Lakoff, “Metaphor and War: The Metaphor System Used to Justify War in the Gulf,” Peace Research 23 (1991): 25–32; and Jessica Weeks, “Strongmen and Straw Men: Authoritarian Regimes and the Initiation of International Conflict,” American Political Science Review 106 (2012): 326–47. 13. Kenneth Schultz, “Looking for Audience Costs,” Journal of Conflict Resolution 45 (2001): 32–60.” 14. Randolph Siverson, “Democracies and War Participation: In Defense of the Institutional Constraints Argument,” European Journal of International Relations 1 (1995): 481–89; and Reiter and Stam, Democracies at War. 15. Bueno de Mesquita et al., The Logic of Political Survival. 16. James Fearon, “Domestic Political Audiences and the Escalation of International Disputes,” American Political Science Review 88 (1994): 577–92. 17. Seung-Whan Choi and Patrick James, “Media Openness, Democracy and Militarized Interstate Disputes,” British Journal of Political Science 37 (2007): 23–46. 18. Kenneth Schultz, Democracy and Coercive Diplomacy. 19. Ibid. 20. Siverson, “Democracies and War Participation”; Reiter and Stam, Democracies at War; and Bueno de Mesquita, Bruce, Michael Koch, and Randolph Siverson, “Testing Competing Institutional Explanations of the Democratic Peace: The Case of Dispute Duration,” Conflict Management and Peace Science 21 (2004): 255–67. 21. Kenneth Schultz, Democracy and Coercive Diplomacy. 22. Bueno de Mesquita et al., The Logic of Political Survival, 241. 23. Sebastian Rosato, “The Flawed Logic of Democratic Peace Theory,” American Political Science Review 97 (2003): 585–602; Desch, Power and Military Effectiveness; and Weeks, “Strongmen and Straw Men.” 24. Chaim Kaufmann, “Threat Inflation and the Failure of the Marketplace of Ideas: The Selling of the Iraq War,” International Security 29 (2004): 5–48; Jane Cramer, “Militarized Patriotism: Why the U.S. Marketplace of Ideas Failed before the Iraq War,” Security Studies 16 (2007): 489–524; and Alexander Downes, “How Smart and Tough are Democracies? Reassessing Theories of Democratic Victory in War,” International Security 33 (2009): 9–51. 25. Desch, Michael. 2008. Power and Military Effectiveness. 26. Ibid., 48. 27. Ibid., 49. 28. Sally Anderson and Mark Souva, “The Accountability Effects of Political Institutions and Capitalism on Interstate Conflict,” Journal of Conflict Resolution 54 (2010): 543–65. 29. Bueno de Mesquita et al., The Logic of Political Survival. 30. Kevin Clark and Randall Stone, “Democracy and the Logic of Political Survival,” American Political Science Review 102 (2008): 387–92. For information on the Polity index, see Jaggers, Keith, and Tedd Robert Gurr, “Tracking Democracy’s Third Wave with the Polity III Data. Journal of Peace Research 32 (1995): 469–82. 31. Gelpi, Christopher, Jason Reifler, and Peter Feaver, “Iraq the Vote.” 32. Rosato, “The Flawed Logic of Democratic Peace Theory.” 33. Reiter and Allan Stam, Democracies at War, 6. 34. Ibid., 6. 35. Michael Nickelsburg and Helmut Norpth, “Commander-in-chief or Chief Economist? The 9.

Democracies and Conflict Initiation

36.

37.

38.

39. 40. 41. 42.

43.

44.

45. 46. 47. 48. 49. 50. 51.

