The Dearth of Jointly Dyadic Democratic Interventions

Transcription

1 International Studies Quarterly ~2002! 46, The Dearth of Jointly Dyadic Democratic Interventions John A. Tures LaGrange College The Democratic Peace Proposition, which states that no two democracies have ever gone to war with each other, has been questioned by scholars who claim that such pacific behavior among free states does not apply to lower forms of conflict. In particular, Kegley and Hermann contend that democracies intervene in the affairs of other liberal states via overt military acts or covert machinations. In many cases, they argue that dyadic democratic interventions ~DDIs! occur more frequently than would be expected given the number of jointly democratic dyads in the international system. I examine their research design and suggest changes to their concepts of states, interventions, and regime type, as well as their sample size and definition of dyads in the international system. I implement these changes and retest such arguments on a sample of interventions from 1945 to I find 11 cases where a democracy intervenes against another democracy, but these cases are rare in comparison to interventions conducted by democratic and0or autocratic states in undemocratic states, or by autocratic states against democratic states. Furthermore, these DDIs are less likely to occur than the presence of democracy in the international system would suggest. Critics of the Democratic Peace have argued that the observed peace between democracies does not apply to lower forms of conflict. In particular, Kegley and Hermann ~1995a, 1995b, 1996, 1997! and Hermann and Kegley ~1996, 1998! contend that democracies have used intervention against their fellow liberal states, and are doing so with increasing frequency. Yet their studies show room for improvement in conceptualizing regime type, intervention, and states. Their studies that use Polity data code states as democratic with a measure that treats governments with highly concentrated amounts of power as democratic. These studies also hold their measure of power concentration constant between the years 1987 and 1991, a time when widespread fluctuations in regime power occurred. Kegley and Hermann s analyses also contain questionable cases, including invited interventions, nonstate entities, and occupied states. To address these problems, I utilize a measure of democracy developed by Thompson and Tucker ~1997! that does not code countries with higher levels of governmental power concentration as democratic. I use Tillema s ~1997! intervention dataset, omitting cases of invited interventions, occupied states, and nonstate entities. I also expand their temporal domain to include the years 1945 Author s note: The author would like to thank Patrick James, Kirsten Rafferty, and four anonymous reviewers for their comments and suggestions International Studies Association. Published by Blackwell Publishing, 350 Main Street, Malden, MA 02148, USA, and 108 Cowley Road, Oxford OX4 1JF, UK.

2 580 The Dearth of Jointly Dyadic Democratic Interventions through While I do find support for Kegley and Hermann s contentions that democracies have intervened against other liberal states, such events are rare, especially in comparison with both the actions of democracies against nondemocracies and the behavior of their autocratic counterparts toward liberal and illiberal states. Statistical tests show democratic dyadic interventions ~DDIs! occurring at rates lower than expected, given the profusion of democracy during the Cold War years. Kegley and Hermann s Research and Findings In challenging the Democratic Peace Proposition, Kegley and Hermann claim that such pacific behavior among liberal states does not apply to lower forms of conflict. In fact, they contend that democracies frequently intervene in the internal affairs of their liberal counterparts ~Kegley and Hermann, 1995a, 1996, 1997!, a troubling finding for the Democratic Peace literature, given that interventionism is, after all, a prelude to war ~Kegley and Hermann, 1996:319!. These results hold across a variety of measures of democracy and intervention from 1974 to Kegley and Hermann ~1995a:1! use Freedom House scores and the Pearson and Baumann intervention dataset to test the connection between regime type and lower forms of conflict. They find 15 instances where free democratic states moved their regular troops into the territory of other free states and 32 instances of free states intervening in partly free states between 1974 and In another study, Kegley and Hermann ~1996! test the link between democracy and intervention using Tillema s intervention data and both Freedom House and Polity data to represent regime type. When using the Polity measure, they found that democratic states were willing to use intervention against other democracies 1.43 times more often than expected; 22 percent of all democratic interventions targeted fellow democracies ~Kegley and