Three Paradigms of International Judicial Review

1 Introduction

The scholarly interest in international courts has been growing steadily in recent years. As new international courts are created and the output and the influence of existing courts is rapidly increasing, many theories have been developed to explain international judicial review. But international courts are not all the same, and even the same court can change radically over the years. A theory of international judicial review must consider the variety of review possibilities available to international courts.

This chapter suggests that the types of review exercised by international courts could be divided into three separate paradigms. The paradigms discussed here provide a general framework for understanding certain types of judicial behavior. As such, the paradigms can help to make sense of the strategic situation faced by the court by highlighting the most relevant actors and incentives. Every court at every time shares some similarity with every paradigm discussed here, but it can usually be explained better under one particular paradigm.

The three paradigms are:^[1] (1) The Westphalian Paradigm: international courts are primarily coordinating the interaction of distinct sovereign states, (2) the Hierarchical Paradigm: international courts are primarily a mechanism for ordering states to change their behavior in specific ways, (3) the Network Paradigm: international courts help to sustain a vast network of sub-national actors and exert their influence on these actors in various ways. While all paradigms are still useful for understanding international courts today, there is a chronological element to the distinction between the paradigms. Traditionally, the Westphalian paradigm explained most of the behavior of international courts, later the Hierarchical paradigm became increasingly useful, while today, the Network paradigm is the strongest explanation of international courts’ behavior.

Understanding an international court through the lens of a specific paradigm helps to highlight the main ways in which the court has an impact on policy. Part II describes the Westphalian Paradigm and uses it to explain the International Court of Justice (ICJ). Part III describes the Hierarchical Paradigm and uses it to explain the traditional role of the European Court of Human Rights (ECtHR). Part IV describes the Network Paradigm and uses it to explain recent practices of the ECtHR as well as the Court of Justice of the European Union (CJEU), and the Inter-American Court of Human Rights (IACHR).

2 The Westphalian Paradigm

The 1648 Peace of Westphalia is often referred to as the starting point of the modern international law system, a system that is based on the interaction of independent sovereign states. But if every state is viewed according to the traditional realist perspective of international relations, as a unified actor striving to increase its power and influence, what role is there for international law? Jack L Goldsmith and Eric A Posner argued that even under this narrow view of international law, it can still shape the behavior of states. States behave according to international law when it is anyway in their own interest to do so, when international law helps states to coordinate their actions, when the law is maintained through mutual threats to use force, or when one powerful country forces others to comply with the law.^[2] In this simplified setting, international courts are the most useful when they help states to coordinate their actions without deteriorating into conflicts and wars.

Tom Ginsburg and Richard H McAdams wrote about the ICJ as a tool for coordinating states.^[3] They describe two mechanisms that allow the ICJ to guide states towards a peaceful solution:^[4]

1. Creating focal points through cheap talk – by clarifying ambiguous provisions in treaties, the court helps suggest a solution that states can accept.

2. Signaling useful information – the court can make decisions about the legal or factual situation that provide information which helps the states to agree on a solution.

For Ginsburg and McAdams reputational sanctions play only a limited role in solving disputes before the ICJ.^[5] Yet the role of reputation may sometimes be significant due to the interaction between the incentives of the states. If every state is willing to comply conditioned on compliance of the other state, then a reputational sanction for noncompliance means that country A would suffer a loss if it doesn’t comply, and it also knows country B is likely to comply because it expects a similar loss in case of noncompliance. The combined force of these incentives may be enough to make states give up on some of their preferences.^[6]

The ICJ strives to strengthen even more the tendency of states to comply by issuing judgments that practice what may be called ‘Solomonic’ justice – they try to give each of the states involved a fair share of the benefits to avoid displeasing too much any of the parties involved.^[7] When a judgment seems to balance the interests of both parties, there is a greater likelihood that states will accept it, especially if there is some reputational cost for noncompliance.

If one were to break the shell of the state and look at the different actors guiding its behavior, other reasons for the ICJ’s role in settling disputes appear. For example, it is possible that political leaders would actually prefer to compromise and avoid conflict but are afraid of criticism by their own public if they give up on too much of their national interest. The ICJ provides these leaders an excuse for compromising because they can claim they are bound by the ICJ’s decisions.^[8] Naturally, if noncompliance involves a large reputational sanction and if the judgment seems evenhanded and also grounded in international law, the ability of political leaders to compromise without suffering public criticism is improved.

