After years of demands and pressure from domestic civil society groups and the international community, on August 3, 2009, the Malian National Assembly unanimously adopted legislation that would transform the country’s family law. Drafted in concert with experts, non-governmental organizations (NGOs), and international donors, the new family law would mark a significant development for women’s rights in Mali. Among other things, it would raise the minimum age for marriage to eighteen for girls, remove legal provisions requiring a wife’s obedience to her husband, and provide equal inheritance for men and women. Yet, much of the public was outraged because the legislation would not guarantee the recognition of religious marriage, and on August 22, 2009, a public demonstration drew an estimated 50,000 protesters.Footnote ¹ Rather than sign the legislation into force, President Amadou Toumani Touré returned it to the National Assembly, which then consulted with Islamic organizations to revise it.

A new Family Code was finally promulgated on December 30, 2011. It set the minimum age for marriage at eighteen for men and sixteen for women, but it permitted marriage from the age of fifteen with parental consent: Either parent could give consent for a boy to marry, but only the father could do so for a girl. Moreover, clergymen were not required to verify consent. The law also provided that a wife must obey her husband, defined the husband as the head of the household, and protected discriminatory inheritance practices, inter alia. In 2016, the Association pour le Progrès et la Défense des Droits des Femmes Maliennes (APDF) – a Malian women’s rights organization – and the Institute for Human Rights and Development in Africa (IHRDA) – a pan-African organization – filed an application at the African Court on Human and Peoples’ Rights (ACtHPR or African Court) that challenged the new Family Code. The APDF and IHRDA asked the African Court to determine whether the Family Code violated Mali’s international human rights obligations, and, if so, whether they were entitled to legal remedies such as an order to reform the law.Footnote ²

In another – yet entirely different – instance, the East African Court of Justice (EACJ or the East African Court) was presented with a case in which it was asked to determine whether Uganda had violated its treaty obligations, including the protection of human rights, due to the government’s handling of protests. On April 11, 2011, activists comprising opposition politicians and civil society actors began staging “walk to work” protests in response to high fuel and living costs against the backdrop of widespread discontent with President Museveni’s election to a fourth term of office.Footnote ³ The police were informed of the planned peaceful protests, which drew thousands of protesters from a broad cross section of Ugandans.Footnote ⁴ Over the ensuing days, the police – under the direction of the government – declared the protests illegal and responded with excessive force, using tear gas, beatings, and live bullets, and detained, arrested, and charged protesters. The violence attracted international media coverage and widespread criticism. By the end of April, the protests and violence had ended, but at least nine people had been killed. The East African Law Society (EALS) filed a petition with the EACJ on May 31, 2011, asking the East African Court to determine whether the Ugandan government had violated its obligations to protect human rights according to the Treaty for the Establishment of the East African Community and the African Charter on Human and Peoples Rights.Footnote ⁵

In both these instances, international courts (ICs) were presented with a dilemma: Should it defer to the responding state, thereby accepting actions or decisions of the state, or should it not defer, thus rejecting how the state decided to exercise its authority? In the case of the Malian Family Code, the ACtHPR rejected Mali’s decision, finding that the government had misconstrued the relevant human rights law and failed in its international obligations, and ordered the government to amend its law. In contrast, the EACJ dismissed the lawsuit against Uganda for its response to the “walk to work” protests, finding the applicant had submitted inadequate or unauthenticated evidence that protesters were beaten, tear-gassed, or killed, despite the notoriety of the events, and therefore, it could not conclude that the government had contravened its legal obligations. This dismissal amounted to the Court’s acceptance of Uganda’s actions. In short, the EACJ deferred to the state, while the ACtHPR did not. How can we account for the differences in these international judicial decisions? What factors contributed to the EACJ’s decision to defer and the ACtHPR’s decision to not defer?

