Testing the Waters: Exploring Why Democracies Have More Maritime Conflict

Abstract

While the democratic peace literature finds that pairs of democracies face lower risks for interstate conflict than other regime pairings, the maritime conflict literature often finds that democratic dyads have the highest risks for maritime diplomatic conflicts out of all regime pairings. Despite this surprising finding, there remains a gap for understanding what specific political mechanisms are driving democracies to engage in maritime conflict with other democracies. We explore whether democracies have more opportunities for maritime conflict generally, whether specific dimensions of democracy influence maritime conflict (domestic interests and executive constraints), including distinguishing between presidential and parliamentary democracies, and whether territorial integrity norms are weaker at sea. We find that coastal states are more democratic than landlocked states and that coastal dyads with maritime claims are more democratic than peaceful coastal dyads. We also show that multiple dimensions for democracy help explain the pattern of democratic maritime conflicts and that dyads with stronger domestic interests and more constrained executives (especially presidential democracies) are more likely to experience maritime conflicts. Finally, we show that territorial integrity norms have weaker effects on maritime conflict behavior, which helps us understand why democracies may be less constrained in their revisionist behavior in the maritime arena.

Si bien es cierto que la literatura en materia de paz democrática concluye que los pares de democracias se enfrentan a menores riesgos de conflicto interestatal que otros pares de regímenes, Daniels y Mitchell (2017) concluyeron que las díadas democráticas tienen los riesgos más altos de todos los emparejamientos de regímenes en lo referente a los conflictos diplomáticos marítimos. A pesar de este sorprendente hallazgo, sigue existiendo una brecha que nos impide comprender cuáles son los mecanismos políticos específicos que están impulsando a las democracias a participar en conflictos marítimos con otras democracias. Estudiamos si esto se debe a que las democracias tienen más oportunidades para el conflicto marítimo, en general, si existen dimensiones específicas de la democracia, las cuales, influyen en el conflicto marítimo (intereses internos y restricciones ejecutivas), incluyendo la distinción entre democracias presidenciales y parlamentarias, y si las normas de integridad territorial son más débiles en el mar. Concluimos que los Estados costeros son más democráticos que los Estados sin litoral, y que las díadas costeras con reivindicaciones marítimas son más democráticas que las díadas costeras pacíficas. También demostramos que las múltiples dimensiones de la democracia ayudan a explicar el patrón de los conflictos marítimos democráticos, y que las díadas con intereses internos más fuertes y Gobiernos ejecutivos más limitados (especialmente las democracias presidenciales) tienen más probabilidades de experimentar conflictos marítimos. Finalmente, demostramos que las normas de integridad territorial tienen efectos más débiles sobre el comportamiento de los conflictos marítimos, lo que nos ayuda a entender por qué las democracias pueden estar menos limitadas en su comportamiento revisionista en el ámbito marítimo.

Bien que la littérature consacrée à la paix démocratique affirme que les paires de démocraties sont moins exposées aux conflits interétatiques que les autres paires de régimes, Daniels et Mitchell (2017) constatent que, toutes paires de régimes confondues, ce sont justement ces dyades démocratiques qui courent les risques les plus élevés en matière de conflits diplomatiques maritimes. Malgré ce résultat surprenant, il reste à identifier les mécanismes politiques spécifiques qui poussent les démocraties à s'engager dans des conflits maritimes avec d'autres démocraties. Nous cherchons ici à répondre aux questions suivantes : les démocraties sont-elles généralement plus propices aux conflits maritimes ? Certaines dimensions de la démocratie, comme les intérêts nationaux et les contraintes de l'exécutif, influencent-elles les conflits maritimes (en faisant notamment la distinction entre les démocraties présidentielles et parlementaires) ? Les normes d'intégrité territoriale sont-elles plus souples en mer ? Nous constatons que les États côtiers sont plus démocratiques que les États enclavés. Par ailleurs, les dyades côtières ayant des revendications maritimes sont plus démocratiques que les dyades côtières pacifiques. Nous montrons que les multiples dimensions de la démocratie contribuent à expliquer le schéma des conflits maritimes démocratiques et que les dyades ayant des intérêts nationaux plus forts ainsi que des cadres exécutifs plus contraignants (en particulier les démocraties présidentielles) sont plus susceptibles de faire face à des conflits maritimes. Enfin, nous montrons également que les normes d'intégrité territoriale ont moins de pouvoir sur le déroulement des conflits maritimes. Cela nous aide à comprendre pourquoi les démocraties semblent plus libres dans leur comportement révisionniste sur la scène maritime.