Under the Westphalian paradigm, international courts are not the main heroes of the drama of settling disputes. Rather, powerful states are trying to find a common solution under the shadow of the court. Sometimes, the ICJ can contribute to peace even when it doesn’t do anything. In fact, I have argued elsewhere that a procedure that was not used even once in the court’s history may have nevertheless helped to settle disputes between states.^[9]

This procedure is the possibility of deciding cases according to justice, ex aequo et bono, mentioned in article 38(2) of the ICJ Statute. In order to allow the ICJ to decide a case according to justice, both parties to the case need to agree on that. So far, this never happened. Viewing the choice given to states between a decision according to the law and a decision according to justice as a screening mechanism can both explain why there are no cases decided according to justice and suggest that the existence of the never-used procedure of decision according to justice is nevertheless useful for the peaceful settlement of disputes.

The screening mechanism works in the following way: under the reasonable assumption that decisions according to justice involve lower litigation costs but reach the legally correct result in a lower probability compared to decisions according to law, states that have a weak legal case have a greater incentive to opt for a decision according to justice. A state that hints it would be interested in a decision according to justice signals to the other side the weakness of its case – it believes it has a low chance to win according to the law – so it does not mind to risk increasing the chances for a legally incorrect decision to save on litigation costs. A state that sends such a signal will get a low settlement offer from the other side that perceives its willingness to settle for a small concession. The state will accept that offer and the case would never reach the ICJ. In this way, all the cases in which at least one party would desire a decision according to justice end up being settled out of court and never lead to an ICJ judgment. At the same time, the mere existence of the possibility of a decision according to justice helps states to signal the weakness of their cases and increases the prospects of a peaceful settlement.

Under the Westphalian paradigm then, the ICJ is able to influence the behavior of states even when it doesn’t act at all. States change their actions not because they are ordered to do so by the court, but because it suits their strategic interests, given the options and the possibilities for potential ICJ intervention.

3 The Hierarchical Paradigm

Probably the most famous paradigm shift in international law is associated with the work of Wolfgang Friedmann. Writing in the 1960s, Friedmann describes a shift of international law from a law of coexistence paradigm in which the core of international law was concerned with regulating conflicts between sovereign states to a law of cooperation paradigm in which international law is mainly used to create effective institutions and to promote collaboration between states.^[10]

The Westphalian paradigm of international courts is very much geared towards Friedmann’s law of coexistence paradigm with international courts playing a secondary role in shaping the power struggles between states. In contrast, Friedmann’s law of cooperation paradigm matches the hierarchical paradigm of international courts. Under the hierarchical paradigm, the main role of international courts is to initiate social change by directly ordering states to amend their policies. A good example for a court that suits the hierarchical paradigm is the ECtHR.

The ECtHR issued thousands of important judgments that forced nearly every state in Europe to change some of its policies even when it was dangerous, expensive, or politically inconvenient to do so. Two design features of the court that became mandatory for all member states since 1998 make this possible: (1) the court has compulsory jurisdiction, making states legally bound to comply with its judgments, (2) the court allows petitions by individuals who were victims of violations of the European Convention of Human Rights, thereby ensuring that the court has the opportunity to intervene in all instances involving major violations.^[11] Cognizant of its power, the ECtHR behaves very much like a constitutional court, it does not try to fix one legal problem at a time but rather tries to promote the protection of human rights all over Europe.^[12]

The ECtHR, at least until it started to openly engage with many sub-state actors, as the next Part relates, can serve as a perfect model for the hierarchical paradigm of international courts. Under this paradigm, the role of the court is to order states to change their behavior. Naturally, ensuring compliance with its judgments becomes the primary concern of the court. Although international courts may be tasked with a variety of other goals such as supporting international organizations and legitimizing certain norms and institutions,^[13] if states do not end up doing what the court tells them to do, the court cannot really perceive itself as an effective head of a normative hierarchy.

The European Convention on Human Rights may give the ECtHR the authority to order states to change their behavior, but the Convention cannot by itself supply states with a reason to comply. The ability of the court to coerce states to comply is severely limited. Enforcement is relegated to the Committee of Ministers of the Council of Europe^[14] whose only real stick is the power to kick a state out of the Council of Europe.^[15] This tool was never used until 2022 when Russia was expelled from the Council of Europe following its invasion of Ukraine.^[16] Usually, this extreme measure does not pose any real threat to states.

So why do states often comply with the ECtHR even when doing so is difficult and uncomfortable? The reason lies in the interest of states to preserve their reputation for compliance with international law. A state that disobeys the ECtHR is sending a signal that it would rather serve its immediate interest than suffer the cost of compliance in order to build a reputation as a country that stays true to its obligations. If a country decides to comply, it is sending a signal that it cares about its ability to credibly commit to other states and is willing to pay a price to maintain that ability.^[17] If the ECtHR wants to increase the prospect of compliance with its judgments, it needs to ensure that the reputational sanction on noncompliant states would be as painful as possible.

I have argued elsewhere that the reputational sanction on states that fail to comply with the ECtHR depends on the reputation of the court at the time, which, in turn, depends on the court’s history.^[18] Every time an ECtHR judgment is complied with, the court’s reputation rises. Noncompliance with a higher-reputation court will cause states a stronger reputational sanction, giving states a greater incentive to comply with the court in the future. In contrast, every act of noncompliance would lower the court’s reputation. The court’s reputation measures the expectation that states will comply with its judgments. Therefore, every act of compliance strengthens the belief that the court will be complied with in the future and simultaneously generates a larger reputational incentive to comply with the court, ensuring that this belief will be fulfilled. At every point, the court’s reputation reflects the series of updates caused by the past responses of states to its judgments.

To increase its reputation, the ECtHR has to act strategically. The court should take advantage of the fact that compliance with different judgments leads to a different reputational benefit for the court. When the judgment is more demanding, compliance will lead to a higher reputational benefit, because it proves that states are willing to comply with the court despite the significant costs of doing so. When the reasoning of the judgment exposes judicial discretion, compliance would also lead to a greater reputational benefit because it proves the court can get compliance even when it does not rely heavily on its legal authority.

Judgments that demand more from states, including judgments that give states less deference and do not allow them a margin of appreciation, and judgments that expose judicial discretion, involve a greater risk of noncompliance, but if compliance is secured, they will yield a greater reputational benefit. What the ECtHR should do is continuously walk on the brink of noncompliance, trying to issue judgments that have the biggest risk of noncompliance, but eventually get complied with. The pattern that emerges from this type of behavior is a continuous shift towards more demanding judgments as long as the court’s reputation is on the rise. The court may not issue the most demanding judgment that will still be complied with in every case. Legal constraints, the preferences of the judges, and the need to maintain the support of certain audiences would often lead to restraint. Nevertheless, the trajectory of the judgments should be maintained – an improvement in the court’s reputation will lead to an increase in its demands from states.

To maintain its position at the top of the hierarchy, the ECtHR should follow an egoistic strategy, behaving in ways that are calculated to increase the court’s reputation over time. But with the acquired power comes a responsibility to use it in a way that is both normatively legitimate and beneficial for the international community. The two conditions are not the same, legitimacy inquires whether the ECtHR can justify taking decisions away from democratic states even though it is an unelected professional body. The question of the quality of judgments is designed to ask if the ECtHR can make better decisions than other domestic bodies, such as national courts.

As far as the question of legitimacy goes, the ECtHR should consider if the state could be counted on to represent the people affected by its actions. It was John Hart Ely who famously observed that even democratic institutions cannot be trusted to work all the time, and consequently, they should not be deferred to all the time. When democratic failures occur, for example, when discrete and insular minorities suffer from prejudice that prevents them from forming political coalitions with other groups, the democratic bodies do not really represent their interests. This is where judicial review is justified.^[19]

Ely’s ideas have been used to explain when the ECtHR can legitimately intervene and when it cannot.^[20] Scholars argued that in cases that involve people who either cannot vote or who are a part of a politically weakened minority, there is a greater justification for the ECtHR to intervene in domestic policies. The margin of appreciation granted by the ECtHR should be narrower in these cases, which means the court would be more reluctant to defer to state policies.^[21] Some scholars confirmed that the ECtHR behaves according to the dictates of this normative theory and defers less in cases that involve politically vulnerable groups.^[22]

Arguing that states are not perfect at representing everyone affected by their actions is not very controversial and many would agree that this gives license for the ECtHR to intervene and protect those that would otherwise be unprotected. But suggesting that the ECtHR can actually issue judgments that are superior to the decision-making abilities of the states is a different matter, this argument would probably be resisted by many, especially in light of the evidence that ECHR judges often suffer from low professional abilities.^[23] No one can seriously claim that ECHR judges are more talented or experienced than their domestic counterparts and they are tasked with addressing challenges across all of Europe, not in one country where they spent most of their lives.

The search for a point in which an ECHR judgment can be superior to a judgment of a domestic court must therefore focus on what makes the ECHR superior as an institution. Is positioning the ECtHR at the top of the hierarchy somehow giving it an epistemic superiority over domestic courts? Elsewhere, I argue that it does.^[24] The ECtHR is able to use comparative law as a tool for improving its decisions in a way that national courts are not able to do simply because of their institutional position.

To understand why the ECtHR has this advantage, it is necessary to follow a unique justification for the use of comparative law that was developed by Eric Posner and Cass Sunstein.^[25] They argued that according to the Condorcet Jury Theorem, if many decision-makers make an independent decision on a certain issue, the decision supported by the majority is more likely to be correct than a decision any one of the decision-makers can reach on their own.^[26] If national courts learn from what the majority of the states are doing, they can capitalize on aggregating the wisdom of all the states that decided on the matter. But if every state has a national court that learns from what other states are doing, then the policies adopted by the states are no longer independent – thus, pulling the carpet under the possibility of using the Jury Theorem to justify learning from comparative law.

The ECtHR can avoid this problem exactly because it sits on the top of the hierarchy and gives European states directions that they must all follow. If the states make their policies first and then the ECtHR decides based on what the majority of states have chosen and sets this decision as a rule for all of Europe, the full potential of the Jury Theorem can be fulfilled. Because the ECtHR’s decision learns from the majority of states, it is likely to be a good decision and because all states would then be directed to follow it, it will change policies across Europe in the right direction. National courts cannot reach this outcome on their own because they either learn from comparative law and distort the independent nature of decisions or decide independently and forfeit the decisional advantage of the Jury Theorem.

The emerging consensus doctrine used by the ECtHR is a perfect manifestation of this model of decision-making that simulates the Jury Theorem.^[27] There is empirical support that, at least in some issues, the ECtHR follows the policies adopted by most of the countries in Europe when it applies the emerging consensus doctrine to determine the standards of human rights it demands in its judgments.^[28] Moreover, European countries do not rush to follow each other’s lead on policy issues and instead wait for a judgment by the ECtHR to change their policy choices.^[29] In this manner, the ECtHR can provide states the best possible guidance derived from aggregating the decisions of 47 independent states.

To conclude, the ECtHR secured its position as a court that can order states to change their actions from a superior hierarchical position. It built its reputation by acting strategically and thereby increased its ability to get compliance from the states. To justify this order of things, supporters of the court can maintain that its decisions often enjoy both normative legitimacy and the epistemic capacity to reach better decisions than national bodies. All this presents the ECtHR as a primary example for how international courts behave under the hierarchical paradigm. The challenges and opportunities that this paradigm affords are the ones that the ECtHR faced throughout most of its existence.

4 The Network Paradigm

The Westphalian paradigm and the hierarchical paradigm have something in common: at the end of the day, the actors that determine what will happen in the world under both paradigms are unified states. Under the Westphalian paradigm these states mostly react to each other and under the hierarchical paradigm they respond to directives from the court, but in both cases the state acts as one body and the courts have no interaction with the individuals that compose it.

The decisions of unified states are made by senior politicians or diplomats: a narrow elite that is supposed to represent the interests of the state as a whole. Recently, there is a growing realization that the members of this elite are not the only people who matter. There is a lot of important activity going on within the state. Judges, bureaucrats, NGO activists, generals, bankers: they are all players that often set their own agendas. Furthermore, these sub-state actors make connections with their counterparts across national borders.^[30] The resulting international order is best captured by the metaphor of a vast network with countless transnational connections between actors far down the hierarchy of their respective states.^[31]

International courts that operate under the network paradigm accept this picture of the international arena. They realize that every decision they make is going to send shock waves across a sensitive and polycentric network. The ultimate outcomes of any judicial decision are very difficult to predict under these conditions,^[32] but international courts have not reacted to this predicament by abdicating their commitment to social change. Instead, they have embraced the inevitable impact of their actions on the network and adapted to the opportunities presented by this new situation.

The first step in acknowledging the importance of the network surrounding international courts is realizing that their judgments are not always fully complied with. Instead, once the judgment is issued, the network is activated, and a complicated and unpredictable chain reaction is set in motion. While every individual action of the numerous people and organizations that are potentially involved cannot be accurately foreseen, scholars have attempted to raise educated guesses about the larger impact of judicial intervention on society as a whole. This Part reviews some potential implications that the CJEU, ECtHR, and IACHR have on their networks.

Lisa Conant argued that the CJEU (or European Court of Justice [ECJ] as it was then called) does not get automatic compliance with all its judgments. When the court issues a judgment, the governments of the affected countries often try to resist it and fail to comply. If the judgment is going to have any real effect on policy, it must marshal support from the public, interest groups, and various organizations who will fight for the enforcement of the judgment and, if necessary, bring new cases to the court designed to make the same social change. Against the supporters of the court, resistance is likely to coalesce as well. The two opposite camps are going to continue their conflict using all the propaganda, lobbying, and litigation tools at their disposal until some compromise solution is reached.^[33]

Some may view this entire process as a colossal inefficiency. What good is a court if its judgments are not complied with? And what good is it to anybody that millions are spent on a social struggle before the issue is finally settled? The answer to these questions is that a properly functioning democratic society is, by definition, inefficient. Vibrant public deliberation is slow and expensive but the funds, the effort, and the public attention poured into it are not wasted. They are used to make sure that the public is informed about political decisions and is willing to take action to protect their rights.^[34]

Without an informed public, a small interest group can easily control politicians and take over the wealth and the political capital of any country. Scholars have observed that courts are not only a mechanism for settling disputes. They are also stirring public deliberation in a way that makes more people informed and motivated to take part in the political game.^[35] When the CJEU issues a judgment that stirs political activists both for and against it, it turns Europe into a healthier society where information and, in turn, political influence and even wealth are more equally diffused.

The vast network of interested people can therefore be used by the CJEU to initiate social change after it issues its judgments, but it is also increasingly used by the CJEU to take on cases and make such judgments possible. The CJEU can accept cases in three separate avenues: (1) national courts that refer the case as a so-called preliminary reference, (2) the Commission initiates a case based on a complaint by a person or an NGO, (3) the Commission starts an infringement procedure on its own initiative. From 2006, the number of cases started by the Commission started to drop while the number of cases started as preliminary references started to grow. At the same time, from 2012, the number of complaints sent to the Commission has only been growing.^[36] Instead of trusting the Commission to bring cases, the system is increasingly geared towards using countless individuals across a vast European network to start cases in national courts that are later referred to the CJEU. The Commission is also being informed by people in its network who continue to bombard it with complaints.

National courts are a crucial part of the network surrounding the CJEU. As mentioned above, the court relies on preliminary references as its main source of new cases. But not all countries have the same willingness to submit preliminary references to the court. Scholars have argued that it is the smaller economies in Europe that need the CJEU to protect them against pressures from the bigger economies. This may explain why the national courts of Belgium and the Netherlands refer more cases to the CJEU, relative to the size of their population, than do the national courts of bigger countries. Enjoying the support of at least some national courts empowered the CJEU and helped it to face resistance to its actions.^[37]

Other scholars pointed to the role that lower-level national courts served in supporting the CJEU. A court that is not a court of last resort has a choice whether to use the preliminary reference tool or not. This allows it to consider if it wants to send the case to the CJEU or to allow the possibility of an appeal that would go directly to the national court of last resort before a decision is made by the CJEU.^[38] Appreciating that this choice gives them strategic power that they can use whenever they think a decision by the CJEU suits better their preferences than one by their own appellate court, lower courts have been the allies that sustained the CJEU.^[39] In contrast, others have argued that the main reason the CJEU flourished is because powerful actors in its network – governments, courts, and academics – did nothing to stop it for a variety of context-specific reasons.^[40] Whether the network surrounding the CJEU strengthened it by act or by omission, nobody can ignore its influence.

Another, more indirect, form of influence was also provided to the CJEU by European parliaments. These parliaments are trying to protect themselves from judicial review and to do that they adopt methods of reasoning such as proportionality and balancing that are common at the CJEU.^[41] A network that is deterred by the prospect of judicial intervention and consequently changes the way it deliberates and thinks about policy problems is also a way for the CJEU to have an impact.

A more systemic way to think about the network that surrounds international courts is to analyze the characteristics of a particular web of connections using the tools of Social Network Analysis. There are two questions that this methodology can answer: (1) is information likely to spread quickly in the network? (2) would the deliberation over this information lead to revealing the truth or to generating falsehoods?

As far as dissemination of information goes, Social Network Analysis recommends a network structure that can spread information quickly to the entire network. This structure is known as ‘dissortative mixing’ – a network in which the ‘core’ (those with many connections) and the ‘periphery’ (those with few connections) connect to each other often.

The ECtHR presents a good example of a network with dissortative mixing. It has a core that includes judges, professional staff, and repeating litigators who are all densely connected to each other. But there is also a large periphery of NGO activists, applicants, and various state bureaucrats that sometimes come into contact with this densely connected core. Information about the court’s judgments passes quickly between the core and the periphery and reaches every part of the network. This way, the principles and values that guide the court spread like wildfire across Europe and have a lasting effect on an enormous community of people.

Once ideas are entrenched enough in the European network, they can also have an impact on people outside of this network and way out of the jurisdiction of the ECtHR. The ECtHR’s decision in Dudgeon v United Kingdom^[42] holding that the criminalization of homosexual sex violates the European Convention on Human Rights was cited and probably had an impact on the American Supreme Court in Lawrence v Texas,^[43] which ruled that outlawing sodomy is unconstitutional. Similarly, the ECtHR decision in Ireland v United Kingdom^[44] that certain interrogation techniques constitute inhuman and degrading treatment was cited and clearly influenced the Israeli Supreme Court, which decided similar physical means of interrogation were illegal.^[45]

The question of whether the numerous nodes in the network that surround the ECtHR are good at processing the information supplied by the court is a trickier one. Can the network ensure that the knowledge of its members about the factual and legal situation is at least as good as what can be derived from the judgment itself or even better? Social Network Analysis calls networks, in which the participants are collectively critical of the information that flows in them and lead to more and more accurate assessments, a ‘bandwidth network’. The opposite is an ‘echo network’ where the biases of all the members of the group tend to grow in the process of communication and end up producing false information.^[46]

Is the ECtHR embedded in a bandwidth or an echo network? To answer this question, I conducted an empirical study on a new website that was designed to increase the collaboration between the ECtHR and its network in order to secure compliance with the court’s judgments. The website is run by the Committee of Ministers, and it publishes reports submitted by NGOs that document noncompliance of European states with the ECtHR’s judgments.

Four years after the website was launched, it already published reports by more than two hundred different NGOs, many of which collaborated with each other and submitted joint reports. My study revealed that the NGOs consistently focused their attention on the most severe violations and the most legally-important ECtHR judgments.^[47] Whatever was going on inside the network, it seemed to process information well, pointing NGOs again and again to cases where their intervention is really necessary. The evidence seems to suggest that the network of NGOs surrounding the ECtHR is a bandwidth network.

To generalize these findings to other international courts, the most pressing question is what turned the NGO network I examined into a bandwidth network? What are the characteristics of a network that is good at processing information? Three factors seem particularly important:^[48]

1. The NGOs in the network were mostly independent from one another. This decreases the risk that an NGO will be subject to pressure by others that will lead them to change their original assessments and result in a cascade of information that can easily lead to falsehoods.^[49] People can follow others because they fear for their reputation or because they try to learn from other people’s choices. Either way, without independent decisions, the collective can soon spiral out of control.^[50]

2. NGOs in the network were very different from one another. They had activists with different backgrounds and different skills. Their size varied significantly. With such diverse organizations, there was a greater propensity for different views within the network. Studies on group deliberation reveal that such different views could balance each other instead of pushing existing biases within the groups to be more and more extreme.^[51]

3. Every NGO in the group has a specific agenda that is transparent and known to other organizations within the group. Experiments have shown that this quality prevents deliberation from pushing the group to biased extremes.^[52]

Not every international court is privileged like the ECtHR to have a network that both spreads information quickly and efficiently and processes it well to promote the truth and avoid falsehoods. A court that possesses these characteristics has a tremendous asset that it can use to impact society. Such a court can focus on the influence it gains through its network and less on the implications of the direct commands it gives to states or on its ability to resolve disputes between states. In other words, courts that are positioned within the right sociological framework can perceive their role more easily as falling under the network paradigm.

Another reason that motivates international courts to gravitate towards the network paradigm is less enviable in nature. If a court is facing prevalent noncompliance, it should realize that it cannot lead to good outcomes simply by ordering the states to do certain things. It must reorient its activities towards facilitating social change gradually by affecting the network that responds to its judgments. The IACHR is a good example of such a strategic choice.

The IACHR has been facing very low compliance with its judgments.^[53] This led the court to adopt strategic behaviors that seem strange for any court that can rely on regular execution of its judgments. For example, the IACHR often does not close a case when a state admits responsibility for its actions. After the state accepts the blame, the court still conducts days of hearing, collects evidence from witnesses, analyzes the relevant facts, and presents them publicly.^[54] If the focus were on telling states what to do, all this would be a gigantic waste of resources given that the state already admitted its wrongdoing to begin with. But scholars have concluded that the IACHR has another agenda, it is trying to disseminate information through its vast network in the hope of initiating social movements that can slowly change the situation for the long term.^[55] Indeed a court that sets out to make a difference in the world and realizes that compliance is difficult to reach acts rationally when it decides to pursue other avenues of influence.

The IACHR does not limit itself to influence through the deliberations in the courtroom. It can build its network in many other ways. For example, the IACHR has an internship program that allows selected people to get useful knowledge about the system of human rights protection under its jurisdiction.^[56] A similar internship program is run by the Inter-American Commission on Human Rights: the organization that refers cases to the IACHR.^[57] These internship programs were instrumental in building a network of motivated, skilled, and well-connected lawyers that assist the IACHR in protecting human rights in the region.

The IACHR has also been engaged in reforming the justice system of the states under its jurisdiction by ensuring that the proper institutions are in place and even that the local judges get the training that they need. The local legal elite can be an asset for the IACHR if the court continuously builds the capacity of everyone in its network.^[58]

To conclude, in recent years many international courts have been operating to a large extent as social hubs that execute social change primarily through their influence on their surrounding network. This makes the network paradigm instrumental in understanding a lot of the choices made by international courts today.

5 Conclusions

The slow shift towards the network paradigm does not reflect the full current practice of international courts. A lot of what international courts do today can still be explained better under the Westphalian or the hierarchical paradigm. Furthermore, even the practice of a single court at a specific time does not fall precisely within the realm of only one paradigm.

Why are paradigms useful then? Paradigms are a way for scholars to make a first educated guess about the behavior of a court that can later be tested. When a scholar wants to predict how an international court would act in a specific situation, the method recommended here is to first consider which paradigm is the most suitable for explaining this court. The paradigm functions in the same way in which assumptions work in any scientific model, it simplifies reality enough to make it possible to make a hypothesis about the way the court would behave. This hypothesis can later be checked against the facts to determine if it is true or not.

For example, if a scholar wants to explain why the IACHR holds hearings on cases where the state admits guilt, it can start by assuming that the IACHR operates under the network paradigm and hypothesize that the purpose of holding these seemingly redundant hearings is to disseminate information in the hope of instigating social change. If an analysis of the way these hearings are held supports the hypothesis that they are used in this manner, the hypothesis will be confirmed and predictions on how the court would act in specific circumstances can be made based on it. If the hypothesis doesn’t fit the facts, it should be rejected. If this and other hypotheses that concur with the network paradigm fail, maybe the IACHR can be explained better under a different paradigm.

Paradigms are useful as a shortcut for developing educated hypotheses. But paradigms have another crucial advantage: they openly admit that they are nothing more than fiction, assumptions that are instrumental in generating hypotheses and not something engrained in the nature of any international court. There is nothing sacred about the assumption that the IACHR cares more about its network than about securing compliance. This is just an assumption that is used to make testable hypotheses on how the IACHR would behave in a given situation.

Elsewhere, I argued that when legal scholars do not make clear what assumptions they use when they analyze a legal question, they fall prey to a pernicious tendency to imagine legal concepts that somehow capture the essence of legal fields or institutions. To avoid the dangers of conceptual thinking, scholars should always start by defining the assumptions they are using to generate hypotheses.^[59] Starting the examination of an international court’s behavior by choosing a paradigm that seems to best fit the practice of that court can serve this goal.