Hermann, 1996:318!. Though the authors claim that using the Freedom House measure provides some support for the Democratic Peace Proposition, they noted that there were six cases of free states intervening in partly free states. Furthermore, partly free states intervened against free states almost twice as often as would be expected. Another Kegley and Hermann study ~1997! uses Freedom House scores and Tillema s data. This analysis finds 51 instances where democracies used intervention against other democratic states: two free states against fellow free states, two cases of free states intervening against partly free states, 14 cases of partly free states intervening in the affairs of free states, and 24 cases of partly free states intervening against other partly free states. Hermann and Kegley ~1996! find that while democracies are less likely to intervene in the internal affairs of other democracies than would be expected by chance ~using Polity data and Tillema s intervention dataset!, 12 dyadic democratic interventions still occurred between 1975 and However, their tests of arguments linking regime type and intervention could be improved. Their conceptualization of states, democracy, and interventions produced a sample with questionable cases of dyadic democratic intervention ~DDIs!. In the next section I identify how I operationalize states, democracy, and conflict. I also discuss how my measures represent a departure from previous studies that test the application of the Democratic Peace Proposition to lower 1 Hermann and Kegley ~1998! do study the entire Cold War period ~ !, but analyze only American interventions during this time frame.

3 John A. Tures 581 Table 1. Differences Between My Study and Kegley and Hermann s Study Concept Kegley and Hermann Tures States Democracy Include Nonstate Entities and Occupied States Include Freedom House, Polity II and III Datasets; the Polity Datasets Use Maoz & Russett s ~1992! Democracy Measure Includes Polity Criteria for Statehood, Excludes Nonstate Entities and Occupied States Includes Polity III Dataset; Uses the Thompson & Tucker ~1997! Measure of Democracy Interventions Include Invited Interventions Excludes Invited Interventions Dyads Analyze All Dyads in the International System Analyzes All Directed, Relevant Dyads Temporal Domain forms of conflict. The differences between my study and Kegley and Hermann s analyses are documented in Table 1. Defining Concepts We need to develop a rigorous methodology for defining concepts and selecting cases in order to assess the democratic peace and its applicability to lower forms of conflict. To accomplish this, we should provide criteria for what qualifies as a state, what represents a democracy, and which cases represent intervention. States Those countries that are part of the Polity dataset are included in the analysis. Interventions involving nonstate entities are dropped from the sample, as are cases where intervention initiators or targets have missing regime scores due to regime transitions or a preexisting foreign occupation. This is important, given that the intervention definition employed by Kegley and Hermann ~1997! includes sovereignty as a key condition. 2 To be included in the Polity dataset, a country must be independent and have a population exceeding 500,000 ~Marshall and Jaggers, 2001!. Given that such rules are less restrictive than those employed by other state measures such as the Correlates of War dataset ~Thompson and Tucker, 1997:448!, fewer countries are excluded from my analysis. However, Namibia, East Timor, the Panama Canal Zone, Trieste, and the Falkland Islands are not considered sovereign states between the years 1945 and Democracy In addition to determining which regime type dataset should be employed, we also need to select a measure of democracy from that dataset. Kegley and Her- 2 In their test of links between intervention and regime type, Kegley and Hermann ~1997! include the Falkland Islands, pre-1991 East Timor, and pre-1991 Namibia as states. They also code countries occupied by a foreign power ~whose independent status can be called into question!, such as Kuwait, as states. Governments undergoing a transition in ruling authority structures also find their way into Kegley and Hermann s case selection. Rather than omit these cases, the authors have chosen to include them in these studies, often as states displaying some liberal characteristics. Including such entities also provides conceptual problems as well. Kegley and Hermann ~1997:87! use Joyner s ~1992! definition for intervention, which claims, Like large scale war, intervention ~a! includes the use of force, ~b! is intrinsically hostile in intent, ~c! frequently entails the loss of soldiers lives, and ~d! violates the rule in international law prohibiting the use of force to alter the internal political institutions of other sovereign states. Do the Falkland Islands, pre-1990s East Timor, and Namibia qualify as sovereign states?