This book aims to understand why international courts can come to such disparate conclusions, as depicted in these two illustrative examples, about whether to defer to states. The book develops a novel theoretical account of judicial decision-making. It approaches decision-making in terms of deference: whether courts, through their decisions, accept a state’s (or other governing entities’) exercise of authority. I argue that deference is determined by an international court’s strategic space – the scope of possible decisions that could satisfy a court’s legal and political constraints. When a court’s strategic space is broader, the potential for nondeference expands. A combination of structural factors and judicial practices shapes this strategic space by contributing to the credibility of state resistance and the legitimacy of ICs. Overall, five factors are integral to explaining judicial deference due to their impact on a court’s strategic space: formal independence, political fragmentation, past resistance, persuasive argumentation, and public legitimation. Formal independence expands the strategic space because it improves the perceived legitimacy of a court and imposes institutional hurdles, both of which make state resistance less credible to courts. Political fragmentation also imposes institutional hurdles, making state resistance less credible, and thereby expands the strategic space. Conversely, past episodes of successful resistance, enabled by weaker independence and political unity among member states, narrows the strategic space by leading a court to view state resistance as practicable in the future. Overall, these political constraints impact an IC’s strategic space, but ICs can exercise agency over this strategic space through practices of persuasive argumentation and public legitimation. These practices dampen legitimacy deficits and the potential of state resistance, and thus enable more nondeference.

An original empirical analysis of three international courts – the EACJ, ACtHPR, and the Caribbean Court of Justice (CCJ) – supports this argument. The analysis reveals that the EACJ has been consistently more deferential than the ACtHPR and CCJ. Among the three courts, it has the weakest safeguards to independence and its membership is moderately fragmented (as preference heterogeneity among members is low-to-moderate), both of which facilitated an important episode of collective resistance. The combined effect of low formal independence, moderate political fragmentation, and successful past resistance has been a relatively high degree of deference by the EACJ. The EACJ, however, does not always defer to states. Rather, to the extent that the EACJ does not defer, nondeference has been enabled by the Court’s practices of persuasive argumentation and public legitimation. In contrast, the ACtHPR enjoys a broad strategic space because of its high formal independence and significant political fragmentation, and thus, it has been less inclined to defer to states. Unilateral resistance by states, however, in the form of exit contributes to this court’s efforts to couple nondeference with persuasive argumentation and public legitimation and its move toward more deference in recent years. The CCJ enjoys the highest levels of formal independence, while its membership has mostly homogeneous preferences, which together provide it with a midrange strategic space. This closely corresponds with its tendency to not defer, except when states’ preferences strongly align. The CCJ also relies on persuasive argumentation and public legitimation to bolster its strategic space and minimize resistance when it does not defer to states.

The Puzzle: Deference by International Courts

The end of the Cold War was a pivotal moment for international cooperation as it ushered in the consolidation of the liberal international order. As part of this transition, we have witnessed a proliferation of international courts that were to be part and parcel with a rule-based order.Footnote ⁶ No longer are international courts concentrated in the liberal, democratic core of the international community. Today, ICs operate in nearly all world regions and have jurisdiction over states with varying histories of judicial power, democracy, and respect for the rule of law.

States, international organizations (IOs), civil society groups (e.g., NGOs), and private actors regularly petition these courts, asking them to interpret and apply the law, to facilitate the peaceful settlement of disputes, to determine whether states and IOs have complied with their legal obligations, and to provide remedies for wrongs. ICs respond to these pleas in varying ways. For instance, the EACJ found no violations in roughly 54 percent of its merits judgments, while the ACtHPR did in only 18 percent. By several accounts, the Court of Justice of the European Union (CJEU) – previously known as the European Court of Justice (ECJ) – frequently expands legal obligations such that is appears to make law, even though it avoids ordering intrusive remedies.Footnote ⁷ On the other hand, the Andean Tribunal of Justice (ATJ) is typically restrained in its legal interpretations.Footnote ⁸ The International Court of Justice’s (ICJ) often relies on restrictive interpretationFootnote ⁹ and nonintrusive remedies.Footnote ¹⁰ The European Court of Human Rights (ECtHR) operated for nearly a decade before declaring a state had violated its obligations, but gradually its interpretations significantly expanded human rights protections,Footnote ¹¹ even though its remedies remained nonintrusive.Footnote ¹² The Inter-American Court of Human Rights (IACtHR) is known for the way it leaves little discretion to states by adopting extensive remedies.Footnote ¹³ This quick review of several international courts suggests that decision-making by ICs varies and raises several questions. How can we conceptualize this variation? Do judicial decisions vary both across courts and over time? What factors account for variation in international judicial decisions?

