Explaining State Adjudication Choices: Limits on Focusing on Visible Processes

In “Judicialization of the Sea: Bargaining in the Shadow of UNCLOS,” Sara Mitchell and Andrew Owsiak make an important contribution to the literature that considers if and how international courts influence the behavior of states beyond the process of adjudication. Their analysis shows that states that declare an Article 287 choice for dispute resolution when signing the UN Convention on the Law of the Sea (UNCLOS) are more likely to solve their conflicts in peaceful ways, compared to signatories who do not, suggesting this relationship is not epiphenomenal. Their finding provides us with yet another piece of the puzzle in understanding how legalization and judicialization shape the behavior of states. We still, however, have much to learn about how states interact with legalization and adjudication. This essay argues that while the article's statistical analysis adds to the evidence that judicialization influences state behavior, the uniqueness of UNCLOS makes it doubtful that we can replicate this finding for other international courts and tribunals. In addition, while the statistical analysis suggests that states that make Article 287 declarations behave differently in the face of maritime conflict, we do not really know why states made the choice to declare a preference for adjudication. Finally, given that states strongly prefer negotiations to using UNCLOS formal adjudication mechanisms, we need to understand better the path to adjudication or arbitration.


Áslaug Ásgeirsdóttir*
In "Judicialization of the Sea: Bargaining in the Shadow of UNCLOS," Sara Mitchell and Andrew Owsiak make an important contribution to the literature that considers if and how international courts influence the behavior of states beyond the process of adjudication. Their analysis shows that states that declare an Article 287 choice for dispute resolution when signing the UN Convention on the Law of the Sea (UNCLOS) are more likely to solve their conflicts in peaceful ways, compared to signatories who do not, suggesting this relationship is not epiphenomenal. Their finding provides us with yet another piece of the puzzle in understanding how legalization and judicialization shape the behavior of states. We still, however, have much to learn about how states interact with legalization and adjudication. This essay argues that while the article's statistical analysis adds to the evidence that judicialization influences state behavior, the uniqueness of UNCLOS makes it doubtful that we can replicate this finding for other international courts and tribunals. In addition, while the statistical analysis suggests that states that make Article 287 declarations behave differently in the face of maritime conflict, we do not really know why states made the choice to declare a preference for adjudication. Finally, given that states strongly prefer negotiations to using UNCLOS formal adjudication mechanisms, we need to understand better the path to adjudication or arbitration.

The Uniqueness of UNCLOS
Currently, there are twenty-four international courts and tribunals operating. Four of these are global bodies, including both courts available to signatories of UNCLOS-the ICJ and the International Tribunal for the Law of the Sea (ITLOS). International courts have issued more than 37,000 rulings as of 2011, with 88 percent handed down after the end of the Cold War in 1999. 1 The contribution of international adjudication or arbitration to international affairs is not trivial, and it will likely continue to increase.
Seeking dispute resolution through international courts and tribunals means states have reached an impasse in diplomacy. Agreeing to adjudication or arbitration signals that states are willing to cede sovereignty and autonomy over the outcome of the dispute to a court or tribunal of their choice. 2 Therefore, understanding how states interact with, and are limited by, international courts and tribunals is a key question for scholars of international affairs.
Under UNCLOS, states must solve disputes using peaceful means, and if negotiations fail, states can opt for one of four adjudication methods. In addition to the ICJ and ITLOS, states can also choose one of two arbitral tribunals. 3 Mitchell and Owsiak's key contribution is that states that file Article 287 declarations when they join UNCLOS, "have fewer maritime claims, more peaceful negotiations, and less need for judicial dispute settlement." 4 This finding adds to our understanding of how judicialization of international affairs can alter state behavior, in this case signing UNCLOS, and declaring an Article 287 preference. It is unlikely though that we can extend this finding to other international courts and tribunals, as UNCLOS is unique in allowing states to choose among different dispute resolution mechanisms.
Although UNCLOS is the most comprehensive international agreement regulating the oceans, it still has its limits. For example, it is primarily a resource allocation treaty: negotiated before the advent of environmental agreements, it created 200-mile Exclusive Economic Zones (EEZs), giving states full rights to utilize resources in these areas, significantly increasing actual and potential resource wealth. This served as a key reason for its wide acceptance, as all coastal states significantly expanded their resource base.
What hampers a better understanding of state engagement with UNCLOS is that we do not yet understand why some countries opted to state a preference under Article 287, while others did not. Additionally, we do not understand why the states that did declare a preference chose the specific adjudication mechanisms they did. Presumably, states made this decision based on current realities around their maritime affairs, as well as future expectations of conflict, but we do not really know. Could their choice have been a part of a broader national strategy toward international courts? Were the politicians who made the decision on behalf of their state of a particular ideological bent? Having better understanding of the "why" would help us interpret Mitchell and Owsiak's results in the context of other international courts and tribunals.