4 582 The Dearth of Jointly Dyadic Democratic Interventions mann ~1996; also Hermann and Kegley, 1996! employ Maoz and Russett s ~1992! formula for coding countries as democracies or autocracies, which is: Regime Score Autocracy Score# Power Concentration Though widely used, the Maoz and Russett measure has some problems. Scholars have faulted the Maoz and Russett measure for its absence of theoretical rationale ~Oneal and Russett, 1997:274! and nonmonotonicity in dyadic measures ~Ray, 1995!. But the most serious allegations come from Thompson and Tucker ~1997:448!, who contend that the Maoz and Russett measure heavily weights the power concentration variable. The Polity II Codebook describes the power concentration measure as based on the institutional characteristics of the polity. Institutional power is least in polities where political competition is divisive and unregulated, where political authority is dispersed among different individuals and institutions, and where peripheral regions have some autonomy from central authority. The opposite traits institutional control of competition and executive recruitment, concentration of national power in a unitary state headed by a strong executive signify a high degree of institutional power ~Gurr, Jaggers, and Moore, 1989!. Countries that have a greater concentration of power in the government are more likely to be coded as democratic using the Maoz and Russett measure, even though states with such institutions ~constrained competition and recruitment! are often considered less democratic. 3 Thompson and Tucker recommend the revised equation: Regime AUTOC! 11#0@log10~PCON 1! 1# I measure democracies using Thompson and Tucker s formula for regime type. Thompson and Tucker ~1997:448! code countries with scores between 10 and 21 as democracies. Those countries receiving scores below 10 are labeled nondemocracies. Thompson and Tucker derived the thresholds using a rigorous analysis of countries. 4 In addition to employing the Maoz and Russett ~1992! measure of democracy, Hermann and Kegley ~1996:442; also Kegley and Hermann, 1996:313! lump information from Polity II ~ ! and Polity III ~ ! together, using the most recent scores for Democracy and Autocracy. But since Polity III data were unavailable for the power concentration measure as well as the variables that constitute the measure at the time, the authors assumed such a value remain constant for five years of tumultuous regime change ~ !. Though Polity III and subsequent variations of the dataset do not include the power concentration scores, the Polity II Codebook ~Gurr, Jaggers, and Moore, 1989! provides the formula for the power concentration variable. The variables 3 Thompson and Tucker ~1997:447! provide an example. Consider... the case of state A and state B. They have democracy and autocracy scores of 10 ~the highest possible rating! and 0 ~the lowest possible score!, respectively. Power in state A is highly concentrated ~9 on a scale of 1 to 10!. Power in state B is moderately concentrated ~2!. State A is scored as a 90, whereas state B is scored as a 20, even though, or because, state A s power is more highly concentrated. The threshold for democracy in the Maoz and Russett ~1993! index is 29. Thus, state A is labeled democratic and state B fails to qualify as a democracy. 4 Thompson and Tucker ~1997:448! described their process of arriving at the thresholds. We sought a democratic threshold that combined a high democratic score and a low autocratic score with, at most, moderate concentration. For autocracy, we looked for break points that combined high autocratic and low democratic scores with at least moderate or higher concentration. After arraying all the scores every 20 years or so, a democratic threshold of 10 and an autocratic0anocratic threshold of 5 appeared to fit these criteria most consistently between 1800 and After specifying our thresholds, our categorical outcomes appear intuitively satisfactory.