These are the questions that motivate this book. I approach them by focusing on a core feature of judicial decisions, namely, deference. Sometimes, international courts defer, whereby they accept a state’s exercise of authority. This acceptance is revealed by rulings that declare a member state has not violated its legal obligations, use restrictive treaty interpretation or limited scrutiny, or deliver remedies that do not intrude on sovereignty. Other times, they do not defer and, thereby, reject a state’s exercise of authority. Such rejection can be seen in court decisions that declare a member state has violated its legal obligations, eschew restrictive interpretations and intensely scrutinize states’ decisions or actions, or deliver remedies that intrude on state sovereignty. In short, I define judicial deference as a court’s acceptance of a state or other governing entities’ (e.g., an organ of an international organization) exercise of authority. This definition builds on a definition proposed by the international relations (IR) scholars Asif Efrat and Abraham Newman and adapts it for international courts.Footnote ¹⁴ Moreover, based on this definition, I operationalize deference along three dimensions of judicial decisions: case outcomes, legal interpretation and reasoning, and remedies.Footnote ¹⁵

The focus on deference distinguishes this book from work that characterizes judicial decisions on an activism/restraint continuum. Shifting the conversation to deference is advantageous for a few reasons. First, activism/restraint is largely in reference to legal interpretation and lawmaking (or lack thereof) by courts. Yet, judicial decisions have other important dimensions of potential interest and through which deference may be visible or practiced, such as remedies.Footnote ¹⁶ In other words, deference, as I conceive of it, is a broader concept and encompasses more aspects of judicial decision-making. Second, judicial activism – and conversely restraint – is a loaded concept. Activism can refer to decisions where a court invalidates another authority’s decision, ignores precedent, engages in lawmaking, or departs from accepted interpretive methods or standards of judging, among others.Footnote ¹⁷ In addition to having multiple meanings, activism is laden with political connotation, and identifying which decisions are activist or not is often subjective. Thus, deference is conceptually cleaner and less subjective or politically charged.

De facto independence is another concept often used to characterize judicial decision-making. While this concept and deference are related, and both refer to judicial behavior, they accord to different units of analysis. Deference is a characteristic of discrete judicial decisions, while de facto independence is a characteristic of a court. I am primarily interested in studying the decisions of courts, for which deference is more suitable. Deference, however, can also be used to describe a court’s case law in the aggregate and is useful for comparative purposes.

The Argument: Strategic Space, Structural Constraints, and Adaptive Practices

The core argument that I advance in this book is that judicial deference is the result of a court’s strategic space. This space is constructed by two structural political constraints – formal independence and political fragmentation – which jointly condition how a court is perceived (i.e., its legitimacy) and the feasibility with which states can collectively resist or curtail it. Courts that lack formal robust institutional safeguards to their independence and have unfragmented memberships (i.e., homogeneous state preferences) are more politically constrained than their peers that have greater formal independence and fragmented memberships; hence, their strategic space is relatively narrow, and they will therefore demonstrate more widespread deference. Without robust institutional safeguards to protect a court, it will appear to be vulnerable to the whims and influence of states or those that wield power over it. State defiance of such a court comes with fewer political costs to politicians, and formal independence also means that institutional opportunities for resistance are greater or more practicable. In this setting, courts perceive state resistance to be more credible or plausible, leading them to exercise more caution. Nondeference is thus inhibited by this political setting, especially if resistance has been success in the past. But, when courts enjoy robust safeguards to independence and its member states are politically fragmented (meaning their preferences are heterogeneous), legitimacy is greater and state resistance more limited. This essentially expands the opportunity to exercise nondeference. In slightly different terms, formal judicial independence and political fragmentation enable less deferential decision-making.

