Title Menu - The International Journal of Peace Studies.  
Return to Table of Contents Return to I J P S Home




Brandon C. Prins



In the post-Cold War world, democratic enlargement has occupied the attention of many political elites, particularly in the United States. Not only have bureaucrats at Foggy Bottom begun to appreciate the importance of regime type in structuring foreign policy relationships, but policymakers on the Hill and in the White House have begun to press for regime change as well. In defending NATO enlargement into Eastern Europe and the Baltics, for example, President Clinton presented a Kantian vision of the future, where security guarantees help cement democratic gains leading in the end to a more stable and peaceful Europe (see Lake, 1994; Albright, 1998; Maynes, 2000).1 Additionally, nurturing Russia's democratic transition was a core aim of Clinton's foreign policy team. "No event in the last half-century," Sandy Berger (2000: 7) insisted "has done more to advance our security than Russia's democratic revolution." Clinton's Deputy Secretary of State Strobe Talbot (1996) also made a good case for democratic enlargement. According to Talbot (1966: 63) "democracies are demonstrably more likely to maintain their international commitments, less likely to engage in terrorism or wreak environmental damage, and less likely to make war on each other. [Further,] only in an increasingly democratic world will the American people feel themselves truly secure."

"The Clinton administration strategy," according to Gowa (1995: 511), "reflects the finding of a rapidly growing body of literature in international relations: democratic states pursue distinctive foreign policies." Yet, in spite of empirical support for the democratic peace, disagreement still exists on the causal factors inhibiting conflict among liberal states (Gowa, 1995; Elman, 1997; Gartzke, 2001). Many democratic peace theorists maintain that democratic norms and political institutions inhibit conflict initiation (for example, Maoz and Russett, 1993; Russett, 1993). Other scholars, however, suggest that the pacific effects of liberal regimes may not so much be a function of their ability to avoid conflict, but rather their capacity to resolve conflict short of armed hostilities (Raymond, 1994; Dixon, 1993). For instance, research by Bercovitch (1996), Dixon (1993), and Raymond (1994) indicates that democratic regimes have a higher propensity to resolve their disputes through the use of third-party intermediaries. That is, democracies appear to be more likely than non-democracies to submit to the peace-making attempts of third-party states. Raymond (1994: 27) concluded that, "disputes between democracies rarely escalate to war because each side expects the other to rely on peaceful means of conflict resolution."

In spite of this evidence, Fearon (1994) has suggested that latent electoral costs can incite democratic leaders to escalate militarized quarrels. Thus, rather than encourage the pacific resolution of disputes, the fear of electoral retribution may convince leaders to escalate conflicts. Needless to say, this incentive to prevent or conceal policy failure may be a particular shortcoming of democratic states. Chan (1997) similarly suggests that democracy may not inhibit the escalation of militarized conflict. He writes


Simple frequency counts of past conflict involvement cannot address the core claim of the democratic peace proposition, because they fail to distinguish between initiators and defenders in international hostilities. The democratic peace proposition contends that for structural or cultural reasons democracies are less able or willing to initiate violence or to start war. It does not argue that, if attacked, democracies will fail to respond in kind (1997: 68).


As evidence supporting Chan's conjecture, Senese (1997) found democratic dyads to be equally as likely as non-democratic dyads to use military force in disputes. It may be, then, that democratic leaders are willing to meet militarized demands with force regardless of the regime type of the initiating state. 

The purpose of this paper is to delve further into this conflict puzzle. While considerable evidence shows democracies avoid militarized violence with similar states, the theoretical arguments of Fearon (1994) and Chan (1997) coupled with the empirical evidence of Senese (1997) alludes to the possibility that cultural norms and institutional constraints, which theoretically should resolve disputes through bargaining and compromise, may be insufficient once a demand has been made. I examine democratic dispute reciprocation from 1816-1992. That is, given a dispute, are democratic states more or less likely to reciprocate when the initiator is another democracy? A focus on the decision to reciprocate seemingly provides an appropriate framework to assess whether democratic regimes condition their behavior on the regime type of the opponent. The results I find indicate that the relationship between regime type and dispute reciprocation supports Chan’s (1997) conjecture, but only for emerging or developing democracies. In fact, bilateral militarized disputes between immature democratic states actually have the highest propensity for conflict reciprocation, 38% higher than disputes between non-democracies. Fully institutionalized democracies, on the other hand, experience few disputes. Plus, when targeted, they tend to tailor their foreign policy behavior to the regime type of the initiating state. The results suggest that the pacific effects of democracy may only take hold once liberal institutions and political culture become sufficiently entrenched.



Democratic Politics and Conflict Propensity


According to Russett and Starr (2000: 93), "democracies very rarely–if at all–make war on each other." This empirical finding has profound normative implications as evidenced by the Clinton administration's attention to democratic enlargement. If recent events are any indication, this policy may help to alleviate historical animosity among even the most rivalrous of states. Indeed, recent additions to the democratic club appear to support the non-violent tendencies of liberal regimes. For example, the development of democratic norms and political institutions in the states of the former Communist bloc has coincided with a significant decrease in the tensions between Cold War foes. In other regions as well increased levels of democracy have appeared to reduce conflict levels. In South America, for example, Argentina and Brazil have stopped their respective nuclear programs and engaged in confidence building measures (Beckman et al., 2000), and Ecuador and Peru have recently signed a treaty delimiting territorial boundaries. While relations between old foes Greece and Turkey have not exactly been friendly, these two states have not seriously fought each other since 1974 and the solidification of democratic norms and institutions should help to mitigate any further outbreaks of serious violence. A similar story can be told for India and Pakistan, although recent democratic setbacks in the latter state and nuclear testing by the Hindu Nationalist Party in India have certainly increased tensions in the region. On par, however, even recent events suggest that democratic regimes are less belligerent and more trustworthy than other types of states.

Norms, Constraints, and the Liberal Rationale


Liberal ideology serves as a foundation for democratic peace arguments. That is, at the core of what prevents liberal states from fighting each other is a widespread belief in individual freedoms, such as speech and religion, tolerance of others, and competitive elections that allow citizens a measure of input into the foreign policy decisions of state leaders (Owen, 1994). This shared liberal ideology, according to Chan (1997: 75), provides the "political foundation for a league of peace in which, over time, norms of reciprocity and expectations concerning a preference for nonviolent procedures develop to regulate interaction."  Thus, both liberal norms and democratic political institutions help shape state behavior by conveying a preference for non-aggression and the rule of law.