Introduction and Motivation

While there have been no interstate wars fought between two fully democratic states (Maoz and Russett 1993, Russett and Oneal 2001), democracies frequently engage in diplomatic and lower level military conflicts with each other. From 1946 to 1992, for example, there were ninety-seven militarized disputes between democracies, almost half of which involved fishing, offshore oil, and offshore mineral resources (Mitchell and Prins 1999). To further test this maritime democratic conflict linkage, Daniels and Mitchell (2017) empirically analyzed a dataset that includes all politically relevant coastal dyads. The authors find that democratic dyads experience higher risks for maritime diplomatic conflicts in comparison to fully autocratic dyads and mixed regime dyads, particularly when dyads have similar energy production capabilities and if they have experienced militarized conflict in the past.¹

Daniels and Mitchell (2017) argue that democracies are more likely to experience maritime conflicts with other democracies for two reasons. First, democracy correlates positively with economic development (Lipset 1994), and most states do not delimit their maritime borders until they achieve a certain level of economic development. As a result of their higher economic development, democracies have more economic interest in maritime regions and more opportunities for maritime economic interests to come under threat, such as when fishing fleets compete for resources with other countries’ fishing fleets. The second reason democracies are more likely to experience maritime conflict involves maritime security interests. Daniels and Mitchell (2017) argue that maritime spaces have become essential to national security, and democracies are keenly aware of security risks like transnational terrorist attacks and economic vulnerability to shipping lanes being targeted (Klein 2011). To protect maritime security interests, democracies may be more likely to engage in interstate maritime conflicts, especially when they have past militarized disputes with maritime neighbors and after 9/11/2001 when international security risks like terrorism increased.

Despite Daniels and Mitchell's (2017) finding that democratic dyads have the highest chance of engaging in diplomatic maritime conflicts among all dyad types, we know less about the variation in maritime conflict behavior within democracies and the specific political mechanisms driving democracies toward maritime conflict. To further illuminate this relationship, we extend the authors’ research by comparing average democracy levels of land-locked dyads to maritime dyads, by replicating Daniels and Mitchell's (2017) analysis using Varieties of Democracy (V-Dem) indices that capture variation in domestic interests and executive constraints, distinguishing between presidential and parliamentary democracies, and by analyzing whether territorial integrity norms are less constraining at sea than on land.

In terms of capturing whether democracies have more opportunities for maritime claims because coastal states may be more democratic, we find that maritime dyads are on average more democratic than landlocked dyads, but we also find that maritime dyads with diplomatic conflict are even more democratic on average. Through dyad-year logistic regression analyses, we find all high-level V-Dem democracy indicators are positively correlated with maritime claims, which shows that Daniels and Mitchell's (2017) results generalize with different operationalizations of democracy. We further explore variation in domestic interests and executive constraints, finding that maritime conflicts emerge between democracies where domestic interests can put greater pressure on the government and where executives are more constrained. We also show that presidential democracies are more conflict-prone than parliamentary democracies in the maritime sphere, consistent with arguments about presidential democracy leaders being less constrained in foreign policy. Finally, territorial integrity norms have weak effects on maritime conflicts, indicating that democracies are less constrained at sea by the norms that help to partially explain the democratic peace in the territorial sphere. Yet, maritime conflicts that challenge exclusive economic zone (EEZ) boundaries are more likely to involve less democratic dyads, which shows some territorial integrity constraints. These results provide fruitful avenues for future maritime conflict research and suggest that international relations scholars need to consider the maritime sphere as a substantively different space with different norms of democratic behavior.

More Opportunities for Democracies?

Maybe democratic dyads have more opportunities for maritime conflicts than other regime pairings because democracies are more likely to be coastal states than autocratic countries. If that were true, then the finding that democratic dyads have more maritime conflicts would simply be an artifact of such states having more opportunities for such exchanges. The idea that economic development levels create a baseline condition for engagement in the maritime sphere fits with this expectation (Daniels and Mitchell 2017), but we can also consider more generally if average democracy levels in landlocked and maritime dyads are different, and if democratic dyads that contest maritime areas are even more democratic than the baseline set of maritime dyads.