451

President in the Eye of the Public,” Electoral Studies 19 (2000): 313–32; Matthew Singer, “Who says ‘It’s the Economy’? Cross-National and Cross-Individual Variation in the Salience of Economic Performance,” Comparative Political Studies 44 (2011): 284–312; and John Aldrich, Christopher Gelpi, Peter Feaver, Jason Reifler, and Kristin Thompson Sharp, “Foreign Policy and the Electoral Connection,” Annual Review of Political Science 9 (2006): 477–502. John Mueller, War, Presidents and Public Opinion (New York, NY: John Wiley & Son, 1973); Jon Hurwitz and Mark Peffley, “The Means and Ends of Foreign Policy as Determinants of Presidential Support,” American Journal of Political Science 31 (1987): 236–58; James Fearon, “Domestic Political Audiences and the Escalation of International Disputes”; and Lawrence Jacobs and Benjamin Page, “Who Influences Foreign Policy?” American Political Science Review 99 (2005): 107–23. Darren Davis, Negative Liberty: Public Opinion and the Terrorist Attacks on America (New York, NY: Russell Sage Foundation, 2007); Adam Berinsky, In Time of War: Understanding American Public Opinion from World War II to Iraq (Chicago, IL: University of Chicago Press, 2009); and Jennifer Merolla and Elizabeth Zechmeister, Democracy at Risk: How Terrorist Threats Affect the Public (Chicago, IL: University of Chicago Press, 2009). Shoon Murray, “Tough Talk, Public Predispositions, and Military Action: Reassessing the Rally-’Round-the-Flag Phenomenon,” in Approaches, Levels, and Methods of Analysis in International Politics: Crossing Boundaries, ed. Harvey Starr (New York, NY: Palgrave, 2006), 141–56. Michael Colaresi, “The Benefit of the Doubt: Testing an Informational Theory of the Rally Effect,” International Organization 61 (2007): 99–143; and Laron Williams, “Flexible Election Timing and International Conflict,” International Studies Quarterly 57 (2013): 449–61. Schultz, Democracy and Coercive Diplomacy. Choi and James, “Media Openness, Democracy and Militarized Interstate Disputes.” David Clark and Timothy Nordstrom, “Democratic Variants and Democratic Variance: Examining Domestic Processes and Interstate Conflict,” Journal of Politics 67 (2005): 250– 70; and Seung-Whan Choi, “Legislative Constraints: A Path to Peace?” Journal of Conflict Resolution 54 (2010): 438–70. Robert Putnam, Making Democracy at Work: Civic Traditions in Modern Italy (Princeton, NJ: Princeton University Press, 1993); and Dan Reiter and Erik Tillman, “Public, Legislative, and Executive Constraints on the Democratic Initiation of Conflict,” Journal of Politics 64 (2002): 810–26. Dan Reiter and Erik Tillman, “Public, Legislative, and Executive Constraints on the Democratic Initiation of Conflict”; and David Leblang and Steven Chan, “Explaining Wars Fought by Established Democracies: Do Institutional Constraints Matter?” Political Research Quarterly 56 (2003): 385–400. Zeeve Maoz and Bruce Russett, “Normative and Structural Causes of Democratic Peace, 1946– 1986,” American Political Science Review 87 (1993): 624–38. Fearon, “Domestic Political Audiences and the Escalation of International Disputes”; Schultz, Democracy and Coercive Diplomacy; and Reiter and Tillman, “Public, Legislative, and Executive Constraints on the Democratic Initiation of Conflict.” Richard Lebow, “Thucydides the Constructivist,” American Political Science Review 95 (2001): 548. Maoz and Russett, “Normative and Structural Causes of Democratic Peace, 1946–1986.” Richard Lebow, “Classical Realism,” in International Relations Theories: Discipline and Diversity, eds. Tim Dunne, Milja Kurki, and Steve Smith (UK: Oxford University Press, 2007), 59.  E. H. Carr, The Twenty Years’ Crisis, 1919–1939 (Ann Arbor, MI: University of Michigan Press, 1946), 235. Rosato, “The Flawed Logic of Democratic Peace Theory”; Giacomo Chiozza and Hein Goemans, “International Conflict and the Tenure of Leaders: Is War Still Ex Post Inefficient,” American Journal of Political Science 48 (2004): 604–19; and Alexandre Debs and Hein

452

52. 53. 54. 55. 56. 57.

58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68.

69. 70. 71. 72.