These structural political constraints, however, do not straitjacket a court. Rather, my argument suggests that judicial power is also constructed through agency. ICs exercise some degree of agency over their strategic space by adopting practices that also affect their perceived legitimacy and the credibility of state resistance. In particular, ICs often employ persuasive argumentation, in the form of unanimous decisions or referencing authoritative sources, and engage in public legitimation to boost their legitimacy and disincentivize state resistance. These judicial practices help to build and maintain a court’s strategic space to have an impact on decision-making.

My argument contrasts, yet complements, the view that judicial decision-making is explained by the strength of a court’s support network. Previous literature suggests that courts with strong support networks will demonstrate nondeference, irrespective of states and institutional rules.Footnote ¹⁸ My argument complements this perspective in two ways. First, it sheds light on decision-making when support networks are forming. Second, it assumes that courts themselves play a role in the construction of their power. Aside from the explanations focused on support networks, I also consider perspectives that highlight ideational factors relating to political norms and legal culture.

An empirical analysis of the EACJ, CCJ, and ACtHPR supports the theoretical argument. An examination of these three courts is valuable for a few reasons. First, it offers insights into newer, less studied international courts. As these are regional courts, it also complements a wealth of research on European regional courts, namely, the CJEU and the ECtHR. The empirical analysis therefore sheds light on the extent to which the European experiences travel to other regions operating within different legal, political, and socioeconomic contexts. The EACJ, CCJ, and ACtHPR are embedded within settings of rich legal pluralism, fragile faith and trust in state (or formal) legal order, colonial legacies, and mostly low- and middle-income economies. In addition, these are relatively new courts. Thus, this study expands knowledge on how ICs develop their initial jurisprudence, struggle for recognition, grapple with political constraints, and more during their nascent stages. The cases illustrate how differences in the formal safeguards that are in place to isolate courts from the political control of states engender opportunities for courts to reject the state’s exercise of authority in accord with legal constraints. Divisions among states also differ, which can be leveraged to determine how political fragmentation enables nondeference.

Three main findings come from the empirical analysis. First, there is substantial variation in the formal independence of international courts. While previous literature has focused on other design features of international courts, such as formal powers and access,Footnote ¹⁹ this book introduces formal independence as a design feature and presents original data on the formal independence of twenty-six international courts. A mapping of this data reveals that ICs are not all created alike in terms of their formal safeguards to independence. These safeguards relate to rules on the selection and tenure of judges, the managerial autonomy of courts, and the terms of office for judges. Most variation in formal independence is driven by the former two types of rules.

Second, an analysis of the decisions taken by the EACJ, ACtHPR, and CCJ reveals that the three courts defer in different ways and to different degrees. We can see unique patterns of deference in terms of their case outcomes, legal interpretation and reasoning, and remedies. This threefold approach to assessing judicial deference not only illustrates how courts can rely on different aspects of their decisions to balance legal and political imperatives, but it also offers a more nuanced account of judicial decision-making. Broadly speaking, the EACJ demonstrates substantial deference, while the CCJ shows moderate deference and the ACtHPR minimal deference.

Third, the analysis shows that the observed variation in deference aligns with each court’s respective strategic space and political constraints in support of my argument. The EACJ’s substantial deference corresponds to its pervasive constraints, which arise from low formal independence and moderate political fragmentation. The CCJ has high formal independence that offsets low political fragmentation, providing for intermediate constraints and thus enabling moderate deference. In contrast, the ACtHPR’s subtle constraints, due to formal independence and political fragmentation that are both relatively high, enable its minimal deference. On the other hand, while each court has an emerging support network that is responsible for the lion’s share of litigation, these networks’ access opportunities, strengths, and activities do not systematically vary with the patterns of deference. That is, stronger and more active networks do not consistently correspond with less deference.