5 John A. Tures 583 Table 2. Power Concentration Measure from the Polity II Codebook Variable and Authority Coding Scale Weight Regulation of Participation Regulated 2 Restricted Regulation of Executive Recruitment Regulated Competitiveness of Executive Recruitment Selection Election Constraints on the Chief Executive None 3 2 ~Intermediate Category! 2 Slight to Moderate Limits Monocratism Pure Individual Executive Centralization Unitary State 2 that make up the power concentration measure are available in Polity III. I recode the power concentration scores using Polity III data and the Polity II formula ~see Table 2! so that the data are consistent from 1945 through Interventions I utilize a dataset of foreign overt military interventions constructed by Tillema ~1997! for the years 1945 to He defines intervention as a distinct category of militarized international behavior that ~1! involves the use of force, ~2! results in the loss of soldiers lives, and ~3! is described by a target as a hostile act ~Tillema, 1989:419!. Such conflicts exclude less blatant forms of international interference, such as covert operations, shows of force, deployments of troops, and cross-border incursions ~Tillema, 1989:419!. To be consistent with this definition of intervention, I exclude all foreign overt military interventions in Tillema s ~1997! dataset that he codes as being invited interventions. 6 Such cases also run counter to Rosenau s conceptualization of intervention as an act designed to alter another s authority structure. If a country comes to the rescue of a 5 Research on the applicability of the democratic peace to the intervention context by Kegley and Hermann ~1995a, 1996, 1997; also Hermann and Kegley, 1996! has been limited to the 1974 to 1991 time frame. In one study ~Hermann and Kegley, 1998! they utilize the entire Cold War temporal domain, but only for those interventions initiated by the United States. 6 I find six cases where a democracy was invited to intervene by another democracy. These include the British, Australian, and New Zealand interventions into Malaysia from 1964 to 1966, the British intervention into Uganda in 1964 during an army mutiny, the British intervention into Mauritius in 1968 ~during a series of riots!, and the British intervention into Cyprus in 1974 during the Turco-Cypriot War. An additional 98 invited interventions were found among dyads that were not jointly democratic directed dyads. Tillema ~1997! also codes several cases as partially invited to represent cases where only one internal faction judged to be a governing entity supported intervention, such as the U.S., and IAPF intervention into the Dominican Republic in The government of Donald Reid Cabral had fled, and while the forces of Generals Elias Wessin y Wessin and Antonio Imbert welcomed the external intervention, their rivals, led by Colonel Francisco Caamano, did not support the American presence ~Atkins and Wilson, 1972!. Only five partial cases are included in this study, and none involves a jointly democratic directed dyad. I code all partial cases as uninvited, given the lack of clear governmental support for the foreign intrusion.

6 584 The Dearth of Jointly Dyadic Democratic Interventions Table 3. The Regime Type of Foreign Overt Military Intervention Participants, DID AN INTERVENTION OCCUR? JOINTLY NOT JOINTLY Totals NO 5012 ~99.84%! ~99.44%! YES 8 ~0.16%! 186 ~0.56%! 194 Totals x , p,.0002 Fisher s Exact Test: The p-exact besieged regime, the act represents more of a preservation of that state s institutions rather than an attempt to change them. 7 It is important to note that this sample of interventions excludes covert cases. Scholars ~Forsythe, 1992; Kegley and Hermann, 1995b; James and Mitchell, 1995! contend that there have been situations where a democracy intervened covertly against freely elected governments. Empirical Analysis To assess whether the Democratic Peace Proposition applies to lower forms of conflict, I construct three 2 2 tables ~Tables 3 5!. The rows are distinguished by whether or not an intervention took place, while the column variable codes whether or not a directed dyad is jointly democratic or not. To identify dyads, Kegley and Hermann define their population of cases using the distribution of regime type of dyads found in the international system during the analyzed time frame ~1974 to 1991!. 