Johann Park Goemans, “Regime Type, the Fate of Leaders and War,” American Political Science Review 104 (2010): 430–46. Bruce Russett and John Oneal, Triangulating Peace: Democracy, Interdependence, and International Organizations (New York, NY: Norton, 2001). John Owen, “How Liberalism Produces Democratic Peace,” International Security 19 (1994): 95. Michael Doyle, “Kant, Liberal Legacies, and Foreign Affairs,” Philosophy and Public Affairs 12 (1983): 230. Fiona McGillivray and Alastair Smith, “Trust and Cooperation through Agent–specific Punishments,” International Organization 54 (2000): 809–24. Katherine Barbieri, The Liberal Illusion: Does Trade Promote Peace? (Ann Arbor, MI: University of Michigan Press, 2002). Erik Gartzke, Quan Li, and Charles Boehmer, “Investing in the Peace: Economic Interdependence and International Conflict,” International Organization 55 (2001): 391– 438; John Oneal, Bruce Russett, and Michael Berbaum, “Causes of Peace: Democracy, Interdependence, and International Organizations, 1885–1992,” International Studies Quarterly 47 (2003): 371–93; and Havard Hegre, John Oneal, and Bruce Russett, “Trade Does Promote Peace: New Simultaneous Estimates of the Reciprocal Effects of Trade and Conflict,” Journal of Peace Research 47 (2010): 763–74. James Morrow, “How Could Trade Affect Conflict?” Journal of Peace Research 36 (1999): 481–89. Quan Li and David Sacko, “The (IR) Relevance of Interstate Militarized Disputes to International Trade,” International Studies Quarterly 46 (2002): 11–44. Anne Krueger, “Trade Policy and Economic Development: How We Learn,” American Economic Review 87 (1997): 1–22; and Sebastian Edwards, “Openness, Productivity and Growth: What Do We Really Know?” Economic Journal 108 (1998): 383–98. Thomas Oatley, International Political Economy: Interests and Institutions in the Global Economy, 2nd ed. (New York, NY: Longman, 2005), Chapter 3. Bueno de Mesquita et al., The Logic of Political Survival; and Christopher Gelpi and Joseph Grieco, “Democracy, Trade and the Nature of the Liberal Peace,” Journal of Peace Research 45 (2008): 17–36. Han Dorussen and Hugh Ward, “Trade Networks and the Kantian Peace,” Journal of Peace Research 47 (2010): 29–42. Karl Deutsch, Sidney Burrell, and Robert Kann, Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience (Princeton, NJ: Princeton University Press, 1957). Katja Kleinberg and Benjamin Fordham, “Trade and Foreign Policy Attitudes,” Journal of Conflict Resolution 54 (2010): 687–714. Gelpi and Grieco, “Democracy, Trade and the Nature of the Liberal Peace.” Barbieri, The Liberal Illusion. Kenneth Waltz, “The Myth of National Interdependence,” in The International Corporation, ed. Charles Kindleberger (Cambridge, MA: MIT Press, 1970), 205–23; and John Mearsheimer, “Back to the Future: Instability in Europe after the Cold War,” International Security 15 (1990): 5–56. Mearsheimer, “Back to the Future.” Michael Colaresi and William Thompson, “The Economic Development–Democratization Relationship: Does the Outside World Matter?” Comparative Political Studies 36 (2003): 381– 403. For a nuanced view, see Daniel Drezner, “The Realist Tradition in American Public Opinion,” Perspective on Politics 6 (2008): 51–70. He argues that Americans are at least as comfortable with realism as liberalism and that elites are more inclined to liberalism than the public. Bruce Russett and Miroslav Nincic, “American Opinion on the Use of Military Force Abroad,” Political Science Quarterly 91 (1976): 411–31.