To determine if opportunities help explain the patterns of maritime diplomatic conflict, we construct a dyadic dataset for all dyads in the international system that includes fully landlocked dyads, maritime dyads, and mixed land-maritime dyads. We use the peacesciencer package in R that allows for the construction of state system datasets using either Correlates of War (COW) or Gleditsch-Ward state systems (Miller 2022). The resulting international COW state system dataset includes 1,385,292 dyad-years from 1900 to 2007.²

Next, we combine the COW state system dataset with the dyad-year dataset from Daniels and Mitchell (2017) which pairs together all coastal states in a COW-designated region as well as pairing major powers with each coastal state in a region. The dataset is composed of 88,212 dyad-years for all paired coastal states in each region (73.2 percent) and an additional 32,230 dyad-years of major powers paired with coastal states in a region (26.8 percent) for a total of 120,442 dyad-years. The regions included in the dataset are the Americas (South, Central, and North), Europe (both Eastern and Western), the Middle East, and Asia. By combining these two datasets together, we can distinguish maritime and nonmaritime dyad-years. Then we generate a dyadic dummy variable that equals one if both states in a dyad are landlocked (439,446 cases or 31.7 percent of all dyad-years). In our models of maritime claims, we exclude the landlocked dyads and land-maritime dyads (59.6 percent of all dyads) because these states have no possibility for engaging in maritime conflict with each other.³

To capture the features of democracy that relate to maritime conflict behavior, we start by analyzing the five V-Dem Project's high-level democracy indices that are designed to capture democracy at the highest or most abstract level: (1) electoral democracy, (2) liberal democracy, (3) participatory democracy, (4) deliberative democracy, and (5) egalitarian democracy (Coppedge et al. 2023). Electoral democracy captures free and fair elections, free press, and the ability for civil society groups to operate freely. Liberal democracy captures the extent to which the rule of law operates in a country and how institutions (e.g., independent judiciary) put checks and balances on the exercise of executive power. Participatory democracy records citizen participation in a democracy through direct democracy and civil society groups. Deliberative democracy relates to the degree of free and open political deliberation in a country. Egalitarian democracy picks up equal protection of groups and their equal access to power and resources. Each of these variables is an interval measure that ranges in value from 0 to 1, with 0 being the least democratic in terms of the feature and 1 being the most democratic. We use a weakest link approach, like many analyses in the democratic peace literature, where we record the least democratic country's score on an indicator in each dyad-year. In our multivariate analyses, we also examine specific V-Dem indicators that capture the mechanisms connecting democracy and maritime conflict behavior (domestic interests and executive constraints).

In table 1, we compare the average democracy scores for the least democratic country in a dyad from 1900 to 2007 using the five high-level V-Dem indicators for three groups: landlocked dyads, peaceful maritime dyads, and conflictual maritime dyads.⁴ We see that landlocked dyads have lower democracy scores across all V-Dem measures compared with maritime dyads (peaceful or conflictual). One-way ANOVA tests (Online AppendixTable A1) across the three groups show that the group democracy means are statistically different from each other for all democracy measures. This shows that dyads with maritime states are on average more democratic than dyads with landlocked states.

Yet we also find significant differences in average democracy scores when comparing coastal dyads without contested diplomatic claims to maritime areas (peaceful maritime dyads) and those with claims (conflictual maritime dyads). As noted earlier, democratic dyads may experience maritime conflicts more frequently because they tend to have higher economic development levels (Daniels and Mitchell 2017), which leads to a greater usage of maritime shipping lanes and more frequent extraction of maritime resources. Less developed countries often fail to claim a territorial sea and EEZ area, which diminishes the risks for maritime conflicts with neighbors. Many resources like offshore oil are costly to extract, requiring heavy investment and technologically advanced operations to be viable economically. Nyman (2015) finds that maritime conflict onset increases significantly post-1947 when offshore oil rig platform technology is developed, suggesting that the capacity to develop offshore resources is a strong indicator of maritime conflict behavior. These factors contribute to why democracies may be more revisionist in their claims against other democracies when potential maritime EEZs with valuable resources overlap. Clashes between democratic states also create greater security risks, with 29 percent of all maritime claims involving one or more militarized disputes (Hensel and Mitchell 2017). Democracies have generally avoided escalating maritime conflicts to severe levels, but they are willing to use their coast guards and navies to protect resource extraction, something we have seen in cases like the Cod Wars between the United Kingdom and Iceland.⁵

What Features of Democracy Encourage Maritime Conflict?