The empirical evidence lends strong support to my argument. Yet, the argument is developed in the context of specific conditions. First and foremost, the argument is a story of new international courts and their early years of development. In this context, institutional design has special importance because relationships with states and societal actors are nascent, and thus the reservoir of support that an IC enjoys at its origin depends more on the appearance of constraints, as presented by institutional design. Appearances matter in the absence of past performance. Moreover, support networks may themselves be works in progress early on. New international courts, their rules and legal framework are less knowable to substate and nonstate actors. Courts and their judges grow into their relationship with states. They begin with not knowing exactly how states will react or how well states can collectively coordinate. While early years can be formative, they are not necessarily determinative. First impressions fade and are replaced with stories of success (or failures), and courts’ unique cultures develop. In the end, evolving political, social, and political realities contribute to the development of law and courts, even when considering self-reinforcing processes of path dependency or organizational culture.

Second, the courts that I look at are embedded within similar socioeconomic and ideational contexts that distinguish them from those studied elsewhere (e.g., CJEU, ECtHR). The EACJ, CCJ, and ACtHPR are regional courts whose members are in the Global South and mostly low- and middle-income countries. Socioeconomic considerations may have a range of possible effects related to law and courts, including state legal capacity, opportunities and resources for individuals to litigate, and legal knowledge and consciousness. Previous research, for example, shows that the rule of law and economic development or wealth are closely connected.Footnote ²⁰ Similarly, international development assistance often emphasizes the rule of law. Additionally, the EACJ, CCJ, and ACtHPR share similar legal cultures characterized by complex legal pluralism, often poor societal confidence in state legal order or insecure judiciaries, and enduring legacies of colonialism. If and how my argument extends beyond these contexts requires further investigation.

Third, my argument does not directly address ICs in their interstate jurisdiction. The empirical analysis does not include an examination of judicial decision-making on interstate disputes, and the theoretical argument is developed primarily with thinking about how courts deal with petitions from private (nonstate) litigants. For this reason, I am cautious to not draw conclusions about judicial decision-making on interstate disputes or claims. While I measure the formal independence of ICs with exclusively or primarily interstate jurisdiction – the Appellate Body (AB) of the World Trade Organization (WTO), the ICJ, or the International Tribunal of the Law of the Sea (ITLOS) – I do not make any claims about how their decision-making is affected by this aspect of their institutional design. I leave this to future empirical analysis to determine whether their deference is best conceptualized in the threefold manner that I rely on and whether deference corresponds to their level of formal independence or political fragmentation. I recognize that a court’s relations with states differ somewhat in this context because parties are always states, and therefore dynamics like the relative power of state parties might be an alternative or additional political constraint that affects decision-making in ways not theorized in this book.

Last, I want to be clear that my argument does not ignore how the law itself constrains, or conversely enables, ICs to generate patterns of deference. I assume that it does. This is most obvious in the attention I give to institutional design, which are in fact legal rules that limit state and court behavior. Beyond this, I am attentive to how the law interacts with the political. That said, my primary aim is to identify and understand political constraints.

Research Design

The argument of this book combines a complex set of factors to explain deference and how it varies across ICs and over time. To empirically assess these complexities, I use a nested and mixed methods research design. This approach is beneficial because it exploits the comparative advantages of different kinds of data and types of analysis to enhance the validity of causal inferences.Footnote ²¹ While nesting and mixed methods are widely advocated, they are utilized less often. The research approach used here therefore forges new methodological ground for the study of international courts. In the following, I describe the research design, beginning with how it is nested and then its use of mixed methods.

The research design is nested, meaning that the same case studies are analyzed in both cross-case and within-case comparison.Footnote ²² On the one hand, I draw inferences about the causal factors contributing to deference by comparing the EACJ, CCJ, and ACtHPR. The courts were selected to ensure variation on the key explanatory variables while controlling for confounding factors – making this a most similar systems research design. The EACJ, CCJ, and ACtHPR are “typical” or “pathway” case studies. Methodologists suggest that selecting case studies according to these principles is appropriate for theory-building,Footnote ²³ which is the aim here. The three courts differ in terms of formal independence and political fragmentation, both of which are key explanatory variables considered by the theoretical argument. The EACJ, CCJ, and ACtHPR enjoy different institutional safeguards to independence. An analysis of formal independence among twenty-six international courts supports this position, and it forms the basis of the case selection. As Chapter 3 illustrates, the EACJ has low formal independence, the CCJ’s is high, and the ACtHPR lies between. The memberships of these three ICs also vary in terms of their political fragmentation, especially when considering developments across time. Generally speaking, political fragmentation among the members of the EACJ is moderate but low in the case of the CCJ and high for the ACtHPR.