What the Statistical Analysis Cannot Tell Us
The debate over whether or not judicialization of international relations changes how states actually negotiate is a lively one among international relations scholars, along with the ongoing conversation about how to bridge the historical divide between those who study international law from a legal vis-à-vis a political science perspective. 5 In addition to thinking about international law in different ways, legal scholars and political scientists utilize different methodologies, with political scientists often embracing statistical analysis in answering their research questions.
Mitchell and Owsiak analyze what is arguably the most comprehensive data on maritime conflict, the Issue Correlates of War data. This dataset identifies all contentious maritime claims , including those that involve settlement of maritime boundaries in the case of adjacent or overlapping EEZs, access issues arising from resource extraction (fisheries and oil) or navigational rights that have arisen between two or more countries. 6 It is the most thorough dataset in identifying observable and ongoing conflicts we have, something that is often difficult to discern. As such, this dataset allows for systematic analysis of conflict across time and issue areas.
Different types of research questions warrant different methods; statistical analysis is but one way of doing good research. Statistical analysis is appropriate when the question centers the average behavior of states, and where there is enough of the appropriate quantitative data to answer questions posed. More tightly focused research questions around processes, and exceptions to behavior identified through statistical analysis, for example, warrant other methods of analysis, such as careful reading of judicial texts, interviewing participants in dispute settlement processes, or multi-method research to hone our understanding of the impact of judicialization beyond the law of the sea.
Analyzing a large amount of data using theory-driven hypotheses testing allows scholars to identify patterns in data that are often too voluminous and complex to analyze in any other way. This is certainly the case with Mitchell and Owsiak's research. In approaching statistical analysis, scholars develop their hypotheses for testing using existing literature, and deduction to state clearly the expected relationships between the behaviors of interests to the researcher. In considering the hypotheses of interest, scholars think through alternative explanations, including variables that help decide which hypotheses are supported by the analysis and which are not.
Statistical analysis is by its nature correlational-it does not confirm causality. Getting at causality in research on international affairs is difficult. In this respect, I agree with Ben Appel's contribution to this symposium. 7 That said, the more evidence we have for things being correlated in predictable ways, the more certain we can be in theorizing how judicialization influences state behavior, and to generalize this to other types of international courts and tribunals.
Statistical analysis should also allow researchers to generalize our findings beyond the research question at hand, and should make prediction possible. Using Mitchell and Owsiak's findings, however, to extend the analysis to adjudication mechanisms beyond the law of the sea is difficult because of the uniqueness of UNCLOS and its conflict resolution mechanism, discussed earlier.
In the quest to identify central tendencies in large datasets, statistical analysis can obscure interesting details among pairs or groups of states. Analysis of large datasets, focusing on carefully crafted hypotheses, leaves out details that can tell a more nuanced story of how a subset of states interact with each other over time and across issue areas. Mitchell and Owsiak's article, for example, tells us nothing about the path to adjudication, which can often be long and circuitous, and it does not consider the outcome of negotiations, adjudications, or arbitrations. Finally, their analysis does not tell us anything about how states complied with settlements, and if they did not, to what extent the violation could be enforced. In order for us to fully understand how states interact with UNCLOS, we need different types of analysis to fill in gaps, analysis that, for example, traces conflict processes over time, and case studies or comparative analyses of how states interact with multiple international agreements.