While Daniels and Mitchell's study shows that democratic dyads experience more maritime claims than autocratic or mixed dyads using Polity IV scores, we still have questions about what features of democratic institutions explain these patterns. The V-Dem high-level democracy indicators provide comparable data to Polity IV for capturing regime characteristics at its highest level and help us replicate the original study's results with a more comprehensive dataset. V-Dem also provides richer information about indicator variables to help us capture different dimensions of democracy that may encourage revisionist maritime foreign policy. We focus on two factors: domestic interests and executive constraints.

Domestic Interests

Democracies might experience more maritime conflicts with other democracies because maritime interests in both states could push governments to pursue claims against each other. Prior research shows that domestic groups can play an important role in how governments negotiate for maritime resource rights. In a study of fisheries negotiations involving Norway and Iceland, Ásgeirsdóttir (2007, 2009) shows that a country with more powerful domestic interests in the fishing industry can help put pressure on the government to extract more in negotiations with other countries, especially if those opposing states have less entrenched domestic fishing interests. Østhagen (2021) also notes that domestic groups pressure governments to protect fishing, energy, and other maritime resources. For example, to convince Russia to take a stronger stance against Norway, the Russian fishing industry criticized its government for being weaker compared to the era of the Soviet Union when negotiating fishing rights in Svalbard in the 1990s. Maritime issues can take on national political significance even though they often affect a small percentage of citizens, as seen in the importance of maritime issues in pro-Brexit arguments in the United Kingdom and the salience of the San Andres Archipelago maritime dispute with Nicaragua in the 2018 Colombian presidential elections (Østhagen 2021) While the high-level V-Dem indicators for electoral, deliberative, and participatory democracy capture some aspects of domestic interests, we also consider four specific indicators that reflect the ability of domestic and regional interests to put pressure on the national government: (1) civil society participation, (2) election paid interest group media, (3) regional offices’ relative power, and (4) division of national/local power.

Executive Constraints

While democracies might have economic and security interests for pursuing maritime claims, supported by domestic interests, leaders can also be constrained differently across democratic institutions. Previous interstate conflict research suggests that executive constraints vary in important ways across democratic regimes, which can influence the propensity for interstate conflict. Leaders in presidential democracies typically have more foreign policy autonomy than leaders in parliamentary democracies, where leaders are more constrained by party elites. In coalition governments, party members also can overturn a conflict decision or alter policy in a way that harms the government's success rate in a foreign policy situation (Auerswald 1999), which puts an additional audience cost check on revisionist actions abroad. As Elman (2000, 99) notes, parliamentary systems based on proportional representation tend to have foreign policy strategies that are “diluted to command the support of various parties. Since the cabinet must please many parliamentary groups to retain legislative confidence, foreign security policies are likely to be tailored toward maintaining consensus.” In coalition systems, serious disagreements can also lead to elections being called, which places additional constraints on leaders in such systems.

To evaluate how differences in democratic regimes influence states’ behavior in maritime issues, we control for presidential/parliamentary systems monadically and compare dyads with presidential democracies to mixed or parliamentary dyads to see whether the former group is less constrained by their legislative and public audiences, and thus freer to challenge other states’ maritime areas and resources. We also utilize four specific indicators from V-Dem that capture various aspects of executive constraints: (1) legislative constraints on the executive, (2) judicial constraints on the executive, (3) executive corruption, and (4) executive election turnover.

Research Design

The dependent variable in these analyses focuses on the occurrence of a maritime claim. The data for this variable comes from the Issue Correlates of War (ICOW) maritime dataset and codes years in which a given dyad has one or more ongoing diplomatic claims over maritime issues that can involve the ownership, access to, or usage of a maritime area (Mitchell 2020). Maritime claims typically involve delimitation of maritime areas (e.g., EEZ boundaries), questions about the baselines from which those lines are drawn (e.g., the status of rocks vs. islands), access to fishing or other resources, or navigational rights in the area. ICOW identifies 221 total dyadic maritime claims from 1900 to 2007 in the Western Hemisphere, Europe, Asia, and the Middle East. Of the 120,442 coastal dyads in the Daniels and Mitchell dataset that could experience a claim, 3.6 percent of the dyad year cases (N = 4,374) involve an active maritime claim.

We use the weakest link approach to capture the lowest electoral democracy index value in a dyad-year and use that as the democratic indicator for the entire dyad (mean = 0.32). To differentiate between presidential versus parliamentary democracies, we use the Democratic Electoral Systems (DES) dataset that provides a simple classification scheme for all electoral systems in the world from 1946 through 2011 (Bormann and Golder 2013). We generate monadic binary variables from the dataset's regime variable to denote states that are parliamentary and states that are presidential. Finally, we generate dyad-level variables indicating if a dyad is jointly presidential, jointly parliamentary, or mixed (presidential−parliamentary) each year. In the maritime dyad sample, 13 percent of dyads are parliamentary, 26 percent are presidential, 32 percent are mixed, and 29 percent are nondemocratic. Among dyads that have ongoing maritime claims, 8 percent of dyads are parliamentary, 31 percent are presidential, 37 percent are mixed, and 24 percent are nondemocratic.⁶

To account for other factors influencing maritime conflict, we estimate each model with the full set of substantive controls that Daniels and Mitchell (2017) use in their analysis. First, we include a binary variable that indicates whether a dyad-year experienced a militarized interstate dispute (1.4 percent). Next, we measure relative energy capabilities using data from version (4.0) of the COW capabilities data set. The COW measures provide information on relative dyadic primary energy consumption by taking each country's energy consumption estimate, which is based on consumption of coal, petroleum, electricity, and natural gas. A dyadic measure is then generated by dividing the more energy-capable state's (i.e., larger energy usage) value by the dyadic total sum. This creates a variable that ranges theoretically from 0.5 (the two states have equal energy consumption) to 1.0 (the stronger state has all the energy consumption) (mean = 0.89). The models also include a binary measure for whether a dyad contains a major power (38 percent).⁷ We also account for temporal and spatial patterns in the dataset by including time polynomials (year and year squared) and binary variables for the four regions in the data: Americas, Europe, Asia, and the Middle East.⁸ Due to the binary nature of the dependent variable, we estimate a series of logistic regression models.

In table 2, we present the results using the five high-level dimensions of democracy as captured by the V-Dem project. We replicate the Daniels and Mitchell (2017) findings and see that all five measures have a positive and statistically significant effect on the occurrence of maritime diplomatic conflict. In figure 1, we show the substantive effects for each democracy measure. Participatory and deliberative democracy produce the largest changes in overall risks for maritime conflicts, which fits with the literature that focuses on how domestic fishing or other interests could lead to conflictual behavior in the maritime arena.

In Online Appendix Table A9 (Models 1–4), we examine four indicator variables that capture the degree to which domestic interests can put significant pressure on the national government. Our measures record the value for the country with the weakest domestic interest pressure. We find that three of the four domestic interest variables are positive and significant: civil society participation, regional offices’ relative power, and division of national/local power. This supports arguments in the maritime literature that domestic interests in the marine sphere can increase maritime conflicts between democracies because both governments are more likely to face such pressures at home for action abroad as the weakest link indicator increases.

Next, we consider how variation in executive constraints may help explain why some democracies are especially contentious in the maritime sphere. In table 3, we examine the monadic effects of presidential/parliamentary democracy on state A and state B in the coastal dyad year dataset (Models 1–2). Parliamentary or presidential democracies have higher risks for maritime conflict (except Presidential, State B). However, when we control for the overall electoral democracy level (Model 2), we find that only presidential democratic countries are more likely to be involved in maritime conflicts (in the State A role). In Model 3, we examine dyadic pairings to see whether parliamentary dyads are distinct from presidential dyads, while also controlling for mixed dyads (one parliamentary and one presidential). As shown in the descriptive statistics earlier, parliamentary dyads have fewer maritime claims than presidential or mixed dyads. We see a similar pattern in the multivariate model (table 3, Model 3). While the results are not entirely consistent across the models, the dyadic results generally support expectations from previous international relations literature about the greater constraining effects of parliamentary systems on interstate conflict.⁹