The case selection controls for confounding factors. In particular, the EACJ, CCJ, and ACtHPR have important similarities in their institutional design, especially as they all feature modes of private access. They also have similarities in the types of legal issues that they tend to decide on: human rights, domestic governance, and rule of law. This is seemingly at odds with their formal mandates and general impressions of these courts. While the EACJ and CCJ are regional integration courts, like the CJEU, and do not formally have human rights jurisdiction, a notable portion of their jurisprudence pertains to human rights, domestic governance, and rule of law issues, like the ACtHPR. Additionally, there are parallels between the socioeconomic and ideational contexts of these courts. Their member states are mostly low- and middle-income countries in the Global South, their legal cultures are pluralistic, and there is a lack of faith in and access to the state’s legal order (to be distinguished from legal order that is rooted in customary or traditional law and dispute resolution). Moreover, the courts were established and operate in postcolonial settings where legacies of the past are enduring. Among other things, colonial legacies contribute to a unique dualism of political norms, whereby ideas of sovereignty, noninterference, and independence are frequently in tension with extant beliefs in regionalism, unity, and cooperation. Last, their newness means that each court’s sui generis legal culture is developing and not yet deeply entrenched. In other words, these similarities present a unique opportunity to control – albeit imperfectly – for alternative explanations.

On the other hand, the book draws causal inferences from within-case comparisons. Within-case analysis entails an in-depth study of each court over time and across legal issues. This part of the research design is essential to showing how political fragmentation, past resistance, and judicial practices contribute to deference. In other words, looking at variations within each court’s jurisprudence allows me to determine whether shifts in political fragmentation, resistance, and changes in judicial practices track with the temporal patterns of deference in each of the three courts.

The within-case analyses provide an additional advantage, namely, that each intensive case study makes an important empirical contribution. The analysis of the EACJ, CCJ, and ACtHPR provides detailed data on their decisions. While more and more research has been produced on these courts, the entire corpus of their jurisprudence has yet to be examined. Rather, researchers have so far focused on more limited aspects of their decisions. For instance, James Gathii has analyzed the EACJ’s decisions in response to opposition politicians, Andrew Heinrich has studied the EACJ’s decisions on governance in Burundi, or Adamantia Rachovitsa has examined the ACtHPR’s jurisprudence on its material jurisdiction.Footnote ²⁴ In contrast, the contribution here is a coverage of all judgments from their first decision up to and through 2020.

The research design also relies on mixed methods, drawing from a mix of data types and methods. Overarching both the within- and cross-case comparisons is a congruence analysis. Congruence analysis differs from both correlation analysis and process-tracing. It relies on deductively positing a relationship between variation on the independent and dependent variables, then examining whether there is evidence of a correlation between these variables.Footnote ²⁵ It also provides (minimalist) theoretical specification and empirical evidence of a causal process in addition to demonstrating a correlation.Footnote ²⁶ In contrast, correlation analysis focuses on correlation, not causation. Put different, correlation analysis treats the causal process as a black box – not to be examined – while congruence analysis treats it as a “grey box” that is partially examined. On the other hand, process-tracing is less focused on correlation and instead focus on completely demystifying the causal process such that it is a transparent box.Footnote ²⁷ As Beach and Pedersen argue, many researchers claim to be doing process-tracing when they in fact are doing congruence analysis.Footnote ²⁸ Thus, congruence analysis maps associations between independent and dependent variables while also revealing aspects of causality.