Studying Formal Adjudication vs. Negotiations
As a scholar of cooperation, who has written about negotiations around shared fish stocks, as well as the settlement of maritime boundaries, where the vast majority of conflicts is settled peacefully, I think we often overemphasize the possibility of militarized conflict in maritime spaces. Very few states have the capacity to engage in sustained military actions on the ocean, and most conflicts either remain unresolved or are resolved through diplomacy-and in rare cases through the use of international courts and tribunals.
Legalization clarifies both rights and obligations. UNCLOS is unique among international agreements in giving states rights to extend their EEZs, while guaranteeing freedom of the seas. These two key features of UNCLOS are also the two areas where we see the most conflict: how and where to draw maritime boundaries where EEZs overlap, and around resource extraction.
States rarely use third party adjudication in solving disputes, either under the law of the sea, or other international agreements. Formal adjudication is costly, requiring well-trained lawyers, and a knowledgeable bureaucracy, and using adjudication increases uncertainty about the outcome, which states are loath to do. As Mitchell and Owsiak note, states use binding resolution in only 3 percent of all peaceful efforts, with the real percentage probably being lower, given that we do not know the universe of conflicts among states. This is of course similar to what we see in domestic legal disputes, with states settling the vast majority of disputes out of court.
It is attractive to study court cases, as they are visible and informative to researchers, because of the materials submitted to judicial bodies for them make decisions. Other international conflicts often remain invisible to researcher. In the absence of formal and public statements of conflict, how do we know a dispute is brewing? Similarly, in the absence of a formal and public agreement to end conflict, scholars do not always know if and how states solved a particular dispute. Therefore, while Mitchell and Owsiak have public statements indicating conflict, it is not clear if their data encompasses all conflicts among states that fall under UNCLOS.
Not all conflicts escalate, sometimes states agree to disagree, and sometimes they reach a solution that is not widely shared. States may have long-standing disputes where tensions sometimes escalate, but where one or both actors are not interested in solving the issue. An example of this is the designation of the Northwest Passage, in which Canada argues that the area are territorial waters, while the United States argues that they are international waters. Since the early 1980s, the two countries have agreed to disagree, and while tensions sometimes escalate, it is unlikely that Canada and the United States will seek to solve this conflict formally.
Another issue to ponder is if states that make Article 287 declarations signal commitment to peaceful resolutions, one that sends an additional signal beyond the signal all states send by signing and ratifying the agreement. Mitchell and Owsiak's analysis suggests that is the case, but I would argue that in order to answer this question more definitively, we need a systematic understanding of why these forty-nine states decided to declare this commitment. Do states make these kinds of commitments because of current or future strategic considerations, or because of domestic political or economic factors? As discussed above, we do not know enough about how and why states decided to make Article 287 declarations in order for us to be sure. In addition, we do not have a good understanding of how likely the signatories are to have conflict with each other. Focusing on the settlement of maritime boundaries for example, and looking at the list of states who decided to make an Article 287 declaration, how many of these pairs of states could realistically have a conflict around settlement of overlapping or adjacent EEZs?
While we have seen an increase in binding conflict resolution mechanisms in international agreements, we still need to understand further how states comply with decisions, and in instances of non-compliance, what enforcement is possible. Conflict resolution at the international level relies on states themselves deciding to obey decisions. If that does not happen, we do not really have good mechanisms for enforcement. The area of international trade is a notable exception, where rulings by the World Trade Organization's (WTO's) and the North American Free Trade Area's dispute settlement mechanisms can be enforced by the victor through the levy of import restrictions, which changes the cost of traded products. UNCLOS has no such provision, instead relying on individual states to obey negotiated or adjudicated conflict resolutions.
At the time of UNCLOS opening up for signature in 1982, the agreed-upon binding dispute resolution mechanism was unique in both its "bindingness," and in the ability of states to choose to declare upon signing the treaty which of the four conflict resolution mechanisms they preferred. It is worth noting, however, that very few states have issued a choice-and states rarely use the dispute resolution. Given this flexibility, why has such a small number of states bothered to issue this choice? As noted in Mitchell and Owsiak's article, of 164 signatories of UNCLOS, only forty-seven states (28.7 percent) have exercised this option, but it is worth mentioning that states party to the treaty can only use its dispute resolution processes to solve conflict, leaving out three other uses for courts and tribunals, namely, enforcement, administrative review, and constitutional review. 8 As mentioned, states rarely use adjudication in settling conflicts that fall under UNCLOS. When they do, adjudication is primarily used for two types of cases: the return of seized ships (eighteen cases in all), which mostly go to ITLOS and maritime boundary disputes (twelve cases), eight of which have been heard by the ICJ. 9 By comparison, states have brought more than 600 disputes to the WTO Dispute Resolution Process since it began accepting cases in 1995, resulting in more than 250 rulings.
Official proceedings mean that scholars have access to processes and outcomes to explore questions associated with choice of venues, process, and procedures, but the fact remains that the vast number of international interactions takes place under the radar, and behind closed doors. For example, states have settled more than 90 percent of maritime boundaries via bilateral negotiations, and we know very little of how states reached specific decisions around where to draw boundaries. 10 This lack of information also holds true for many other bilateral and multilateral negotiations, and arbitration proceedings conducted away from the public eye. There is often a great deal at stake in not negotiating in public, as doing so can inflame existing tensions and derail agreements. This level of secrecy obscures a deeper way to understand international interactions: if we cannot observe the processes, procedures, and outcomes of the vast majority of international interactions, what are we missing?

Conclusion
It is exciting to see more attention paid to UNCLOS, its functions, its shortcomings, and its impacts. Mitchell and Owsiak's article adds to our understanding of how agreeing to a specific binding conflict mechanism signals different state behavior around maritime issues. As the authors acknowledge, they focus primarily on the impact of declaring a choice for adjudication when signing the treaty, leaving to other researchers to explore how states use UNCLOS to mitigate maritime conflicts.
At the same time, there are worrying trends when we look toward the future of UNCLOS and its dispute resolution mechanisms. First, powerful signatories have recently disagreed with rulings and refused to be bound by them, potentially undermining the binding conflict resolution mechanisms available under UNCLOS. This happened with two recent rulings: the 2015 Permanent Court of Arbitration (PCA) ruling in the Arctic Sun case, where Russia does not acknowledge the illegal seizure of the vessel, and the 2016 PCA ruling in favor of the Philippines in the South China Sea, which China does not acknowledge. Beyond these trends, areas beyond national jurisdiction are still not governed effectively. For example, Illegal, Underreported and Unreported fishing, often carried out by ships registered with flags of convenience, has been a difficult problem to tackle. Finally, the treaty cannot really tackle other types of criminal behavior on the oceans, such as piracy, smuggling, or illegal transport, behaviors that impact the safety and health of the seas.