In Online Appendix Table A9 (Models 5–8), we examine four indicator variables that capture differences in executive constraints. Our measures record the value for the country with the weakest executive constraints. Two of the four executive constraint variables are positive and significant for maritime conflicts: legislative (Model 5) and judicial (Model 6) constraints on the executive. Institutional checks do not prevent democratic leaders from pursuing their states’ maritime interests. Executive corruption, which has a negative and significant coefficient, is also consistent with our other findings because it shows that dyads with leaders that have more freedom to take bribes or embezzle state funds (e.g., are less democratic) are less likely to engage in maritime claims. The effect of recent elections that produced executive turnovers is not significant (Model 8). Overall, the results confirm that different features of democracy push towards greater revisionism at sea, but that institutional structures may create differences between democracies in how they pursue maritime claims.

Are Territorial Integrity Norms Less Constraining at Sea?

Most wars in history involved contests over territory. Territorial integrity norms emerged in the international system following attempts to prohibit territorial aggression that failed (e.g., the 1928 Kellogg–Briand Pact) and costly experiences that countries felt after the World Wars (Zacher 2001). Under the territorial integrity norm, states are discouraged from using force to alter interstate boundaries and should instead respect their neighbors’ territorial borders. As Zacher (2001) describes, this norm emerged out of the concepts of nationalism and self-determination, and limited states' ability to transfer territory without respect for the will of the inhabitants. Hensel, Allison, and Khanani (2009) note that the territorial integrity norm encompasses norms against conquest and norms against the use of violence to pursue territorial changes. Altman (2020, 494) also points to the importance of nonviolence, arguing that under the territorial integrity norm, it is “impermissible to deploy a military force to seize territory from another state without that state's consent in furtherance of a claim to sovereignty over the territory.”

Democracies have been compliant with territorial integrity norms. Only 2 percent of militarized disputes between fully democratic countries since World War II involved territorial issues (Mitchell and Prins 1999). Enduring rivalries, which often continue because of border disputes, are more likely to terminate when one or both states become democratic (Hensel, Goertz, and Diehl 2000). This is consistent with the theoretical logic that because democracies have larger selectorates and winning coalitions, they pursue policies that provide public goods to a larger group, rather than territorial grabs that might provide resources to a smaller support group (Bueno de Mesquita et al. 2005). The territorial peace literature questions the ordering of the relationship, positing that territorial peace precedes democracy (Gibler 2012; Owsiak and Vasquez 2023), but there is little doubt that fully democratic countries rarely challenge their neighbors’ territorial borders.

On the other hand, our results show that pairs of democracies are most likely to challenge the issue status quo when it comes to maritime issues. One possibility is that the territorial integrity norm, which was shaped in terms of land border claims, does not influence how countries negotiate over maritime areas. If this were the case, then we would expect to find a null relationship between measures of the territorial integrity norm and maritime conflict behavior. On the other hand, some maritime claims are more “territorial” than others and thus might mimic the pattern of less democratic states challenging those “borders” more often. Owsiak and Mitchell (2019) suggest that EEZ or maritime border delimitation claims are similar to territorial claims because they deal with how a boundary is determined. The authors show that peaceful conflict management strategies for EEZ maritime claims are more like territorial claims than for the management of non-EEZ maritime issues, with for example, countries avoiding adjudication to settle disputes and using bilateral negotiations more frequently in EEZ claims.

To test these ideas, we operationalize territorial integrity norms using three different system-level measures from ICOW's supplemental Multilateral Treaties of Pacific Settlement dataset (Mitchell and Hensel 2007). The first variable is a dummy variable set to one for dyad-years after 1976, with the date being based on Zacher's (2001) historical analysis of the territorial integrity norm. The second variable measures the average number of global or regional Pacific settlement treaties with five or more members (mean = 3.9). In these treaties, countries promise to settle interstate disputes with peaceful conflict management tools. The third variable is a similar measure that captures the average number of global or regional treaties with territorial integrity clauses and includes five or more members (mean = 2.2). In these treaties, the member states agree not to violate the territorial integrity of the other member states, showing support for territorial integrity. We also create a maritime claims occurrence variable for EEZ/border delimitation claims only where the primary issue at stake involves delimitation of the EEZ area (2,174 dyad years or 1.8 percent of the maritime dyad dataset).

We see in table 4 that none of the territorial integrity measures has a significant effect on diplomatic maritime conflicts. The Zacher measure has a positive coefficient (p = 0.068), suggesting that maritime claims are more likely in the strongest era of territorial integrity norms (1976+). This is understandable given that the United Nations Convention on the Law of the Sea (UNCLOS) treaty was not signed until 1982 and did not come into legal force until 1994, which created incentives for countries to stake out claims to maritime areas before the more precise legal rules articulated in the treaty would be constraining on UNCLOS states parties.¹⁰ Membership in regional or global organizations that call for peaceful dispute settlement and territorial integrity has little constraining effect on the behavior of coastal states. Thus, democracies may engage in maritime conflicts more frequently than territorial ones because they are less constrained by the norms that govern land areas.

In the Online Appendix (Table A4), we present One-Way ANOVA analyses comparing the average democracy scores for EEZ claim dyad-years compared to no-claim/non-EEZ dyad-years. For each measure, we observe that the average democracy score is lower in the EEZ claim dyad years, with statistically significant differences between groups for four of the five V-Dem measures (except deliberative democracy). These patterns are consistent with a type of territorial integrity norm at sea, whereby challenges to proposed maritime borders occur more often in less democratic dyads. This shows that democratic countries may pursue access to fishing resources, as Britain did in the Cod Wars, but be less revisionist once maritime areas are delimited.¹¹

However, a different picture emerges when we split the sample into EEZ and non-EEZ claims (Online Appendix Table A12a) and interact dyadic democracy score with the territorial integrity norms variables (Online Appendix Tables A8 and A12b). The findings are mixed, as the Zacher measure is positive and significant for EEZ claims (Model 1), but negative and significant for non-EEZ claims (Model 2). For dyads that jointly belong to more Pacific settlement treaties, we see a reduction in the chances for EEZ/border-type maritime claims (Model 3), consistent with territorial integrity norms. On the other hand, treaties with territorial integrity clauses have no effect on EEZ (Model 5) or non-EEZ claims (Model 6). In Online AppendixTable A12b, we interact the V-Dem electoral democracy score with each of the territorial integrity norm measures. Here we obtain results that are more consistent with the constraining effects of such norms, even at sea, as the negative and statistically significant interaction terms show that EEZ claims become less likely as territorial integrity norms strengthen (Models 1, 3, and 5). On the other hand, we see no similar constraining effects of the norms on non-EEZ claims (Online Appendix Table A12b, Models 2, 4, and 6). Even if it becomes harder to revise settled maritime borders over time, states are not being constrained in pursuing other types of claims, such as demanding access to areas with fishing or energy resources outside their EEZ areas. However, the types of maritime conflicts that we observe between democracies could evolve over time.

Conclusion

In this study, we evaluate three major reasons why democratic dyads experience many diplomatic maritime conflicts with each other. We assess whether democracies simply have more opportunities for maritime conflict, showing that while coastal dyads are more democratic than landlocked ones, the maritime conflict dyads have even higher average democracy scores. In addition to showing that democratic dyads experience more maritime conflict than nondemocratic dyads using V-Dem high-level democracy indicators, we also examine how domestic interests and executive constraints help to explain these patterns. We find that regimes that face more pressure from domestic interests and have less constrained leaders (e.g., presidential democracies) are more conflict-prone. We also examined whether territorial integrity norms were less constraining at sea, finding that territorial norms have little effect on maritime claims, but that challenges to EEZ boundaries tend to occur in less democratic dyads and that democratic countries become more constrained to challenge maritime boundaries as territorial integrity norms strengthen. Our results both confirm Daniels and Mitchell's (2017) empirical findings with a broader set of democracy measures and provide more insights into why democracies behave differently when contesting maritime areas.

Author Biography

Chase LaSpisa is a Doctoral Candidate of Political Science at the University of Iowa. His research seeks to further explore the effect that natural resources have on international conflict and cooperation, especially in the maritime context, and how natural resources influence regime behavior and social mobilization at all levels of analysis. He graduated from Oklahoma State University in May 2020 with a BA in Political Science, as well as minors in Economics and Russian Language.

Sara McLaughlin Mitchell is a Professor of Political Science and Collegiate Fellow at the University of Iowa. She published eight books and more than seventy journal articles and book chapters. Her areas of expertise include international conflict, political methodology, and gender issues in academia.

Footnotes

References