Congruence analysis is beneficial to research where there is limited previous documentation of a causal relationship and when process-tracing is impracticable.Footnote ²⁹ These considerations apply to this research project. Deference has been minimally studied in terms of a causal relationship, and even less research has speculated on the link between deference and formal independence and political fragmentation. Moreover, there are significant barriers to process-tracing because the researcher cannot sit in the room when judges deliberate and make decisions, deliberations are not documented, and judges legally and ethically cannot divulge details of deliberations or their exchanges with one another. Put differently, this may be an instance where “grey-boxing” is as good as it gets. As Beach and Pedersen argue, congruence analysis is appropriate in such cases.

Causal inferences can be strengthened in congruence analysis when combined with cross-case comparison.Footnote ³⁰ Even better is when congruence tests are clustered, meaning that a battery of evidence is provided to support a proposition. In other words, it is a cumulation of evidence, as opposed to any smoking gun, that enables a “relatively strong overall confirmation of the theory.”Footnote ³¹ For this reason, I rely on a series of tests and data types. First, I produce original data on the decisions of each court. I hand-coded the judgments of the three courts, from their first judgments to those finalized by the end of 2020, along several dimensions (case outcomes, remedies, citations, vote distribution, etc.).Footnote ³² This data is triangulated with information drawn from each of the courts’ judgments, official reports, websites, and public communication. Interview data and secondary research also complement the quantitative data for the within-case and cross-case analyses. Interviews were conducted between 2015 and 2016 in Arusha, Tanzania; and Port of Spain, Trinidad; and additional interviews were conducted virtually in 2023 and 2024.Footnote ³³ Finally, I also rely on secondary cross-national data related to human rights, democracy, rule of law, and so on. In all, the concurrent use of qualitative and quantitative data improves not only measurement validity but also the robustness of the causal inferences.

While details on data and measurement are specified in each of the empirical chapters, I take this opportunity to briefly describe the measurement strategy for the key explanatory factors. To measure deference, I use my data on each court’s decisions to calculate descriptive statistics on case outcomes (e.g., rates at which violations are found) and remedial orders (e.g., the portion of judgments that include intrusive remedial orders). Deference through legal interpretation and reasoning is assessed qualitatively. Formal independence is measured with an original index on formal independence, described in Chapter 3. I measure political fragmentation as the degree of variation in state preferences, where state practices on issues such as human rights, democracy, and the rule of law serve as a proxy for preferences. For this purpose, I use Varieties of Democracy (V-Dem) and Freedom House data.Footnote ³⁴ I use my original data on the courts’ decisions to assess persuasive argumentation by calculating the rates at which each court issues unanimous judgments and descriptive statistics on citation practices. I assess public legitimation qualitatively from the courts’ official reports, press releases, websites, and social media. I make special use of Facebook profiles, extracting all posts from each court’s Facebook page with a web-scraping tool. Text analysis of the posts is used to measure the relative prominence of narratives in the public communication of the courts. Finally, support networks are assessed qualitatively, but also using my data on the courts’ jurisprudence to determine the proportion of legal cases supported by repeat lawyers, NGOs, and similar.

Theoretical Implications of the Argument

This book sits squarely at the crossroads of legal scholarship and political science and speaks directly to a growing body of interdisciplinary research on international courts. It has implications for debates on the performance of international courts,Footnote ³⁵ showing that political constraints structure performance – that is, ICs judge under political constraints.Footnote ³⁶ While I am not the first to come to this conclusion, this book uniquely contributes by demonstrating how formal independence and political fragmentation combine to structure these political constraints. It suggests that future research should revisit how institutional design and state preferences interact to affect IC performance. Similarly, it points to a need to improve our understanding of why substate and supranational actors do (not) engage with international courts and how these actors themselves face political constraints that may limit their capacity to utilize the international judiciary.Footnote ³⁷ In addition, the book has implications for debates about how hierarchy shapes international judicial politics. While previous work suggests that hierarchy – akin to domestic separation of powers – affects international judicial politics,Footnote ³⁸ the case studies reveal impacts of global hierarchies and inequalities (that have little to do with formal authority).Footnote ³⁹ This suggests that future research on international courts, ranging from questions about their origins, accessibility, and practices, might benefit from a closer consideration of the role of global hierarchies and inequalities.Footnote ⁴⁰