Democracies and Conflict Initiation

453

73. Bruce Jentleson and Rebecca Britton, “Still Pretty Prudent: Post–Cold War American Public Opinion on the Use of Military Force,” Journal of Conflict Resolution 42 (1998): 399–400. 74. Richard Eichenberg, “Victory Has Many Friends: U.S. Public Opinion and the USE of Military Force, 1981–2005,” International Security 30 (2005): 140–77. 75. Peter Feaver and Christopher Gelpi, Choosing Your Battles: American Civil–Military Relations and the Use of Force (Princeton, NJ: Princeton University Press, 2004). 76. Christopher Gelpi, Peter Feaver, and Jason Reifler, Paying the Human Costs of War (Princeton, NJ: Princeton University Press, 2009). 77. Scott Bennett and Allan Stam, “EUGene: A Conceptual Manual,” International Interactions 26 (2000): 179–204. 78. David Cingranelli and David Richards. “The Cingranelli–Richards (CIRI) Human Rights Dataset,” (2014), available at http://www.humanrightsdata.com (accessed August 8, 2017). 79. We focus only on this sub-class of human rights, as opposed to others such as empowerment rights and women’s political rights, because inflicting arbitrary physical harm on citizens will be seen as the most offensive to the universal ideas on human dignity. See Margaret Keck and Kathryn Sikkink, Activists beyond Borders: Advocacy Networks in International Politics (Ithaca, NY: Cornell University Press, 1998); and James Walsh and James Piazza, “Why Respecting Physical Integrity Rights Reduces Terrorism,” Comparative Political Studies 43 (2010): 551–77. 80. Monty Marshall, Ted Gurr, and Keith Jaggers, “Political Regime Characteristics and Transitions, 1800–2015: Dataset Users’ Manual,” (2016), available at http://www.systemicpeace.org/inscr/ p4manualv2015.pdf (accessed August 8, 2017). 81. Russett and Oneal, Triangulating Peace. 82. Kristian Gleditsch, “Expanded Trade and GDP Data,” Journal of Conflict Resolution 46 (2000): 712–24. 83. James Fearon and David Laitin, “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97 (2003): 81. 84. John Gallup, Andrew Mellinger, and Jeffrey Sachs. “Geography Datasets: General Measures of Geography: 1) Physical Geography and Population,” (2001), available at http://www.cid. harvard.edu/ciddata/geographydata.htm (accessed May 23, 2011) 85. Karen Rasler and William Thompson, “Rivalries and the Democratic Peace in the Major Power Subsystem,” Journal of Peace Research 38 (2001): 659–83. 86. Christopher Layne, “Kant or Cant: The Myth of the Democratic Peace,” International Security 19 (1994): 5–49. 87. William Thompson, “Identifying Rivals and Rivalries in World Politics,” International Studies Quarterly 45 (2001): 557–86. 88. David Carter and Curtis Signorino, “Back to the Future: Modeling Time Dependence in Binary Data,” Political Analysis 18 (2010): 271–92. 89. The marginal effect of an independent variable on the probability of the dependent variable is different over its different values and different values of all other independent variables even with no explicit interaction term in logit models. See Churong Ai and Edward Norton, “Interaction Terms in Logit and Probit,” Economics Letters 80 (2003): 123–29. We use average marginal effects computed as “the average of discrete or partial changes over all observations” of all independent variables without setting them at an arbitrary fixed value, following Tamas Bartus, “Estimation of Marginal Effects Using Margeff,” Stata Journal 5 (2005): 310. For interactive effects, we examine how the average marginal effect of an independent variable changes across all different values of our conditioning variable, Challenger Democracy. See Pinar Karaca-Mendic, Edward Norton, and Bryan Dowd, “Interaction Terms in Nonlinear Models,” Health Services Research 47 (2012): 225–74. 90. Schultz, Democracy and Coercive Diplomacy, 99. 91. Debs and Goemans, “Regime Type, the Fate of Leaders and War.” 92. Schultz, Democracy and Coercive Diplomacy. 93. Christoph Bluth, “The British Road to War: Blair, Bush, and the Decision to Invade Iraq,”

454

Johann Park

International Affairs 80 (2004): 871–92; and Andrew Flibbert, “The Road to Baghdad: Ideas and Intellectuals in Explanations of the Iraq War,” Security Studies 15 (2006): 310–52. 94. Nick Assinder, “Blair Makes Significant Tone Change,” BBC News, July 17, 2003; Marc Sandalow, “News Analysis: Record Shows Bush Shifting on Iraq War,” San Francisco Chronicle, September 29, 2004; and Tom Raum, “Bush Keeps Revising War Justification,” Associated Press, October 15, 2006. 95. Reiter and Stam, Democracies at War. For critics of Reiter and Stam’s findings, see Desch, Power and Military Effectiveness; and Downes, “How Smart and Tough are Democracies?” Reiter and Stam dispute these critics’ approach. See Dan Reiter and Allan Stam, “Correspondence: Another Skirmish in the Battle over Democracies and War,” International Security 34 (2009): 194–200.

Notes on Contributor Johann Park  is an assistant professor of Political Science and International Relations at Incheon

National University. Previously, he worked at U.S. institutions such as the University of California– Merced and Mississippi State University as an assistant professor. Recently, he returned to Korea to teach at his alma mater, Incheon National University. His research on international conflict and terrorism has been published in major Political Science and International Relations journals such as International Studies Quarterly, Journal of Conflict Resolution, Electoral Studies, Conflict Management and Peace Science, and Foreign Policy Analysis, among others. He holds a Ph.D. in Political Science from Michigan State University and an M.A. from the University of Missouri– Columbia.