8 But they do not provide important details about the dyads they use. Do Kegley and Hermann include cases where both members of the dyad could intervene in each other s internal affairs? Do Kegley and Hermann include all dyads, or those considered politically relevant? In generating a list of dyads, I use the EUGENE software ~Bennett and Stam, 1998! to create a list of directed dyads from 1945 to I select politically relevant dyads, defined by scholars ~Weede, 1976; Maoz and Russett, 1992, 1993! as contiguous states or pairs of countries where one state is labeled a major power by the Correlates of War project. 9 This reflects not only the distribution of regime type dyads in the interstate system, but also the cases where a country has the ability to intervene in the affairs of another country. In addition, my units of observation are defined by directed dyad cases, reflecting the potential 7 Kegley and Hermann ~1997! include a variety of invited interventions in their study, including a 1982 plea by Gambian president Jawara for Senegal to send troops to help suppress a Marxist-Leninist coup led by Kukoi Samba Sanyang ~Tillema, 1991!. Other invited interventions coded as hostile acts or overt aggression include the Panamanian request for United States assistance in quelling Herrera s Mutiny in 1990 ~Tillema, 1997!, British interventions to help the Jamaican government against the Henry Rebellion in 1960, and Malay-Thai cooperation on border operations ~Tillema, 1991!. 8 The analyses utilized by Kegley and Hermann ~1995a, 1996, 1997; also Hermann and Kegley, 1996! focus only on the regime type of dyads in the interventions or the number of regime type dyads within the international system. They do not consider whether or not the dyads can even interact with each other. Hermann and Kegley ~1996! conduct a regression analysis designed to determine the effects of polity type and several potentially confounding factors on being chosen as a target of intervention, but this does not address the issue of politically relevant dyads, their regime type and intervention. 9 In measuring contiguity, I include countries that either share a land border or are separated by 24 miles of water or less ~Bennett and Stam, 1998!.

7 John A. Tures 585 Table 4. The Regime Type of Foreign Overt Military Intervention Participants, DID AN INTERVENTION OCCUR? JOINTLY NOT JOINTLY Totals NO 6081 ~99.95%! ~99.48%! YES 3 ~0.05%! 134 ~0.52%! 137 Totals x , p,.0000 Fisher s Exact Test: The p-exact for both states to intervene in each other s internal affairs ~i.e., Pakistan s relationship to India is observed separately from India s relationship to Pakistan!, given that Tillema s ~1997! dataset codes interventions in such a manner. To determine if there are any temporal effects, I split the sample between and , in order to observe whether the rapid increase in democracy in the international system from 1974 to 1991 ~ Jaggers and Gurr, 1995! has any impact on the intervention proclivities of democracies and autocracies. Findings Jointly Democratic Dyadic Intervention: A Statistical Rarity The results show 11 DDIs ~cases where a democratic country intervened against a liberal target! between 1945 and Eight of these 11 DDIs occurred between 1945 and 1973, with the other 3 occurring during the second phase ~ !. This preliminary finding would appear to give support to Kegley and Hermann s argument that the Democratic Peace does not necessarily apply to lower forms of conflict such as interventions. Despite their occurrence, the presence of a DDI is quite rare in comparison with other types of interventions. Only 3.3 percent of the 331 interventions in this study were uninvited interventions by one democratic state against another democratic state. DDIs are also scarce in comparison with the number of jointly democratic directed dyads in the international system during the Cold War. Democracies intervened in other democracies only 0.09 percent of the time from Table 5. The Regime Type of Foreign Overt Military Intervention Participants, DID AN INTERVENTION OCCUR? JOINTLY NOT JOINTLY Totals NO ~99.9%! ~99.41%! YES 11 ~0.099%! 320 ~0.59%! 331 Totals x , p,.0000 Fisher s Exact Test: The p-exact