The book also contributes to the literature on international organizations in contemporary world politics.Footnote ⁴¹ Two dynamics appear to have new effects on IOs today – a heightened public awareness and salience of IOs and the politics of noncooperation and contestation (e.g., backlash, pushback, delegitimation, competitive regime creation). This book integrates these dynamics into its theoretical framework. Drawing from scholarship on the legitimacy of IOs as well as IOs and public opinion,Footnote ⁴² the book offers an account of IO decision-making that considers how IO legitimacy and public perceptions affect the relationship between IOs and states and thus IO decisions. Additionally, the book contributes to research on IO responses to noncooperation and institutional resilience.Footnote ⁴³ While recent research provides insights into the causes and consequences of noncooperation and contestation,Footnote ⁴⁴ we know less about IO responses. This book suggests that institutional resilience amid backlash and resistance derives – at least partially – from their modifications to decision-making and adaptive practices.

Structure of the Book

The remainder of this book develops the theoretical framework and empirical analysis. The theoretical framework is the focus of Chapter 2. Chapters 3–6 empirically evaluate this framework.

The theoretical framework is presented in Chapter 2, which proceeds in three steps. It first specifies the key concept of interest (or dependent variable) – judicial deference. After describing how I conceptualize and operationalize judicial deference, the chapter outlines two main theoretical explanations for judicial deference. The first explanation, which forms the basis of my argument, focuses on the strategic setting in which ICs operate. It maintains that a court’s strategic setting is shaped by formal judicial independence and political fragmentation. Together, these two factors share a negative relationship with deference as they reduce the likelihood of state resistance and improve IC legitimacy. Put differently, deference declines as independence and fragmentation increase. However, ICs can also exercise agency over their strategic space by adopting practices in the form of persuasive argumentation and public legitimation. These practices also affect an IC’s strategic space by way of the credibility of state resistance and the court’s legitimacy. The second explanation focuses on the role of judicial support networks and suggests that ICs are less deferential when they have strong support networks. The chapter describes each of these theoretical logics and concludes by identifying observable implications for each explanation.

Chapter 3 asks whether international courts are designed with institutional safeguards to preserve independence. It first conceptualizes formal independence as the formal rules that aim to safeguard autonomous judicial decision-making. Based on this definition, the chapter then develops an original measure of formal independence, which is used to estimate the formal independence of twenty-six ICs from 1945 to 2015. Next, the chapter compares formal independence across these ICs and reveals significant variation in formal independence. It shows that rules pertaining to the selection and tenure of international judges as well as the managerial autonomy of ICs are the sources of most of the observed variation.

Chapters 4–6 examine the judicial decision-making of the East African Court of Justice, the Caribbean Court of Justice, and the African Court, respectively. Each of these chapters begins by situating the courts within the context of their historical origins and surrounding political norms and legal culture. The chapters then describe deference as it is revealed by case outcomes, interpretation and reasoning, and remedies of each court. Following the description of deference, the chapters consider the extent to which deference aligns with the strength and activity of the courts’ support networks. Each chapter then analyzes the court’s formal independence and political fragmentation, followed by the practices that help to build or maintain its strategic space. These analyses focus on drawing out the relationship between each of these factors and each court’s distinct patterns of deference.

These chapters reveal three varieties of deference and related constraints. The EACJ is best described as having substantial deference due to its pervasive constraints, which result from low formal independence and medium political fragmentation. On the other hand, the CCJ has been moderately deferential as a result of its intermediate constraints. While the CCJ enjoys high formal independence, political fragmentation among its member states is low, and thus, the CCJ’s strategic space is especially affected by the degree of homogeneity of state preferences on the matters before the Court. Finally, the ACtHPR is characterized by minimal deference because of its more subtle constraints, which derive from having generally high formal independence and high political fragmentation.

Chapter 7 summarizes the empirical findings on the EACJ, CCJ, and ACtHPR by comparing the three international courts and reviews the core results of the within-case comparisons. It also outlines the book’s implications for interdisciplinary research on international courts and IR literature on IOs in contemporary world politics.