According to some international relations scholars, a core element helping to explain the observed lack of violent militarized encounters between democratic states centers on the widespread acceptance of bargaining and compromise as the legitimate means of resolving domestic conflicts of interest (Russett, 1993). At the domestic level, citizens rely upon legislatures and courts to resolve interpersonal disputes. And this domestic-level preference for adjudication and negotiation is then transferred to interstate interactions (Maoz and Russett, 1993). Rather than relying upon gunboat diplomacy or power politics, democratic political elites expect to resolve disputes through mutual accommodation. Leng (1993), for example, has found that democracies use less coercive foreign policy strategies in crises than non-democratic states. Thus, democratic political culture may help to create an expectation of mutual respect and non-violent conflict resolution (Russett, 1993; Elman, 1997).2

Domestic political institutions, which are often created to formalize political or cultural norms, are also thought to contribute to the mitigation of violent conflict between democratic states. This is because the checks and balances inherent in democratic polities, as well as the openness of the democratic political process, help to prevent leaders from engaging in unprovoked military actions.3 These checks and balances between different individuals and institutions promote conflict resolution strategies that avoid violent confrontations with the leaders of other states. According to Maoz and Russett (1993), the decision to use force becomes more difficult as the need to secure political support from multiple domestic groups increases.4 For example, the multiple layers of possible participation presented by the American federal system encourages executives to seek political support from bureaucratic agencies, legislators, and important interest groups.5 Consequently, as Russett (1993: 80) observed, “Federalism restricts the ability [of executives] to mobilize economic and military resources rapidly in the event of a serious international dispute.”

Likewise, Bueno de Mesquita and Lalman (1992) maintain that democratic institutions and values present visible manifestations of constraint that are likely to be seen by other democratic states. In an anarchical international environment where it is often difficult to discern the preferences of states, democracy is a simple and effective way of distinguishing friend from foe (Elman, 1997). Given the high political costs involved in using force, leaders will recognize that democratic institutions present an impediment to violent engagement. Indeed, Schultz (1998, 2001) and Bueno de Mesquita, Morrow, Siverson, and Smith (1999) suggest that the political costs democratic leaders face for policy failure contributes to efficient signaling of preferences and thus makes militarized demands more credible. Democratic leaders avoid violent escalation because they rarely bluff about their preferences. "The most important quality that a republican government brings to [the] table,” Huntley (1996) writes, “is not a 'peaceful disposition,' but rather a capability to be trusted" (quoted in Chan, 1997: 81).6


Are Democracies More Pacific?


While the evidence in support of a democratic peace is substantial, many scholars continue to question the statistical results. Some critics insist that the lack of war between democratic states is a statistical anomaly that is driven by both the rarity of war and the rarity of democracy, as well as by the arbitrary operationalizations of both concepts (Layne, 1994; Oren, 1995; Gowa, 1995). Other scholars question the exclusion of both extra-systemic wars (Henderson, 2000) and covert attempts to undermine democratic political institutions (Spiro, 1994) and insist that power considerations are substantively more critical (Farber and Gowa, 1995).7

A perhaps more insightful critique of democratic peace begins with the causal logic grounding the two prevalent explanations. While the normative account has received considerable empirical support at the dyadic level, it still suffers from theoretical imprecision. For instance, it is not clear why the non-violent norms of diplomacy practiced by democratic states should necessarily be dominated by the arguably deceitful practices of authoritarian elites. The high costs of war should naturally compel leaders to use diplomatic means first to resolve contentious issues. In fact, given the power-potential of most democracies, it is surprising that scholars continue to insist that autocrats target democracies as a result of their perceived domestic political weaknesses.8 Indeed, evidence exists which suggests that salient features of the domestic environment can push democratic leaders into disputes (Ostrom and Job, 1986; Morgan and Bickers, 1992; Smith, 1996). Economic weakness and the prospects of reelection may convince democratic leaders to use military force to demonstrate foreign policy acumen. While autocrats may repress, democratic political elites look abroad to create a rally at home. Targeting democracies during periods of domestic weakness, then, may present leaders with the perfect opportunity to evince foreign policy leadership (see Leeds and Davis, 1997).

Plus, Fearon (1994) has argued that the domestic costs for backing down in the face of a challenge may convince democratic leaders to escalate disputes in hopes of avoiding legislative and electoral sanctions for policy failure. Russett (1990), as well, acknowledges that constraints on an executive may not serve to impede minor uses of force, such as those associated with the diversionary literature. Indeed, Russett (1990: 43) found that the U.S., and democratic states more generally, were more likely to engage in international disputes during economic downturns. "Faced with [considerable domestic] discontent," Russett (1990: 24) wrote, "even a democratically elected government may feel some temptation to try to divert hostility toward foreign adversaries."9 Thus, targeting a democratic state during a period of domestic weakness may embroil an autocrat in a fight he doesn’t really want and most likely can’t afford.

Arguments highlighting institutional constraints on the use of force also suffer from theoretical holes. First, the structural argument is a monadic-level explanation that would seem to fall apart because of the lack of evidence for a monadic-level peace (see Gartzke, 2001). If checks and balances prevent conflict between democratic leaders, why then is there little evidence demonstrating that these structural constraints inhibit conflict between democracies and non-democracies? Moreover, if democratic leaders are able to circumvent normal political processes when facing non-democratic elites, then these supposed structural constraints surely could be manipulated in other instances as well. This tends to cast some doubt on the theorized relationship between democratic institutions and constraints on the use of force.


A Progressive Debate


While scholars continue to refine the causal logic grounding democratic peace, others have begun to address some of the empirical anomalies uncovered thus far. First, recent evidence indicates that democracies are more prone than non-democracies to come to the aid of like regimes in peril. Both Raknerud and Hegre (1997) and Werner and Lemke (1997) find empirical support for joining behavior on the part of democracies.10 Plus, Prins (1998) finds that the U.S. is considerably more likely to offer its diplomatic assistance in resolving interstate crisis situations when a democracy is involved. Not only may these empirical results help explain the lack of evidence for a monadic-level democratic peace, but these studies also highlight how political similarities tend to propel democracies towards partnership rather than hostility. They also demonstrate that third-party states may play a crucial role in crisis bargaining.11 Smith (1996) has even suggested that the decisions of state leaders to initiate hostilities cannot be fully understood without accounting for the potential third-party participant.

Some scholars have also begun to explore within group differences with regards to foreign policy behavior. Prins and Sprecher (1999), for instance, find that conflict propensities differ depending on the nature of the parliamentary government in power. Specifically, they observe that coalition parliamentary governments have a higher probability of reciprocating disputes than single-party cabinets, and that increases in the polarization of the legislature tend to decrease reciprocation rates. Schjølset (1996) has also noted that the federal/centralized distinction has an important impact on conflict propensity. Her evidence indicates that centralized and parliamentary democracies are generally more conflict prone than federal and presidential systems.12 And recently, Palmer and Regan (1999) have found that the liberal-conservative orientation of democratic governments can additionally play a role in the propensity to use force (also see Hagan, 1994). Needless to say, these analyses begin to refine our understanding of the role played by democratic political processes and institutions.

A further extension of the democratic peace project has been to specify the initiating actor. According to Chan (1997: 68), “even though the role of initiator of violence does not necessarily mean the country in question is the aggressor in a particular conflict, it is still the most important discriminating indicator for examining the democratic peace proposition” (also see Ray, 1999). Chan (1997: 82) reminds us that once in a crisis, democratic institutions and norms will not necessarily prevent conflict between liberal states. Consequently, as Chan (1997: 82) goes on to write, “it is important not to conflate ‘the effects of democracy on the emergence of crises with its effects on the escalation of crises.’” However, empirical evidence does exist which suggests that democracies continue to condition their behavior once a crisis or dispute threshold has been breached. Certainly the findings of Bercovitch (1996), Dixon (1993), and Raymond (1994) demonstrate that democratic states pursue distinctive conflict resolution techniques, which attempt to avoid coercion. Moreover, Rousseau et al. (1996: 527) concluded, “Once a democracy is involved in an international crisis, it carefully distinguishes the type of state with which it is bargaining and adjust its bargaining behavior accordingly.”

Despite the rich literature on democratic peace and the large amount of empirical evidence that has already been collected, theoretical and empirical questions still remain. The research design presented below is intended to address two specific aspects of the democratic peace puzzle. First, are democracies equally as likely as non-democracies to reciprocate dispute situations? Second, is the decision to reciprocate influenced by the regime type of the initiating state? Based on the discussion above, two hypotheses are proposed.


H1:   Democratic targets are equally as likely as non-democratic targets to reciprocate militarized interstate disputes (monadic proposition).


H2:  Bilateral disputes involving democracies are less likely to experience dispute reciprocation than non-democratic or mixed regime type disputes (dyadic proposition).



Research Design


In assessing the decision to reciprocate, attention is devoted to the regime types of both the initiating and defending states.13 Senese (1997) only considered whether the highest level reached in a dispute varied by dyad type (democratic-democratic vs. other). While this research design allows one to assess whether democratic dyads are less violence prone than other types of regime pairs, it does not account for how the targeted state reacts to the initial demand. That is, democratic disputes may initially involve a demand made with the use of military force. Non-democracies, on the other hand, may initiate disputes with a lower hostility level. Without considering the response of the targeted state, a complete picture of the strategic bargaining that takes place is elusive. Democracies may rarely reciprocate with a militarized demand, preferring to resolve the quarrel through mutual accommodation. Non-democracies, in contrast, may meet militarized demands with militarized demands. Consequently, without specifying both initiation and reciprocation we are left unsure of the relationship between regime type and dispute behavior.


Dependent Variable: Dispute Reciprocation


To test the relationship between regime type and crisis bargaining, the reciprocation of a threat, show, or use of force is used.14 This measure captures whether the targeted state has opted to escalate the militarized quarrel, rather than seek a less violent resolution to the dispute. To be sure, the response by the targeted state may not be as militarily aggressive as the initial action. However, the continued militarization of the quarrel most certainly increases the chances of a more severe conflict breaking out. Further, this measure does indicate that a non-violent conflict resolution strategy has been rejected in favor of more militarization.


Exogenous Variables


Regime Type. The primary independent variable of concern is regime type. Are democratic states less likely to reciprocate disputes when the initiating state is a fellow democracy? Data on regime type come from Polity IIId (McLaughlin et al., 1998). The level of democracy indicator is an 11-point index (0-10) based on three relevant aspects of democratic polities: constraints on the chief executive, competitiveness of political participation, and openness of executive recruitment. To capture basic threshold effects, the 11-point coding has been re-coded to a dichotomous democracy-non-democracy (see for example Dixon, 1994).15 Polities with a democracy score of 0-5 are coded as non-democracies, while those with scores of 6 or greater are defined as democratic states.16

Major Power. A distinction is made between major and minor powers. While this is perhaps more an empirical distinction than a theoretical one, evidence does indicate that significant differences exist in the foreign policy behaviors of each. For example, Morgan and Campbell (1991) find that higher political constraints only reduce the propensity to use force for major power states. In fact, they find that higher political constraints tend to increase the war-proneness of minor power states. Therefore, to prevent mis-interpreting the relationship between regime type and reciprocation, a control for major power is included (see Singer and Small, 1982).17

Contiguity. To control for the costs of projecting force or influence abroad, a distance measure is included.18 Boulding (1963), for example, insists that that ability to exercise power effectively depends in part on the location of an opponent. This is because, as Bueno de Mesquita writes (1981: 41), “Combat over a long distance (a) introduces organizational and command problems; (b) threatens military morale; (c) invites domestic dissension; and (d) debilitates soldiers and their equipment.”  Consequently, these costs tend to deter non-contiguous states from escalating disputes (Senese, 1997: 11). The sharing of borders additionally tends to contribute to the presence of historical animosities. Contiguity, in fact, is not only related to war, but persistent confrontations as well (Vasquez, 1993). Since repeated confrontations often exacerbate relations, disputes between neighbors should have a higher probability of reciprocation. To measure geographical proximity, a dichotomous variable is constructed. The coding of this indicator follows the Correlates of War contiguity data set: (1) contiguous by land; (2) contiguous up to 12 miles of water; (3) contiguous up to 24 miles of water (4) contiguous up to 150 miles of water and; (5) contiguous up to 400 miles of water. For this analysis, states that share land borders or are separated by less than 25 miles of water are coded as contiguous. All others are considered non-contiguous.

Preference Similarities. Similarity in alliance portfolios is included to gauge preference alignment (see Thompson and Tucker, 1997, fn. 7). The Kendall Tau-b Regional measure is used, which only calculates the rank-order correlation for the states’ relevant region (see Bueno de Mesquita, 1981 for a full description of regional composition). The variable ranges from –1, indicating completely opposed alliance portfolios, to +1 indicating complete agreement. Presumably, the likelihood of reciprocation increases as alliance similarities decrease.

Controls. Two additional exogenous variables are included to help prevent mis-interpreting the relationship between regime type and dispute reciprocation. The first control accounts for the previous bargaining relationship between the two states. This variable, "peace years", measures the number of years since the two states last experienced a dispute. The second control captures dispute salience. Toset, Gleditsch, and Hegre (2000) suggest using casualties to distinguish different types of disputes. Since 80% of militarized disputes result in no casualties, this salience criterion cannot be used here without losing a large number of cases. Rather, a salience distinction based on actions is utilized. The seizing of ships are classified as low-salience MIDs inasmuch as these cases typically represent the confiscation of fishing trawlers for violating maritime boundaries (see Mitchell and Prins, 1999). To control for salience, then, a dummy variable is included that equals 1 if the dispute involves a naval seizure and 0 otherwise.



Data Analysis


Over 46% of the bilateral disputes included in the MID data file resulted in some level of reciprocation. Interestingly, this rate does not vary drastically over the two centuries of available observations. Averaging by decade from 1816-1992, reciprocation rates remain between 40% and 50% almost without exception (see Figure 1). Reciprocation rates were at their highest during the 1970s and their lowest during the 1880s. The low variation in reciprocation rates stands in marked contrast to the dramatic increase in the frequency of militarized disputes (see Figure 2). While the number of disputes remains fairly steady from 1816 to the early 1900s, a significant jump occurs as a result of the international tensions surrounding World War I. Another significant increase in the number of disputes occurs as a result of World War II. After World War II, however, the annual frequency of militarized disputes raises to a new and higher level. From 1816-1945 the average number of annual disputes was close to 5, although this figure is slightly elevated as a result of the world wars. After 1945, the annual average leaps to nearly 23. Part of the explanation for this dramatic increase has to do with data accessibility. The reporting of events and the spread of information technologies has certainly helped to fill out the datafile on disputes. One might have supposed, however, that reciprocation rates might have also changed over the time period.

Overwhelmingly, as other scholars have noted, democracies are targeted in militarized disputes (see for example, Rousseau et al., 1996; Gleditsch and Hegre, 1997). When a democracy is involved, over 60% of the time it is as a target. Of those disputes, democratic states are targeted over 80% of the time by non-democracies. When democratic leaders initiate hostilities, over 75% of the targets are non-democracies. Non-democracies, in contrast, are not as discriminate when it comes to the regime type of an opponent. In 62% of disputes initiated by non-democracies, another non-democracy was the targeted regime. Non-democratic disputes account for the largest single category of militarized disputes. In fact, over 45% of all bilateral disputes have non-democratic states on both sides. Now mixed regime type disputes do account for a slightly greater percentage (48%), but 56% of those were initiated by non-democracies against democratic states. In less than 7% of the total bilateral disputes from 1816-1992 was a democracy targeted by another democracy.


Regime Type and Reciprocation


While comparing initiation and targeting rates across regime types is interesting, it does not address the issue motivating this paper. Reciprocation is the foreign policy choice of interest. While democracies are more often than not the targets of disputes initiated by non-democracies, reciprocation rates do not vary much as a result of regime differences. Table 1 demonstrates that there is almost no difference between democracies and non-democracies when it comes to rates of dispute reciprocation. In fact, democracies reciprocate 46.8% of disputes while non-democracies reciprocate nearly the same percentage (46.6%). This monadic-level evidence supports hypothesis 1 above.


Figure 1.

Figure 2.


Controlling for the regime types of both the initiating and defending states should clarify the relationship between regime type and reciprocation. It is here, at the dyadic level, that the effects of democracy will be apparent. That is, democratic leaders may respond aggressively to threats made by non-democracies but accommodate without coercion the demands of other democratic states. Looking at Table 2, the rate of reciprocation across regimes does not vary much when the initiating state is a non-democracy. When targeted by autocrats democratic leaders respond similarly to other autocrats. That is, non-democratic disputes are reciprocated 50% of the time, while mixed regime type disputes where the democracy is the target are reciprocated nearly 47% of the time. These results appear to support the basic argument that democracies, to prevent exploitation, adopt the tactics of non-democracies when confronting such states in militarized quarrels.


Table 1. Regime Type and the Reciprocation of Bilateral MIDs, 1816-1992


Targeted State

Reciprocated Dispute



























Table 2. Regime Type, Initiation, and Reciprocation, 1816-1992




  Reciprocated Dispute












































Unsurprisingly, dispute reciprocation is lower when the initiating state is a democracy. A closer look reveals that this lower overall reciprocation rate is not due to democratic-democratic disputes, but because of the low rate of reciprocation of the mixed regime type disputes. In fact, democratic reciprocation rates are higher when the initiating state is a fellow democracy. Of the 114 bilateral disputes between democratic states, 54 or 47.4% were reciprocated. In contrast, only 39.2% of the mixed regime type disputes with the democracy as the initiator were reciprocated. Not only does it appear that democracies are equally as likely, if not more likely, than non-democracies to reciprocate bilateral disputes, but democratic leaders appear more willing to escalate low-level disputes when the opponent is another democracy. In fact, given a democratic initiator, targeted democracies are over 20% more likely to reciprocate than non-democracies.19

Temporal differences interestingly emerge when examining reciprocation. In fact, reciprocation rates were notably higher in the 19th century as compared to the 20th. Particularly with disputes between democracies, reciprocation was high if not certain in the 19th century. Of the 13 democratic-democratic disputes, 10 or 77% experienced some form of escalation.20 In the 20th century, reciprocation appears invariant to the regime type of the initiating state. While disputes initiated by non-democracies have a slightly higher probability of being reciprocated compared to disputes initiated by democracies, the targeted state reciprocates regardless of ideology. This supports Chan’s (1997) conjecture that democratic states should be equally as likely as non-democratic states to respond once a demand has been issued.


Multivariate Analysis


To assess the impact of regime type on reciprocation, a multivariate analysis is conducted to control for other salient exogenous variables. With a dichotomous dependent variable, like dispute reciprocation, a logistic specification is an appropriate model for estimating the posited relationships. Because the coefficients are difficult to interpret directly, the marginal effects of each independent variable on Y are assessed as well. These marginal effects offer one tractable method for evaluating the substantive impact of the explanatory variables.

Table 3 presents the results of four logit models that assess the impact of power, contiguity, regime type, preference similarities, rivalry, and dispute salience on bilateral conflict reciprocation. Model 1 contains both monadic and joint measures of democracy and major power, while models 2 and 3 estimate the monadic and joint measures separately. Model 4 once again estimates the relationships with the joint measures but drops the Tau-Regional variable from the analysis. Due to missing data, 240 cases are dropped with the alliance measure included. Model 4 is estimated without this measure to assess whether these 240 observations alter the substantive findings to any significant degree. 

Table 3. Logit Regression of Dispute Reciprocation, 1816-1992





Model 1


Model 2


Model 3


Model 4

Marginal Effects Model 1

Marginal Effects Model 4

Initiating State Democracy

















Targeted State Democracy

















Joint Democracy

















Initiating State Major Power
















Targeted State Major Power

















Joint Major Powers

















Tau Regional





































Peace Years



















Seizure Cases



















































c2 (p)







Pseudo R2







% Correctly Pred./Null Model











Note: Robust standard errors are in parentheses. The dependent variable is whether or not the dispute was reciprocated. Columns six and seven represent the change in probability Y, after fluctuating each independent variable one standard deviation below the mean to one standard deviation above the mean (except dichotomous variables which are fluctuated from 0 to 1) while holding all other independent variables at their mean values. **p<.01; *p<.05.

What is obvious from each of the models is that regime type has little impact on the decision to reciprocate a low-level quarrel. Neither the monadic nor the joint measures of democracy are significantly related to the decision by the target state to reciprocate a militarized dispute. Interestingly, the signs of the democracy initiator and democracy target variables are in the directions that Fearon’s (1994) model predicts. A democratic target should increase the likelihood of reciprocation (due to audience costs), while a democratic initiator should tend to decrease this likelihood (because the target will recognize that the opponent is facing audience costs). This is exactly what model 1 demonstrates, although the relationship is statistically significant only for targeted democracy in model 2.

While the joint measure of major power is only statistically related to dispute reciprocation in model 3, the monadic measures tell a more complete story. It is in fact major power targets that show no difference in dispute reciprocation from minor power targets. Major power initiators, in contrast, have a strong deterrent effect on dispute reciprocation. When facing a demand made by a major power, target states opt against inflaming the quarrel by making a militarized counter-demand of their own. This is further support for the importance of power in crisis bargaining situations.

Contiguity and peace years also affect dispute reciprocation in the directions hypothesized. First, as the distance between the two states involved in the dispute increases, the probability of reciprocation decreases. The impact of contiguity on dispute reciprocation is large. The probability changes by over 50% (.374 to .568) going from non-contiguous states to contiguous ones. Thus, distance does tend to restrict the projection of force. Second, the peace-years variable indicates that longer periods of peace between two states results in less escalatory foreign policy strategies. A change of one standard deviation (18.9 years) decreases the probability of reciprocation by nearly 15% (.508 to .437).

Perhaps the most significant factor influencing dispute reciprocation is whether the dispute involved a naval seizure. As one might expect, these low-salience disputes rarely evoke retaliatory militarized demands. The confiscation of fishing trawlers for boundary violations, while formally a use of force, only infrequently lead state leaders to escalate interstate quarrels. Indeed, non-seizure disputes have slightly over a 50% chance of reciprocation. Seizure disputes, in comparison, have less than a 30% chance.

Interestingly, preference similarities have little influence on dispute reciprocation. Although the Tau-Regional measure, as expected, is inversely related to the decision by the target state to escalate (that is, preference similarities tend to decrease the probability of conflict reciprocation), its substantive impact is quite small. It appears that while alliance agreement may have an important impact on the outbreak of war, similarly aligned states demonstrate little reluctance to reciprocate militarized disputes.

Immature Democracies


Recent empirical evidence suggests that fully institutionalized democracies tend to behave quite differently than their under-developed counterparts. In fact, Mansfield and Snyder (1995) find that states in transition from autocracy to democracy have a much higher propensity to engage in warfare than more institutionally stable polities.21 They suggest that elites in these emerging democratic polities have yet to be fully constrained by democratic norms of compromise and conflict avoidance or institutional impediments to the use of military force. As a consequence, these immature regimes face a precarious set of circumstances where conflict escalation helps divert attention away from domestic turmoil and arouses nationalist sentiment, which can be exploited by the regime in power (see Snyder, 2000). In order to assess whether the lack of findings for the regime type variables is being driven by the foreign policy behavior of emerging democratic states, I control more precisely for level of democracy. That is, I distinguish between those states with democracy scores of 0-5, those with scores of 6-9, and those mature democratic states with democracy scores of 10. The results I find strongly suggest that fully institutionalized democracies and emerging democracies behave very differently when it comes to crisis bargaining.

Table 4 demonstrates that young or emerging democratic states have the greatest propensity for issuing militarized counter-demands. Bilateral disputes involving two immature democratic states are particularly conflictual, with nearly 70% of the disputes experiencing reciprocation. Emerging democracies appear nearly as belligerent when facing non-democracies as well. In over 60% of the disputes where non-democratic states targeted emerging democracies, counter-demands were issued. This is considerably higher than disputes between non-democratic states (49.9%) and mixed regime type disputes involving mature democracies and non-democracies (39.3%). Fully institutionalized democracies, in contrast, actually demonstrate the lowest rate of dispute reciprocation. Only a third of the time will mature democratic states retaliate, and less than a third when the initiating state is another mature democracy.

The results in Table 4 suggest that states react differently to foreign policy crises. In particular, fully institutionalized democracies rarely react to militarized demands with militarized demands of their own. However, this does not mean that these mature democracies are not distinguishing the regime type of their opponent. Indeed, the rate of reciprocation for mature democracies increases as the democracy level of the initiating state decreases. These results also suggest that the institutionalization and stability of polities may contribute to more efficient signaling and thus less escalatory decision-making. The most conflictual pairings of polities appear to be emerging democracies where institutions remain in flux. It appears, then, that the initial results on regime type were heavily affected by the foreign policy behavior of emerging democratic states.



Table 4. Regime Type, Initiation, and Reciprocation, 1816-1992




  Reciprocated Dispute













Immature Democracy






Mature Democracy











Immature Democracy






Immature Democracy

Immature Democracy





Immature Democracy

Mature Democracy











Mature Democracy






Mature Democracy

Immature Democracy





Mature Democracy

Mature Democracy















The arguments of Fearon (1994) and Chan (1997) stand in marked contrast to the empirical evidence showing democratic polities to use less coercive foreign policy strategies (Leng, 1993) and to resolve quarrels non-violently (Dixon, 1993; Raymond, 1994). To evaluate these divergent perspectives on democratic foreign policy decision-making, I examined the relationship between regime type and dispute reciprocation. The evidence uncovered here supports the findings of Leng, Dixon, and Raymond. Not only are fully institutionalized democracies unlikely to reciprocate militarized disputes, but they also appear to condition their foreign policy behavior on the regime type of the initiating state. The rate of mature democratic reciprocation increases steadily as the democracy level of the initiating state decreases. In fact, the rate of mature democratic dispute reciprocation increases by 35% as the regime type of the initiating state changes from a mature democracy to a non-democracy.


Emerging democracies, in contrast, display very different dispute behavior. Immature democratic polities frequently reciprocate disputes and are almost certain to issue counter demands when the initiating state is similarly under-developed. These results support previous research that finds young democratic polities to be more conflict prone than their more fully institutionalized relatives. While these results do not speak to war proneness, they do suggest that emerging democratic states are more willing to respond to militarized demands with militarized demands of their own. This type of power politics behavior is likely to be associated with dispute escalation.

In addition to regime type, the logit results indicate that contiguity, dispute salience, and years in peace all have important effects on dispute reciprocation. These results suggest that the strategic environment and the bargaining relationship certainly influence dispute reciprocation. But, regime type similarities also play an important role in crisis bargaining, at least for mature democratic states.





  • The Clinton administration used a similar line of argument to defend normalizing trade with the People's Republic of China. Increased economic openness on the part of China would eventually lead to the emergence of democratic political institutions and values, and since democracies don't fight, this would be a prescription for peace.
  •  According to Maoz and Russett (1993), when it comes to relations between democratic leaders and autocrats this expectation does not exist. In fact, to prevent exploitation democratic political elites approach authoritarian regimes with deep suspicion. This is because as Doyle (1986: 1186) writes, "nonliberal governments are in a state of aggression with their own people" creating a natural presumption of enmity on the part of democratic leaders.
  • The “political penalty,” as Chan (1997: 80) writes, is particularly severe for democratic leaders who engage in armed hostilities (Bueno de Mesquita and Siverson, 1995; Bueno de Mesquita et al., 1999). Even without policy failure, democratic political leaders often suffer electorally as a result of resorting to military force (see Bueno de Mesquita and Lalman, 1990).
  • Gleditsch and Ward (1997) additionally found evidence tying institutional constraints to pacific foreign policy behavior. In their disaggregation of the Polity scales, Gleditsch and Ward discovered that the executive constraints component of the democracy score had the largest influence on a state’s final democracy ranking. Given the lack of violent conflict between democratic nations, Gleditsch and Ward’s findings tend to support the argument that constraints on the chief executive keep in check foreign policy adventurism, at least when it comes to relations with similar regimes (see Benoit, 1996 for monadic-level evidence).
  • According to Lee Hamilton, the former ranking minority member on the House International Relations Committee, one of the president’s responsibilities is to help reconcile the differences that exist between the professional foreign policy decision-makers and a less activist American public (The Economist, October 30th, 1993: 24).
  • While not denying the importance of these analytical distinctions (norms and structures), Owen (1994) insists that liberal perceptions necessarily precede the impact of institutional structures and cultural norms. That is, “the liberal commitment to individual freedom,” as Owen (1994: 124) writes, “gives rise to foreign policy ideology and governmental institutions that work together to produce democratic peace.” For Owen, perceptions (or misperceptions) help explain anomalous cases of dyadic democratic conflict. If political elites fail to recognize the liberal ideology grounding a regime, then the presumption of amity disintegrates. For example, Owen (1994: 108-109) insists that the War of 1812 between the U.S. and Great Britain (arguably a war between two democracies) should not be considered an exception to the rule inasmuch as most Americans refused to recognize the U.K. as a liberal state. Interestingly, Huntley’s  (1996) view seems to contrast with Owen’s (1994). For Huntley, the perception of trustworthiness stems from democratic political norms and institutions and not necessarily from a “liberal commitment to individual freedom,” as Owen (1994: 124) writes.
  • There also is the problem of endogeneity broached by some scholars (see Gates et al., 1996; Thompson, 1996).
  • Lake (1992) finds that democratic states win wars, although it remains unclear whether this result stems from morale and leadership differences or power differences. Reiter and Stam ( 1998) similarly find that democratic states win the wars they choose to initiate. Both careful selection of weak opponents and military leadership and morale issues most likely contribute to this finding. If these results are robust, then rationally democratic states should not be targeted by autocracies given the typical outcome, especially since Reiter and Stam (1998) also observe that democratic targets are more likely than non-democratic targets to win wars.
  • At times, the need to maintain domestic political support may force democratic leaders to adopt aggressive or hardline policy positions. Naturally, how democratic politics affects foreign policy decision-making has profound normative implications for the U.S. political process. As Lindsay et al. (1992: 5) wrote, "The defense of the imperial presidency rests on the claim that presidents are more rational and more immune to the tide of public opinion than is Congress. The claim of superior presidential decision making crumbles, however, if presidents use foreign policy to serve their own political ends."
  • Hewitt and Wilkenfeled (1996) also found that democratic crises are less violent than their non-democratic counterparts.
  • Monadic-level, that is state-centered, conflict behavior may require further empirical research. Most of the evidence collected to date shows democratic regimes to be equally as conflict-prone as non-democratic ones. However, if democracies demonstrate a propensity to aid like regimes in peril, the monadic-level results may be heavily affected by joining behavior.
  • Waltz (1967) argued that institutional differences accounted for some of the variation observed in the foreign policies of Great Britain and the United States.
  • To test whether democratic dispute reciprocation is invariant to the regime type of the initiating state, the Militarize Interstate Dispute (MID) data set is utilized. The MIDs data set provides a large number of events short of war where the initiating actor is clearly identified. For a dispute to be included in the data file, the threat, display or use of military force must have occurred. Jones, Bremer, and Singer (1996: 166), define MIDs as “confrontations that [lead] politicians to invest energy, attention, resources, and credibility in an effort to thwart, resist, intimidate, discredit, or damage those representing the other side.” Furthermore, the coded acts “must be explicit, overt, nonaccidental, and government sanctioned” (Gochman and Maoz, 1984: 586). The basic unit of observation in the empirical analyses is the interstate dispute. The updated version of the MID data file (version 2.10, 1996) contains 2034 total disputes from 1816-1992. For the analyses, only bilateral MIDs are included. Thus, any dispute that did not begin and end as a one-on-one confrontation is removed from the analysis. This is done to prevent any confounding effects from dispute joining. The decision-making process of third parties may be quite different than that of originating states. Therefore, to isolate the effects of regime type on reciprocation, only bilateral MIDs are evaluated. This decision deletes 329 disputes.
  • Drawing on Leeds and Davis (1997), initiator and target are operationalized using the ‘Side-A’ and ‘Originator’ codings in the MID datafile. An initiating state is one that is involved on the first day of hostilities and is considered the state which first militarized the dispute. A defending state (target) is involved on the first day, but is not on the side that first militarized the quarrel. Of the 1705 bilateral disputes, 46.6% (795) experienced some level of reciprocation. This leaves 910 other disputes that were resolved without a violent dispute response.
  • Partell and Palmer (1999: 396) use 5 as the cutpoint.
  • The Polity data file also specifies whether regimes are in a period of transition, experiencing political collapse, or are interrupted such as by military intervention. These cases were coded as non-democracies.
  • Major powers include: U.S. 1899-1992; U.K. 1816-1992; France 1816-1940 and 1945-1992; Germany 1816-1918 and 1925-1945; Austria-Hungary 1816-1918; Italy 1860-1943; Russia 1816-1917 and 1922-1992; China 1950-1992; Japan 1895-1945.
  • Interestingly, Gochman (1990) argues that technological advances should make contiguity less important in future militarized confrontations. Projecting force, in other words, has become easier and less costly. He finds, however, that contiguity is more strongly related in the 20th century than in the 19th.
  • To test the effects of audience costs on dispute reciprocation, I also examine constraints on the chief executive (see for example, Partell and Palmer, 1999). The scale of this Polity III variable ranges from 1 (unlimited authority) to 7 (executive parity or subordination). Similar to Partell and Palmer (1999: 396), a state with a score of 3 or greater is coded as constrained, while anything below is considered not constrained. I find that using either the executive constraints dummy or the democracy dummy does not have a large impact on reciprocation rates. Unconstrained dyads are the most likely to experience reciprocated disputes. This evidence does not support Fearon’s (1994) conjecture regarding escalation and audience costs. It would seem that mixed dyads would have the greatest potential to escalate. Once a dispute has been initiated, a highly constrained state would be compelled to escalate, perhaps even more compelled if the initiating state is not similarly constrained.
  • Of these 13 democratic-democratic disputes in the 19th century, the United States was involved in 10. Seven disputes involved Great Britain and three involved Spain. The three remaining disputes involved the U.K. and France in 1893, 1896, and 1898.
  • Thompson and Tucker (1997) find that Mansfield and Snyder’s (1995) results are sensitive to the methodological choices these authors made.





Albright, Madeleine K. 1998. "The Testing of American Foreign Policy." Foreign Affairs, Vol. 77, No. 6, pp. 50-64.

Beckman, Peter R., Paul W. Crumlish, Michael N. Dobkowski, and Steven P. Lee. 2000. The Nuclear Predicament: Nuclear Weapons in the Twenty-First Century. Upper Saddle River, N.J.: Prentice Hall.

Berger, Sandy. 2000. "A Foreign Policy for the Global Age." Foreign Affairs, Vol. 79, No. 6, pp. 22-39.

Benoit, Kenneth. 1996. "Democracies Really are More Pacific (in general): Reexamining Regime Type and War Involvement." Journal of Conflict Resolution, Vol. 40, No. 4, pp. 636-58.

Bercovitch, Jacob, ed. 1996. Resolving International Conflicts: The Theory and Practice of Mediation. Boulder: Lynne Rienner Publishers.

Boulding, Kenneth. 1963. Conflict and Defense: A General Theory. New York: Harper and Row.

Bueno de Mesquita, Bruce. 1981. The War Trap. New Haven: Yale University Press.

Bueno de Mesquita, Bruce and David Lalman. 1992. War and Reason: Domestic and International Imperatives. New Haven: Yale University Press.

Bueno de Mesquita, Bruce and David Lalman. 1990. "Domestic Opposition and Foreign War." American Political Science Review, Vol. 84. No. 3, pp. 747-765.

Bueno de Mesquita, Bruce, James Morrow, Randolph Siverson, and Alastair Smith. 1999. "An Institutional Explanation of the Democratic Peace." American Political Science Review, Vol. 93, No. 4, pp. 791-808.

Bueno de Mesquita, Bruce and Randolph M. Siverson. 1995. "War and the Survival of Political Leaders: A Comparative Study of Regime Types and Political Accountability." American Political Science Review, Vol. 89, No. 4, pp. 841-53.

Chan, Steve. 1997. "In Search of Democratic Peace: Problems and Promise." Mershon International Studies Review, Vol. 41, No. 1, pp. 59-91.

Dixon, William J. 1996. "Third-Party Techniques for Preventing Conflict Escalation and Promoting Peaceful Settlement." International Organization, Vol. 50, No. 4, pp. 653-681.

Dixon, William J., 1994. "Democracy and the Peaceful Settlement of International Conflict." American Political Science Review, Vol. 88, No. 1, pp. 14-32.

Dixon, William J. 1993. "Democracy and the Management of International Conflict." Journal of Conflict Resolution, Vol. 37, No. 1, pp. 42-68.

Doyle, Michael. 1986. "Liberalism and World Politics." American Political Science Review, Vol. 80, No. 4, pp. 1151-1161.

The Economist, October 30, 1993, p. 24.

Elman, Miriam Fendius, ed. 1997. Paths to Peace: Is Democracy the Answer? Cambridge: The MIT Press.

Farber, Henry S and Joanne Gowa. 1995. "Polities and Peace." International Security, Vol. 20, No. 2, pp. 123-146.

Fearon, James. 1994. "Domestic Political Audiences and the Escalation of International Disputes." American Political Science Review, Vol. 88, No. 3, pp. 577-592.

Gates, Scott, Torbjorn L. Knutsen, and Jonathon W. Moses. 1996. “Democracy and Peace: A More Skeptical View.” Journal of Peace Research, Vol. 33, No. 1, pp. 1-10.

Gartzke, Erik. 2001. "Two-Level Liberalism and the Logic of Interstate Contests: Determinants of the Democratic Peace." Paper presented at the Annual Meeting of the International Studies Association, 20-24 February, Chicago, IL.

Gleditsch, Kristian S. and Michael D. Ward. 1997. "Double-Take: A Re-examination of Democracy and Autocracy in Modern Politics." Journal of Conflict Resolution, Vol. 41, No. 3, pp. 361-383.

Gleditsch, Nils Petter and Håvard Hegre. 1997. "Peace and Democracy: Three Levels of Analysis." Journal of Conflict Resolution, Vol. 41, No. 2, pp. 283-310.

Gochman, Charles. 1990. "The Geography of Conflict: Militarized Interstate Disputes Since 1816." Paper presented at the Annual Meeting of the International Studies Association, 29 August-2 September, Washington, D.C.

Gochman, Charles and Zeev Maoz.  1984.  "Militarized International Disputes, 1816-1976: Procedures, Patterns, Insights." Journal of Conflict Resolution , Vol. 28, No. 4, pp. 585-616.

Gowa, Joanne. 1995. "Democratic States and International Disputes." International Organization, Vol. 49, No. 3, pp. 511-522.

Hagan, Joe D. 1994. "Domestic Political Systems and War Proneness." International Studies Quarterly, Vol. 38, No. 2, pp. 183-208.

Henderson, Errol A. 2000. "Paradise Cost: Is the Democratic Peace and Imperialist Peace?" Paper presented at the Annual Meeting of the American Political Science Association, 30 August-3 September, Washington, D.C.

Huntley, Wade L. 1996. “Kant's Third Image: Systemic Sources of the Liberal Peace.” International Studies Quarterly, Vol. 40, No. 1, pp. 45-77.

Jones, Daniel M., Stuart A. Bremer, and J. David Singer. 1996. "Militarized Interstate Disputes, 1816-1992: Rationale, Coding Rules, and Empirical Patterns." Conflict Management and Peace Science, Vol. 15, No. 2, pp. 163-213.

Lake, David A. 1992. "Powerful Pacifists: Democratic States and War." American Political Science Review, Vol. 86, No. 1, pp. 24-38.

Layne, Christopher. 1994. "Kant or Cant: The Myth of the Democratic Peace." International Security, Vol. 19, No. 2, pp. 5-49.

Leeds, Brett Ashley and David R. Davis. 1997. "Domestic Political Vulnerability and International Disputes." Journal of Conflict Resolution, Vol. 41, No. 6, pp. 814-834.

Leng, Russel J. 1993. "Reciprocating Influence Strategies in Interstate Crisis Bargaining." Journal of Conflict Resolution, Vol. 37, No. 1, pp. 3-41.

Lindsay, James M., Lois W. Sayrs, and Wayne P. Steger. 1992. "The Determinants of Presidential Foreign Policy Choice." American Politics Quarterly, Vol. 20, No. 1, pp. 3-25.

Mansfield, Edward, and Jack Snyder. 1995. "Democratization and the Danger of War." International Security, Vol. 20, No. 1, pp. 5-38.

Maoz, Zeev and Bruce Russett.  1993.  "Normative and Structural Causes of the Democratic Peace, 1946-1986." American Political Science Review, Vol. 87, No. 3, pp. 624-638.

Maynes, Charles William. 2000. "U.S. Role in the World: What are the Choices." Great Decisions, pp. 9-18.

McLaughlin, Sara, Scott Gates, Håvard Hegre, Ranveig Gissinger, and Nils Petter Gleditsch. 1998. "Timing the Changes in Political Structures: A New Polity Database." Journal of Conflict Resolution, Vol. 42, No. 2, pp. 231-242.

Mitchell, Sara McLaughlin and Brandon C. Prins. 1999. "Beyond Territorial Contiguity: Issues at Stake in Democratic Militarized Interstate Disputes." International Studies Quarterly, Vol. 43, No. 1, pp. 169-183.

Morgan, T. Clifton, and Kenneth N. Bickers. 1992. "Domestic Discontent and the External Use of Force." Journal of Conflict Resolution, Vol. 36, No. 1, pp. 25-52.

Morgan, T. Clifton and Sally Campbell. 1991. "Domestic Structure, Decisional Constraints, and War." Journal of Conflict Resolution, Vol. 35, No. 2, pp. 187-211.

Oren, Ido. 1995. "The Subjectivity of the ‘Democratic Peace’: Changing U.S. Perceptions of Imperial Germany." International Security, Vol. 20, No. 2, pp. 147-184.

Ostrom, Charles W. and Brian Job. 1986. "The President and the Political Use of Force." American Political Science Review, Vol. 80, No. 2, pp. 541-566.

Owen, John M. 1994. "How Liberalism Produces Democratic Peace." International Security, Vol. 19, No. 2, pp. 87-125.

Palmer, Glenn and Patrick M. Regan. 1999. "Power, Domestic Structure, and Parties: Untangling Entangling Democracies." Paper presented at the Annual Meeting of the Nordic Political Science Congress, 19-21 August, Uppsala, Sweden.

Partell, Peter J. and Glenn Palmer. 1999. "Audience Costs and Interstate Crises: An Empirical Assessment of Fearon’s Model of Dispute Outcomes." International Studies Quarterly, Vol. 43, No. 2, pp. 389-405.

Prins, Brandon C. 1998. "Opportunities, Domestic Politics, and the Decision to Intervene. U.S. Third Party Involvement in Interstate Crises, 1918-1994." Paper presented at the Annual Meeting of the Peace Science Society (International), 16-18 October, Rutgers, N.J.

Prins, Brandon C. and Christopher Sprecher. 1999. "Institutional Constraints, Political Opposition, and the Escalation of Interstate Disputes: Evidence From Parliamentary Systems." Journal of Peace Research , Vol. 36, No. 2, pp. 271-287.

Raknerud, Arvid and Håvard Hegre. 1997. "The Hazard of War: Reassessing the Evidence for the Democratic Peace." Journal of Peace Research, Vol. 34, No. 4, pp. 385-404.

Ray, James Lee. 1999. "Comparing Conflict Propensities of Autocracies and Democracies: A Directed Dyadic Approach." Paper presented at the American Political Science Association, 2-5 September, Atlanta, GA.

Raymond, Gregory A. 1994. "Democracies, Disputes, and Third-Party Intermediaries." Journal of Conflict Resolution, Vol. 38, No. 1, pp. 24-42.

Reiter, Dan and Allan C. Stam. 1998. "Democracy, War Initiation, and Victory." American Political Science Review, Vol. 92, No. 2, pp. 377-89.

Rousseau, David L., Christopher Gelpi, Dan Reiter, and Paul K. Huth. 1996. "Assessing the Dyadic Nature of the Democratic Peace, 1918-1988." American Political Science Review, Vol. 90, No. 3, pp. 512-33.

Russett, Bruce. 1993. Grasping the Democratic Peace: Principles for a Post-Cold War World. Princeton: Princeton University Press.

Russett, Bruce. 1990. Controlling the Sword: The Democratic Governance of National Security. Cambridge: Harvard University Press.

Russett, Bruce and Harvey Starr. 2000. "From Democratic Peace to Kantian Peace: Democracy and Conflict in the International System." In Manus I. Midlarsky, ed., Handbook of War Studies II. Ann Arbor: University of Michigan Press.

Schjølset, Anita. 1996. "Are Some Democracies More Peaceful Than Others?" Paper prepared for the Annual Meeting of International Studies Association, 17-21 April, San Diego, CA.

Schultz, Kenneth A. 2001. Democracy and Coercive Diplomacy. Cambridge: Cambridge University Press.

Schultz, Kenneth A. 1998. "Domestic Opposition and Signaling in International Crises." American Political Science Review, Vol. 92, No. 4, pp. 829-44.

Senese, Paul D. 1997. "Between Dispute and War: The Effect of Joint Democracy on Interstate Conflict Escalation."  Journal of Politics, Vol. 59, No. 1, pp. 1-27.

Singer, J. David and Melvin Small.1982. Resort to Arms. Beverly Hills: Sage Publications.

Smith, Alastair. 1996. "Diversionary Foreign Policy in Democratic Systems." International Studies Quarterly, Vol. 40, No. 1, pp. 133-153.

Snyder, Jack, 2000. From Voting to Violence: Democratization and Nationalist Conflict. New York: Norton.

Spiro, David E. 1994. "The Insignificance of the Liberal Peace." International Security, Vol. 19, No. 2, pp. 50-86.

Talbot, Strobe. 1996. "Democracy and the National Interest: Idealpolitik as Realpolitik." Foreign Affairs, Vol. 75, No. 1, pp. 47-63.

Toset, Hans Petter Wollebæk, Nils Petter Gleditsch, and Håvard Hegre, 2000. "Shared Rivers and Interstate Conflict." Political Geography, Vol. 19, No. 6, pp. 971-996.

Thompson, William R. 1996. "Democracy and Peace: Putting the Cart Before the Horse?" International Organization, Vol. 50, No. 1, pp. 141-174.

Thompson, William R. and Richard Tucker. 1997. "A Tale of Two Democratic Peace Critiques." Journal of Conflict Resolution, Vol. 41, No. 3, pp. 428-454.

Vasquez, John.  1993.  The War Puzzle.  Cambridge: Cambridge University Press.

Waltz, Kenneth N. 1967. Foreign Policy and Democratic Politics: The American and British Experience. Boston: Little, Brown and Company.

Werner, Suzanne and Douglas Lemke. 1997. "Opposites Do Not Attract: The Impact of Domestic Institutions, Power, and Prior Commitments on Alignment Choices." International Studies Quarterly, Vol. 41, No. 3, pp. 529-546.