8 586 The Dearth of Jointly Dyadic Democratic Interventions 1945 through The rate of interventions among not jointly democratic dyads is 5.95 times greater during this same time frame. In the early years of the Cold War ~ !, democracies intervened against other liberal states eight times, or 0.16 percent of the time. Among the not jointly democratic dyads during these same years, the percentage of intervention observations is 0.56 percent, or 3.5 times greater than the rate of interventions among jointly democratic dyads. The concluding years of the Cold War ~ ! provide an even greater disparity between intervention rates in the two types of dyads. The percentage of intervention observations found among the not jointly democratic dyads is 0.52 ~134025,682!. This is 10.4 times greater than the percentage of intervention observations found among the jointly democratic directed dyads ~0.05 percent, or 306,084!. Tables 3 through 5 demonstrate that the scarce number of DDIs and the lower rate of intervention among jointly democratic dyads are no accident. The chisquare statistics for all three time periods ~ , , and ! are all significant at the.001 level, indicating strong support for the impact of regime type on who intervenes and who is the target. These findings are confirmed by results using Fisher s Exact Test. Democracies rarely ever intervene against one of their liberal brethren and the results cannot be written off to chance or the well-proclaimed paucity of democracy in the international system, an argument frequently employed by critics of the Democratic Peace. These eleven cases, including the participants, year~s! of conflict, and conflict subjects, are listed in Table 6. Additional Examination of Initiators and Targets Analysis of the intervention initiators and targets provides some insight into the regime type of attackers and defenders between 1945 and Here I look at the regime type of those who conduct and receive interventions, and compare the findings from the time period to the time frame. The results can be found in Figure 1. Scholars studying Democratic Peace have noted that democracies often engage autocratic countries in combat ~Maoz and Russett, 1992, 1993; Kegley and Hermann, 1995a, 1996, 1997; Bueno de Mesquita et al., 1999!. My results show there have been 87 cases where a democracy chose to intervene in the internal affairs of a nondemocracy; the number of such interventions has declined since 1974 ~only 23 occurred, as compared to the 64 cases of democracy opposing autocracy Table 6. Cases of Intervention Within a Jointly Democratic Directed Dyad Intervention Initiator Intervention Target Dates of Intervention Name of Intervention Pakistan India Kashmiri Raid Israel Syria Kinneret Shelling Syria Israel Kinneret Shelling Israel Syria Kinneret Raids India Pakistan Rann of Kutch Conflict Pakistan India Rann of Kutch Conflict Israel Syria Border Shelling Syria Israel Border Shelling Turkey Cyprus Turco-Cypriot War Peru Ecuador Paquisha Incident Ecuador Peru Corrientes Incident

9 John A. Tures 587 Fig. 1. Interventions by Dyad ~Regime Type of Initiators and Targets! Between 1945 and 1991 between 1945 and 1973!. Situations involving autocrats targeting democrats have shown similar patterns; after witnessing 42 cases of nonliberal states attacking democracies between 1945 and 1973, I find only 19 occurred since Some of the explanation behind the behavior of mixed dyads might be attributed to Cold War patterns. In the early stages of the Cold War, the political landscape was marred by conflict between the democratic West and autocratic East, as well as their proxies. After a period of détente, Cold War tensions subsided and neither side reacted with the same level of hostility that characterized the previous time period. More than half ~51.9 percent! of all foreign overt military interventions consisted of an autocratic initiator and an autocratic target. An even more disturbing trend can be found in Figure 1. Whereas all other types of dyadic interventions experienced a decline in interventions after 1974, the number of autocratautocrat interventions increased during this time frame ~from 80 between 1945 and 1973 to 92 between 1974 and 1991!. Conclusions In this study, I find that democracies rarely engage in a military intervention against their democratic counterparts. Only in 11 cases does a democracy intervene against another democracy. These cases of DDI are rare compared with interventions by democratic and0or autocratic states in undemocratic states, or by autocratic states against democratic states. At a time when scholars proclaimed the presence of a Third Wave of Democracy ~ Jaggers and Gurr, 1995!, the number of DDIs fell by more than half their number from the previous decades ~ !. These results counter the pessimistic tone of Kegley and Hermann ~1995a, 1996, 1997!, who not only find that democracies are engaging in intervention at rates not significantly lower than those found by observing not jointly democratic directed dyads, but claim that such cases of DDI are on the rise. Furthermore, three dyads ~India-Pakistan, Israel-Syria, and Ecuador-Peru make up 10 of the 11 cases of DDIs, while Chan ~1997! questions the credentials of the Turkey-Cyprus case as a conflict between

10 588 The Dearth of Jointly Dyadic Democratic Interventions democracies. 10 These 11 DDIs pale in comparison to the 172 cases of interventions between autocracies. Couple this information with the rate of interventions among the not jointly democratic dyads in comparison to the rate of interventions among jointly democratic dyads and a strong conclusion might be reached about the relative dearth of dyadic democratic interventions. The difference in my findings from Kegley and Hermann s concerns specifications for states, democracies, interventions, dyads, and time frame. By expanding the sample to include cases from 1945 to 1973, and by employing a measure of democracy that does not code systems with heavy government power concentrations as democracies, including politically relevant directed dyads and omitting invited interventions, nonstate entities, and occupied states, I find support for the application of the Democratic Peace Proposition to lower forms of conflict. Suggestions for Future Research The evidence in this analysis shows that few democracies intervene in the internal affairs of other democracies, confirming support for the application of the Democratic Peace to lower forms of conflict. However, scholars have recently suggested that the relationship between democracy and peace may be endogenous. James, Solberg, and Wolfson ~1999! contend that simultaneous causation may be occurring, although their data analysis reveals a relatively weak substantive effect in each direction. Peace may also be producing democracy, providing stable conditions for such institutions to emerge. If this is the case, traditional models implemented in analyzing the Democratic Peace may be underspecified. These arguments have been supported by other researchers ~Layne, 1994; Gates, Knutsen, and Moses, 1996; Chan, 1997!. Mousseau and Shi ~1999! dispute these findings, testing the relationship between war and autocratization. They find that all countries are as likely to become more institutionally autocratic as they are to become even more democratic in the periods before the onset of war ~Mousseau and Shi, 1999: !. Their results show that there is relative institutional stability in democracies in their war preparation. However, these studies have focused primarily on wars. Do these arguments of reverse causality apply to conflicts short of war? At a minimum, the contentions of James, Solberg, and Wolfson ~1999! are too important for this growing body of international relations literature to ignore. Critics have responded with an emphasis on issues related to levels of analysis, excluding variables and measurement ~Oneal and Russett, 2000; for a reply see James, Solberg, and Wolfson, 2000!. Subsequent research on the subject of liberalism and lower-level conflicts should focus on the anterior, concurrent, and posterior effects of military intervention upon democracy. In other words, do democracies autocratize before, during, or after a foreign overt military intervention? Such findings will tell us whether democracies and foreign overt military interventions suffer the same endogeneity concerns that others feel may be creeping into our analysis of democracy and war. References Atkins, G. P., and L. C. Wilson ~1972! The United States and the Trujillo Regime. New Brunswick, NJ: Rutgers University Press. 10 Chan ~1997! questions the coding of the Turkey-Cyprus conflict as an intervention between democracies. Additional research shows that Cypriot democracy was ousted in a military coup on July 16, 1974, before the Turkish invasion ~McLaughlin et al., 1998!. Since democracy was not restored in Cyprus until February 14, 1975, the conflict dyad is difficult to justify as jointly democratic.

11 John A. Tures 589 Bennett, D. S., and A. C. Stam III ~1998! EUGENE: Expected Utility Generating Program, Version University Park: Pennsylvania State University. Bueno de Mesquita, B., J. D. Morrow, R. M. Siverson, and A. Smith ~1999! An Institutional Explanation of the Democratic Peace. American Political Science Review 93: Chan, S. ~1997! In Search of Democratic Peace: Problems and Promise. Mershon International Studies Review 41: Forsythe, D. ~1992! Democracy, War and Covert Action. Journal of Peace Research 29: Gates, S., T. Knutsen, and J. W. Moses ~1996! Democracy and Peace: A More Skeptical View. Journal of Peace Research 33~1!:1 10. Gurr, T. R., K. Jaggers, and W. H. Moore ~1989! The Polity II Codebook. Boulder: University of Colorado Center for Comparative Politics. Hermann, M. G., and C. W. Kegley, Jr. ~1996! Ballots, a Barrier Against the Use of Bullets and Bombs: Democratization and Military Intervention. Journal of Conflict Resolution 40: Hermann, M. G., and C. W. Kegley, Jr. ~1998! The U.S. Use of Military Intervention to Promote Democracy: Evaluating the Record. International Interactions 24: Jaggers, K., and T. R. Gurr ~1995! Tracking Democracy s Third Wave with the Polity III Data. Journal of Peace Research 32: James, P., and G. E. Mitchell II ~1995! Targets of Covert Pressure: The Hidden Victims of the Democratic Peace. International Interactions 21: James, P., E. Solberg, and M. Wolfson ~1999! An Identified Systemic Model of the Democracy-Peace Nexus. Defence and Peace Economics 10~1!:1 37. James, P., E. Solberg, and M. Wolfson ~2000! Democracy and Peace: A Reply to Oneal and Russett. Defence and Peace Economics 11: Joyner, C. C. ~1992! International Law. In Intervention into the 1990s: U.S. Foreign Policy in the Third World, edited by P. J. Schraeder, pp Boulder, CO: Lynne Rienner. Kegley, C. W., Jr., and M. G. Hermann ~1995a! Military Intervention and the Democratic Peace. International Interactions 21:1 21. Kegley, C. W., Jr., and M. G. Hermann ~1995b! The Political Psychology of Peace Through Democratization. Cooperation and Conflict 30:5 30. Kegley, C. W., Jr., and M. G. Hermann ~1996! How Democracies Use Intervention: A Neglected Dimension in Studies of the Democratic Peace. Journal of Peace Research 33: Kegley, C. W., Jr., and M. G. Hermann ~1997! Putting Military Intervention into the Democratic Peace: A Research Note. Comparative Political Studies 30: Layne, C. ~1994! Kant or Cant: The Myth of the Democratic Peace. International Security 19~2!:5 49. Maoz, Z., and B. Russett ~1992! Alliance, Contiguity, Wealth, and Political Stability: Is the Lack of Conflict Among Democracies a Statistical Artifact? International Interactions 17: Maoz, Z., and B. Russett ~1993! Normative and Structural Causes of the Democratic Peace, American Political Science Review 87: Marshall, M. G., and K. Jaggers ~2001! Polity IV Project: Political Regime Characteristics and Transitions, College Park: University of Maryland CIDCM. McLaughlin, S., S. Gates, H. Hegre, R. Gissinger, and N. P. Gleditsch ~1998! Timing the Changes in Political Structures. Journal of Conflict Resolution 42: Mousseau, M., and Y. Shi ~1999! A Test for Reverse Causality in the Democratic Peace Relationship. Journal of Peace Research 36~6!: Oneal, J. R., and B. Russett ~1997! The Classic Liberals Were Right: Democracy, Interdependence, and Conflict, International Studies Quarterly 41: Oneal, J. R., and B. Russett ~2000! Comment: Why An Identified Systemic Model of the Democracy- Peace Nexus Does Not Persuade. Defence and Peace Economics 11: Ray, J. L. ~1995! Democracy and International Conflict: An Evaluation of the Democratic Peace Proposition. Columbia: University of South Carolina Press. Thompson, W., and R. Tucker ~1997! A Tale of Two Democratic Peace Critiques. Journal of Conflict Resolution 41: Tillema, H. K. ~1989! Foreign Overt Military Intervention in the Nuclear Age: A Clarification. Journal of Peace Research 26: Tillema, H. K. ~1991! International Conflict Since 1945: A Bibliographic Handbook of Wars and Military Interventions. Boulder, CO: Westview Press. Tillema, H. K. ~1997! Foreign Overt Military Interventions, : OMILIST Codebook. Columbia: University of Missouri. Weede, E. ~1976! Overwhelming Preponderance as a Pacifying Condition Among Contiguous Asian Dyads, Journal of Conflict Resolution 20: