1.2.1 Compliance as Politics and Enforcement: The Legal Scholarship

The question of why state agents change their behaviour in response to statements that international lawyers consider authoritative, final, or binding used to elide international legal scholars. Over much of the twentieth century, international lawyers concentrated their efforts on the expansion of judicial dispute settlement, without much heed to the question of compliance. One author has observed that major international law textbooks traditionally did not even include entries for ‘compliance’ or ‘enforcement’.Footnote ²⁰ A survey of the literature on the topic indicates that the most widely used definition for compliance – ‘a state of conformity or identity between an actor’s behavior and a specified rule’Footnote ²¹ – does not have its roots in international legal thinking but in the literature on international relations.Footnote ²²

As regards lawyers and legal scholars, they have traditionally set aside the question of compliance, usually taking one of three paths.Footnote ²³ Many devoted their attention to the process leading to binding judgments, assuming for the purposes of their scholarship that compliance should follow essentially because the tenet pacta sunt servanda so requires. This approach can be associated with the idea that international law constitutes a form of ‘compelling morality’ for statesFootnote ²⁴ and, especially once clarified by an adjudicator, triggers a ‘compliance pull’,Footnote ²⁵ or ‘propensity to comply’,Footnote ²⁶ based on ‘legal legitimacy’.Footnote ²⁷ This view is largely echoed in a statement by the PCIJ in Chorzów Factory, repeated multiple times by the International Court of Justice (ICJ): the Court ‘neither can nor should contemplate the contingency of the judgment not being complied with’.Footnote ²⁸ Among those who sought to justify the assumption of states’ inclination to comply with a rule once it is clarified, perhaps the most striking statement came from Hersch Lauterpacht: ‘The moral and political effect of the judgment of a judicial tribunal is so tremendous that its legal finality has a tendency to approach actual finality’.Footnote ²⁹

Others have acknowledged that problems of non-compliance, including voluntary non-compliance with clear, adjudicated commitments, exist, while claiming that these problems are political rather than legal. The ICJ itself stated, in Northern Cameroons, that ‘the use which the successful party makes of the judgment is a matter which lies on the political and not on the judicial plane’.Footnote ³⁰ Shabtai Rosenne argued that ‘both compliance with a decision and the enforcement of a decision is in each case a political operation’.Footnote ³¹ Laurence Boisson de Chazournes and Antonella Angelini asserted that ‘the enforcement of judgments … acquires a political rather than a legal character’.Footnote ³² Constanze Schulte concurred that ‘the efficacy of the post-adjudicative phase will not be determined by another judicial examination, but by immediate political action’.Footnote ³³

As regards the ICJ, this view aligns with the conclusion of the Committee of Jurists that drafted the Court’s Statute. The ‘execution of the decisions of the Court’, the Committee reasoned, ‘was not the business of the Court itself’ but ‘concern[ed] rather the Security Council’ of the UN.Footnote ³⁴ This view is reflected in subsequent ICJ statements regarding compliance. In Haya de La Torre, responding to a request for clarification regarding the implementation of a prior ruling, the Court stated that a judgment entails for the parties ‘only the obligation of compliance therewith’, and ‘it is not part of the Court’s judicial function to make … a choice’ among the various paths available to the parties following its rulings. Such a choice ‘could not be based on legal considerations, but only on considerations of practicability or of political expediency’.Footnote ³⁵ This echoes the approach adopted by Edvard Hambro, perhaps the first among modern scholars concerned with implementation of adjudicators’ rulings, whose conclusion was that states must ‘execute judgments in good faith’ but are otherwise ‘free with regard to execution’.Footnote ³⁶ Perhaps not coincidentally, Hambro was the Registrar of the Court during the Haya de la Torre case.

A third strand of scholars sought to follow the traditional domestic law approach to the issue of compliance: to examine, or develop, means to ensure ‘implementation’, ‘enforcement’, or ‘execution’ of judgments. Michael Reisman pointed out that the mere reference to ‘enforcement’ should not serve to shield questions of implementation from legal considerations. The distinction between adjudication and enforcement, he argued, ‘has tended to put the latter into a nonlegal category … [and] to insulate the crucial problem of the realization of authority pronouncement from lawyers. By implication, courts and lawyers are not concerned with postadjudicative processes’.Footnote ³⁷

Discussion of such post-adjudicative processes, however, is usually limited by the notion that international adjudicators lack control over material resources.Footnote ³⁸ Rather than seeking judicial responses at the international level, scholars have often looked to domestic courts and political bodies for means of enforcement. A 1960 article by Oscar Schachter provided the tone for much of the subsequent scholarship. Schachter divided the ‘remedial measures’ available to a successful party, in case of non-compliance with a judicial decision, into two branches. First were ‘methods of enforcement by the successful party’, including diplomatic and economic pressures, unilateral attachment of property belonging to the debtor state, enforcement through the debtor’s municipal courts, and use of armed force. Second was ‘enforcement through international organization’.Footnote ³⁹ This collective enforcement involved essentially recourse to the UN Security Council, with a subsidiary role for other UN bodies and UN agencies. This enforcement, then, was essentially a matter for states and international organisations, not for international adjudicators.Footnote ⁴⁰

The idea that ICs should themselves be concerned with compliance, which would gain traction from the 1990s, was perhaps first expressed in 1981 by Roger Fisher. In Improving Compliance with International Law, Fisher proposed not recourse to sanctions (‘the basic technique … of the criminal law’) but ‘legal machinery’.Footnote ⁴¹ First, international law needs ‘ways of obtaining authoritative determinations with respect to alleged noncompliance’.Footnote ⁴² Second, there must be means of assessing ‘compliance with an authoritative decision … where there has been a dispute’.Footnote ⁴³

Fisher divided compliance into ‘first-order compliance’, that is, states’ overall compliance with international commitments, and ‘second-order compliance’, that is, compliance in situations where there is disagreement regarding what the law requires. As Section 2.2 will clarify, there is little dispute that overall state compliance with international law is high. But ‘the weakness in the international legal system lies in its lack of an orderly and lawful way of coping with alleged departures from what the law provides’.Footnote ⁴⁴ Thus, he concluded, the key objective of those seeking to improve compliance with international law should be to develop the set of responses to the eventuality of non-compliance.Footnote ⁴⁵

Assessing what legal responses to non-compliance are available and may improve compliance, however, requires understanding what motivates states to comply with their commitments in the first place. This question, often set aside by traditional international lawyers focusing on adjudication,Footnote ⁴⁶ has been examined in depth by a literature adjacent to traditional legal scholarship – ‘international law and international relations’. This literature examines the role and application of international law through an interdisciplinary lens, adopting concepts, methods, and language from economics, political science, psychology, or sociology, and applying these disciplines’ insights to ascertain and explain the influence of norms on state behaviour.

1.2.2 Compliance as Behaviour: The International Relations Scholarship(s)

The study of the motives for state compliance with international law is often traced back to Louis Henkin’s 1968 book, How Nations Behave.Footnote ⁴⁷ Henkin’s project was to bridge the gap between international lawyers’ view of international law as something to be complied with at all costs and the perception of international law as largely irrelevant for state behaviour, popular in the ‘realist’ strand of international relations scholarship inaugurated by Hans Morgenthau. In his 1948 book Politics among Nations, Morgenthau had summarised the realist attitude to norms with a quote: ‘You may cover whole skins of parchment with limitations, but power alone can limit power’.Footnote ⁴⁸

Henkin’s central claim was condensed in what came to be known as the Henkin formula: ‘[i]t is probably the case that almost all nations observe almost all principles of international law and almost all of their obligations almost all of the time’.Footnote ⁴⁹ In itself, this assessment was shared by Morgenthau, who acknowledged that ‘the great majority of the rules of international law are generally observed by all nations without actual compulsion’.Footnote ⁵⁰ However, Morgenthau and other political realists saw compliance with scepticism, arguing that it occurs essentially when it coincides with states’ national interests.Footnote ⁵¹ Henkin, on the other hand, affirmed the existence of a causal link between rule and conduct, of ‘forces which induce nations to observe law’.Footnote ⁵² States would prefer compliance, first, for ‘psychological’ internal reasonsFootnote ⁵³ and, second, to avoid damaging their relations with other states. ‘Internal impulsions and international inducements … promote the observance of law and obligation’.Footnote ⁵⁴

The Henkin formula would later be referred to as ‘the sentence that launched a thousand articles’.Footnote ⁵⁵ Over the next decades, a sophisticated compliance scholarship developed around the questions of why, when, and how international legal rules may influence state conduct.Footnote ⁵⁶ Contrasting with the somewhat sweeping statements sometimes made in previous contributions, the post-Henkin compliance scholarship provided nuance and considered with method the various elements that may affect a state’s decision to comply.

Its sophistication notwithstanding, the key debate within the compliance scholarship concerns the principal motive for state compliance. Some, following Morgenthau, emphasise the coercive element and reject outright any claims that norms have an independent effect on behaviour.Footnote ⁵⁷ Others seek to demonstrate how international norms are internalised by states, guiding their behaviour and delineating, or ‘constructing’, their identity and interests.Footnote ⁵⁸ Yet others seek to establish that states, though remaining essentially self-interested actors – or focal points for decision-making by domestic politiesFootnote ⁵⁹ – have a rational self-interest in complying with rules. Once they reach normative understandings, concerns with retaliation, reciprocal non-compliance and reputational costs could lead to rational compliance.Footnote ⁶⁰

For present purposes, there is no need to choose one strand of the scholarship over the others, nor to devise ‘compliance metrics’ as a benchmark for the significance of international law and its various subfields.Footnote ⁶¹ More relevant than the question of who is ‘correct’ is the articulation that each strand provides of the different factors contributing to compliance with international rules. Authors in the various strands at the end of the day recognise some merit in others’ arguments. Thus, realist George Downs acknowledges that ‘wrongheaded’ realism sometimes prevents sophisticated analysis, noting that enforcement has strengths and limitationsFootnote ⁶² and recognising a (limited) value in the concept of reputational costs.Footnote ⁶³ On the other hand, legitimacy-focused Thomas Franck sees rational choice as useful, arguing, however, that long-term national interests include strengthening the rule of law.Footnote ⁶⁴

The divergences between the strands often boil down to emphasis. As Beth Simmons notes, ‘the less one is willing to strawman the arguments of the major proponents, the clearer become the numerous points of overlap’.Footnote ⁶⁵ As Daniel Drezner points out (regarding the debate about the relevance of non-state actors), in explaining why their theories work ‘[b]oth sides tend to generalize from their most favorable cases’.Footnote ⁶⁶ Analysing the different strands of thought in this field, Benedict Kingsbury finds that they can be viewed as rationalisations of different conceptions of international law, with different scholarly theories relating closely to a scholar’s views on the role international law should play in the domestic and international order.Footnote ⁶⁷

In short, their disagreements notwithstanding, the various authors in the compliance scholarship elaborate on elements that are complementary in explaining compliance with international legal rules, within an international system characterised by the absence of an authority superior to that of states and capable of ensuring that legal commands are ‘armed with a sanction’.Footnote ⁶⁸ The different elements emphasised by the different schools essentially match the two forces identified by Henkin as motivating compliance: internal impulsions within states and external inducements produced by the prospect of other actors’ responses to their behaviour.

Importantly for present purposes, the compliance scholarship usually focuses on the internal perspective of states and their bilateral relations with other states. In so doing, it often collapses into the category of ‘external’ motives forces within the international society that are of very different characters, emanating from injured states, third states, and international organisations. For the purposes of understanding the instruments at the disposal of adjudicators to influence state conduct, it is helpful to consider in more detail the various types of external forces that promote norm-conforming and sanction non-conforming behaviour.

1.2.3 Compliance as System: Five Layers of Social Control

Seeing compliance as a single phenomenon with a principal cause can obscure the variety of forces that impel states (or any actors) to adopt certain behaviour in function of norms. Rather than having a single motive, compliance usually relies on the application of, and interaction between, various levels of self-imposed and externally imposed constraints. This may be all the more so with regard to states, whose conduct is rarely determined by a single individual. Even where a single state agent appears to make a decision, her being in a position to make this decision is itself the product of a series of social processes. And the decision, besides reflecting the agent’s own beliefs and preferences, is subject to various political and legal constraints at the domestic level; is influenced by the expectations of the decision-maker regarding the responses of other actors (domestic and foreign); and is subject to these actors’ reactions once the decision is announced or the relevant conduct carried out.Footnote ⁶⁹ An understanding of compliance and its relationship to judicial rulings demands an examination of the various layers of forces that may lead to what observers ultimately perceive as norm compliance and of the articulation between them.

An assessment of the articulation of these forces in an international setting can be developed on the basis of the typology outlined by Robert Ellickson in Order without Law: How Neighbors Settle Disputes, a study of a community of ranchers and farmers in the relatively isolated Shasta County, California.Footnote ⁷⁰ Ellickson found that the community’s members interacted primarily on the basis of informal social norms rather than state-enforced norms. Indeed, calling upon state agents to interfere with a dispute was considered a sanction under the ranchers’ informal normative order.Footnote ⁷¹ To explain the operation of this order, Ellickson developed a typology of classes, or ‘levels’, of norms that constitute ‘a comprehensive system of social control’, going beyond ‘law’ (a term he reserves to state-enforced, government-created norms) and encompassing an array of other social processes, or ‘sanctions’, that are mobilised or available to promote norm-conforming behaviour.

These processes range from ‘Level 1’ control by self-sanction (with ‘sanctions’ such as shame or guilt) to ‘Level 5’ control by government authorities, under the threat of enforcement by the state apparatus. Besides these, Ellickson identifies three other processes. Level 2 or bilateral enforcement prevails in ‘contractual’ rules, in the sense of promises made bilaterally between actors, sanctioned by means of self-help through various forms of retaliation. Besides bilateral sanctioning, norms are also subject to diffuse ‘Level 3’ enforcement by actors in the community, with sanctions such as gossip, and consequential harm to the violator’s reputation, or active third-party sanctions by other community members (‘vicarious self-help’). Finally, operating at Level 4 are organisations with voluntary participation, which often set up their own norms, applicable to their participants and controlled by the organisations themselves. Table 1.1 presents Ellickson’s typology.

Ellickson’s account of a comprehensive normative order (or ‘system of social control’) provides a useful starting point for the analysis of international law compliance for three key reasons. First, by presenting four levels of social control alternative to the ‘law’ enforced by the state under Level 5, this account offers a meaningful disaggregation of the forces that motivate actors’ compliance with norms besides, and regardless of, the intervention of a centralised coercive apparatus. Second, rather than seeking a main driving force behind compliance, it acknowledges that norm-oriented behaviour occurs for a multiplicity of reasons. For the purposes of understanding how adjudication can impact state compliance with international law, identifying the various layers of motives for compliance and having an account of the relevant compliance-inducing mechanisms is far more valuable than seeking ways to scientifically determine their relative importance. Third, it does away with the binary distinction between ‘legal sanctions’ applied by state agents and ‘social sanctions’ applied by other actors. Ellickson’s reconstruction of a comprehensive normative order still identifies sanctions as the distinctive element characterising a normative order.Footnote ⁷³ At the same time, it not only acknowledges that self-applied, interpersonal, and organisational sanctions can be more effective at shaping behaviour than state-administered sanctionsFootnote ⁷⁴ but also accords to this diverse array of sanctioning actors the character of norm enforcers.Footnote ⁷⁵

In short, the five levels provide an account of how a society without a centralised organisation of an authority structure and enforcement apparatus, and in this sense ‘anarchical’, may not be anarchical in the sense of being ‘devoid of rules’.Footnote ⁷⁶ Chapter 2 will develop a variation of the five-level system framework, applicable to the international society, using the term ‘layers of compliance control’. There, Table 2.1 synthesises the five layers of compliance-inducing forces of international law, each corresponding to one type of controller or enforcer. Together, these layers form international law’s compliance control system. The disaggregation of international law’s compliance control system illuminates the diversity of forces that prevent this horizontal social order, without a hierarchically organised legal order enforced through coercion by state agents, from being lawless. In the international society as in other social orders without a centralised enforcer, norms remain pervasive. They observably frame and guide social relations and, in cases of violation, are regularly upheld through decentralised sanctions, imposed by the various actors both on themselves and on each other.

The resulting normative framework is no doubt distinct from one in which an organised state apparatus, whose authorities have overwhelming control over the community’s material resources, is tasked with hierarchically interpreting and applying rules, as well as with enforcing them against vis-à-vis all actors. It is nonetheless characterised by sanctions, once the concept of ‘sanctions’ is extended beyond the sanctions applied centrally by such authorities. It then becomes clear that in horizontally organised societies a normative framework also exists, and that is constantly referred to, applied, and enforced through sanctions by its various participating actors.

1.3.1 Normative Indeterminacy and Semantic Struggles

The existence of a variety of institutions claiming distinctive normative authority challenges the picture of international law as a wholly horizontal legal order. Indeed, as a matter of its normative make-up, international law has become highly and diversely institutionalised over the past century.Footnote ⁷⁷ Through treaties and within international bodies, states have established and developed a number of fora where they can publicly exchange views and keep records of common understandings, establish procedures for collective decision-making, create agencies for independent scrutiny of conduct, and set up bodies with the ability to adjudicate disputes. As a matter of its means of social control, however, international law remains heavily reliant on decentralised sanctions by its component states. Though international institutions and bodies may provide interpretative guidance and occasionally condemn unlawful or undesirable conduct, states retain quasi- if not fully monopolistic control over the political and economic resources that can be mobilised to meaningfully affect other states’ interests.

As Ellickson’s framework clarifies, even complete decentralisation does not prevent a normative order from operating, guiding behaviour, or structuring interactions. A horizontal normative order operating in a wholly decentralised manner functions largely through self-regulation and self-enforcement.Footnote ⁷⁸ Indeed, in the international order, there is little doubt that states are overwhelmingly able to agree on the applicable rights and obligations and behave accordingly, settling most emerging disagreements through consultations and negotiation. Thus, international aviation depends on concordant interpretation of complex legal provisions and agreement by a multitude of actors regarding the permissible trajectories of thousands of aircraft; yet, tens of thousands of flights cross the Earth without incident every day. International trade depends on port workers applying rules and procedures agreed to in a dozen different international fora, which are consulted, argued about at the domestic level, and complied with (in the sense of not leading to inter-state disputes regarding compliance) in the overwhelming majority of cases. Even state borders enjoy the respect of virtually all states at virtually all times. A state that avowedly breaches another’s widely recognised borders (usually, by asserting a diverging interpretation of the applicable normative framework) will likely, at this very point, be respecting ‘almost all’ of its internationally recognised borders rather than rejecting the overall territoriality principle and the duty of states to respect borders.

As a result, if compliance is defined as state conduct being unopposed by other actors, international law is observed by ‘almost all nations … almost all the time’.Footnote ⁷⁹ However, by the same token, it is probably the case that almost all states are, almost all of the time, engaged in at least one interpretative struggle with at least one other state regarding their borders, their mutual trade obligations, or their duties with respect to one another’s citizens – or their own citizens.

This is to be expected. In any normative order, due to either genuine belief or self-interest (often, a mixture of both), different actors hold different interpretations of what norms require, permit, or prohibit. And, in the horizontal international legal order, a state’s ‘norm entrepreneurship’Footnote ⁸⁰ is kept in check solely by the ability of others – in particular, other states – to respond with claims of violation, demands for redress, and, ultimately, measures capable of compelling a change of heart in those determining the conduct of the norm entrepreneur state. Each state is able not only to state its own interpretations of international commitments and assessments of how facts fit under these commitments but also to act upon these self-interpretations. As Carmen Wunderlich notes, even a state disputing an overwhelming normative consensus (a ‘rogue state’) can be thought of, in a horizontal order with no commonly recognised norm interpreter, as a particularly strident norm entrepreneur, acting on its own understanding of the desirable normative framework, against other states’, equally valid, understandings.Footnote ⁸¹

This is not a mere academic construction. The Arbitral Tribunal in Air Service Agreement noted:

It is on these occasions, in which states with conflicting interpretations are under no special obligations to follow an institution’s pronouncements, that the international legal order faces some of its greatest difficulties. Whenever a dispute emerges between states regarding the lawfulness of a specific conduct, the horizontality of international law becomes patently clear. All states being equally sovereign, none has the authority to impose their interpretation of the applicable legal framework on another. Contrary to hierarchical legal systems, in which a distinction between ‘authorities’ and ‘subjects’ can be drawn so that the former’s interpretation prevails (in principle and until a contradictory interpretation is pronounced by a hierarchically higher authority), in international law there is (in principle) no criterion for determining whose interpretation, among those voiced by the different sovereign participants in an interaction, should prevail. When faced with horizontal interpretative disputes, states are locked in what Ingo Venzke termed ‘semantic struggles in which actors craft legal interpretations in an attempt to implement meanings of legal expressions that are aligned with their convictions or interests’.Footnote ⁸³

It is of course possible to seek to increase the clarity or specificity of legal rules and ‘narrow[] the scope for reasonable interpretation’.Footnote ⁸⁴ However, the authors of the leading article conceptualising legalisation in international relations go too far when they state that ‘[a] precise rule specifies clearly and unambiguously what is expected of a state or other actor … in a particular set of circumstances’.Footnote ⁸⁵ Semantic struggles are an inevitable consequence of the indeterminacy of normative text, in the sense that, while text can be informative, it is never exhaustive. It is impossible to establish that a certain conduct is a violation of legal commitments merely by considering the text of a treaty provision (or a pattern of conduct allegedly generating norms). Sentences (or patterns of conduct) may be vague or ambiguous, sometimes in ways that only become clear once an event occurs whose characterisation is disputed. There are no complete contracts.

The impossibility of full contractual precision arises, first, because a term’s meaning is inextricable from its context and from the purpose of the communication. This embeddedness of terms in their linguistic ecosystem is reflected in Article 31.1 of the Vienna Convention on the Law of Treaties (VCLT): ‘A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’.Footnote ⁸⁶ Similarly, with regard to customary law, the objective of ‘ascertain[ing] whether there is a general practice that is accepted as law’Footnote ⁸⁷ raises all forms of questions, which the literature has explored in depth.Footnote ⁸⁸ Even seemingly clearly worded rules invariably contain exceptions, sometimes provided for in other provisions of a treaty or in special practice, often inferred from a preamble, document or practice that exists in connection with the rule.

Additionally, even if a treaty provision – usually, the most easily identifiable source of a ‘rule’ – appears to establish an obligation unambiguously when read in isolation, a variety of treaty and extra-treaty instruments and legal mechanisms may render this obligation inapplicable, in general or to a particular situation. The provision may be subject to an exception under the treaty or a related instrument. Interpretation of the treaty or broader normative framework may need adjustments to reflect subsequent practice, subsequent agreement, systemic integration, and normative evolution.Footnote ⁸⁹ An obligation may have ceased to exist or be inapplicable to a situation due to lex posterior (desuetude or subsequent treaty), lex specialis (a more specific treaty rule), or lex superior (jus cogens or a norm deriving from the UN Charter).Footnote ⁹⁰ Finally, even an ostensibly clear violation may be justified if a circumstance precluding wrongfulness applies. The International Law Commission’s (ILC) Articles on State ResponsibilityFootnote ⁹¹ feature six of them: consent, self-defence, countermeasures, force majeure, distress, necessity, and compliance with peremptory norms.Footnote ⁹²

In short, each of the key stages of the legal assessment of a situation – the interpretation of a provision, the consideration of its surrounding normative context, the assessment of justifications and circumstances precluding wrongfulness, and the application of the interpretative outcome to the facts of a case – involves interpretation. Together, they provide plenty of room for the emergence of multiple, sometimes equally defensible, assertions regarding the content of obligations and the characterisation of conducts. And, even if states can agree on the applicable legal interpretation and on the characterisation of conduct, they may still disagree on a violation’s legal consequences for the affected parties.

International law thus operates under a fundamental difficulty: the lack of legal certainty in the application of its layers of social control. As Leslie Johns put it, ‘[f]or states to cooperate, they must first have common expectations about appropriate behavior and the consequences for inappropriate behavior’.Footnote ⁹³ This is not to say that decentralised interpretation impedes the operation of a normative system. In Ellickson’s community of cattle ranchers, sanctioning seemed to be applied by the various actors on the basis of their self-interpretation of the rules. However, while Ellickson downplays the relevance of uncertainty for the operation of an informal normative framework,Footnote ⁹⁴ a core difficulty for international law is the uncertainty inherent to the application of norms in the many complex and overlapping areas in which this horizontal normative order applies. It is on this point that the intervention of adjudicators qualitatively changes the operation of international law.

1.3.2 Settling Interpretative Disputes: The Role of Adjudication

The function of international adjudication is often described as that of settling interpretative disputes.Footnote ⁹⁵ It is true that, as a matter of their current set of functions in the international society, ICs have come to do far more than that. Armin von Bogdandy and Ingo Venzke noted that ‘dispute settlement’ suggests a state-oriented, bilateralist framing of the role of ICs. This framing leaves aside key functions performed by contemporary international adjudicators, including acting as organs of the international community and developing and shaping international legal regimes.Footnote ⁹⁶ Their broader formulation of the role of adjudication is that ICs ‘generate and stabilize normative expectations and control and legitimize other institutions’.Footnote ⁹⁷ At the same time, international adjudication continues to operate essentially according to the model of third-party dispute settlement. Even international criminal courts seek to organise their procedure as a series of ‘disputes’ between the prosecution and the accused, to be adjudicated impartially by the IC.Footnote ⁹⁸ The broader norm-setting authority of ICs depends, as a matter of procedural rules governing the exercise of jurisdiction, on their ability to hear and adjudicate upon an underlying, fact-laden, and usually bilateral, dispute.Footnote ⁹⁹

A dispute was famously defined by the PCIJ as ‘a disagreement on a point of law or fact, a conflict of legal views or of interests between two persons’.Footnote ¹⁰⁰ The settlement of the dispute by an adjudicator implies the end of valid disagreement for the purposes of the relevant legal order. It is, as a number of constitutive instruments of international adjudicators state, ‘final’,Footnote ¹⁰¹ ‘binding on the parties’,Footnote ¹⁰² and ‘without appeal’.Footnote ¹⁰³ As explained by the ICJ, ‘a judicial decision … has the force of res judicata and clarifies the rights and obligations of the parties’.Footnote ¹⁰⁴

From a legal standpoint, ICs are thus in a position to eliminate lawful disagreement. In a wholly horizontal order, sovereign equals’ claims regarding the applicable norms and the characterisation of facts are of equal value. This absolute horizontality disappears once an adjudicator validly exercises jurisdiction. As a matter of law, binding adjudicatory pronouncements perform what Robert Cover called a ‘jurispathic’ or ‘law-killing’ function.Footnote ¹⁰⁵ The adjudicator’s pronouncements set aside, for the purposes of the legal system, all but one of the competing claims regarding the normative framework, and its application to acts and events, put forward by the various actors in their interpretative struggles. Rosalyn Higgins described the judicial function as ‘to subject such disputes as the parties request to an authoritative legal process’ and ‘not finding “the correct rule” but rather making a choice between alternatives – and alternatives which … are often each meritorious’.Footnote ¹⁰⁶

At the same time, the decentralisation of the international order means that ICs lack a key element usually presumed in conceptualisations of adjudication developed having in mind domestic legal orders: the connection between adjudication and coercion. Shapiro considers this the distinctive feature of judges when compared to arbitrators: ‘having means to tap the organized forces of coercion in the society to enforce their solution’.Footnote ¹⁰⁷ In domestic legal orders, as Cover noted, ‘“interpretation” suggests a social construction of an interpersonal reality’,Footnote ¹⁰⁸ but judicial interpretation is marked by the shadow of violence: ‘as the judge interprets … she also acts – through others – to restrain, hurt, render helpless, even kill’.Footnote ¹⁰⁹

Assessments of the role and effects of adjudication usually connect judicial interpretation to the use, or at least threat or ‘shadow’, of coercion by the state apparatus to enforce judicial decisions.Footnote ¹¹⁰ The characterisation of courts as ‘jurispathic’ thus extends beyond the symbolic plane in which claims to normativity are made and proposed interpretations are offered. As Franklin Snyder explained, it means that ‘the judge will select one of the squalling brood of conflicting legal meanings to elevate and to enforce with the violence of the state – and will slay the rest’.Footnote ¹¹¹ Cover observes: ‘Legal interpretive acts signal and occasion the imposition of violence upon others’.Footnote ¹¹²

International adjudicators cannot rely on such a connection. Their rulings bind states as collective entities and establish their rights and obligations under international law. For the purposes of each state’s internal legal hierarchy, however, the determinative views will emerge not from adjudicators but from domestically authoritative persons and bodies – in particular, those within that polity who control the deployment of coercion by the state apparatus. Of course, states may choose to accord value in thei domestic legal hierarchy to international judicial rulings. As the Caribbean Court of Justice (CCJ) put it in TCL v. Guyana, ‘[i]ncorporation into domestic law may make it possible to invoke the coercive powers of the State in support of orders made by the Court and give efficacy to its orders’. By the same token, however, this incorporation determines the extent of the recognition of the IC’s authority domestically. If a dispute arises, ‘a court (presumably a local court) may be called upon to decide’.Footnote ¹¹³

Ultimately, the impact of ICs’ rulings depends on their influence on states’ domestic legal-political orders. It is the assessment of the situation by those within the domestic hierarchical structure, interacting under domestic political and legal processes, that will lead the various individuals acting on behalf of the polity to adopt the measures that international lawyers interpret as ‘state conduct’, and international relations scholars as ‘state behaviour’. The ability to influence these assessments, without being able to rely on a coercive apparatus, can be characterised as an IC’s authority.

1.4.1 Authority in Horizontality: Auctoritas Without Potestas

Authority is an essentially contested concept, ‘an object of dispute and political contestation’Footnote ¹¹⁴ wherever it has emerged. Its very existence as a separate element from power is sometimes challenged.Footnote ¹¹⁵ The contemporary term, as used in Latin-influenced languages with various shades of meaning, descends from the Roman auctoritas. Auctoritas is usually associated with an ineffable notion of moral authority, the ‘possession of something more than power’,Footnote ¹¹⁶ and stands in contrast with potestas, the more determinable legal authority within an organised legal/societal hierarchy. Potestas in Roman society has been described as ‘a legal power or authority to which a person was entitled’.Footnote ¹¹⁷ Auctoritas, by contrast, eludes a unified definition, and has been employed in a variety of expressions and occasions, conveying the notion that something ‘enlarges, confirms, or gives to a thing its completeness and efficient form’.Footnote ¹¹⁸

When referring to international judicial authority, I will use the term auctoritas where useful to avoid the association between authority and hierarchy, domination, subordination, and reflexive obedience. This association is pervasive in modern investigations of the concept. Max Weber stated, with regard to the state as a hierarchical order, that ‘[i]f the state is to exist, the dominated must obey the authority claimed by the powers that be’.Footnote ¹¹⁹ Hannah Arendt saw in the concept of authority two elements: an individual’s freedom to decide, on the one hand, and this individual’s voluntary surrendering of this freedom to another, on the other hand. She noted that authority exists ‘in contradistinction to both coercion by force and persuasion through arguments’.Footnote ¹²⁰ The authoritative relationship (she uses ‘authoritarian relation’) is therefore one characterised by obedience that ‘rests neither on common reason nor on the power of the one who commands’ but on ‘the hierarchy itself, whose rightness and legitimacy both recognize and where both have their predetermined stable place’.Footnote ¹²¹

Herbert Marcuse’s outline of the concept goes in a similar direction: authority is ‘the tying of will (indeed of thought and reason) to the authoritative will of an Other’, in a manner ‘which is not based purely on coercion’ but involves a degree of ‘voluntariness’, so that the authority relationship involves both ‘a certain measure of freedom’ and ‘the surrender of autonomy’ by the subordinate, a ‘tying of the subject’s reason and will to pre-establish[ed] contents … as the obligatory norms for his (sic) reason and will’.Footnote ¹²² Joseph Raz’s conception draws from the same dissociation between coercion and authority: ‘the justified use of coercive power is one thing and authority is another’.Footnote ¹²³ Authority exists, as Ingo Venzke put it, ‘between power and persuasion’.Footnote ¹²⁴ An authoritative pronouncement is one to which the addressee feels bound without needing to be rationally persuaded but also without the need for externally inflicted coercion (or, in some accounts, even the prospect of coercion).

These reflections on authority integrate, and in a sense assume, an underlying hierarchy recognised by both parties. This is not only the particular and circumstantial hierarchy between the two parties to the authoritative relationship. It is also a broader hierarchy, on which the particular hierarchy stands: a hierarchical ordering of society, where the authoritative actor’s instructions cause the addressees of its pronouncements to reflexively surrender their decision-making autonomy and comply with determinations made by the authoritative actor.Footnote ¹²⁵ As Kostiantyn Gorobets notes, the ‘standard view of authority … assumes that authority is necessarily somebody’.Footnote ¹²⁶

Raz’s notion of ‘practical authority’ aims to avoid assuming this underlying hierarchical structure, drawing a distinction between voluntary ‘surrender of judgment’ and ‘blind obedience’. Raz describes the role of practical authority as to ‘mediate between ultimate reasons and the people to whom they apply’.Footnote ¹²⁷ An authority examines an actor’s already-applicable reasons to act one way or the other and determines how these reasons will apply in a specific context. An authority can thus direct another person’s actions not due to the latter’s ‘blind obedience’ but because, and when, ‘it is better to conform … than to decide for oneself, unaided by authority’.Footnote ¹²⁸ As Raz explains, ‘all authoritative directives should be based, in the main, on reasons which already independently apply to the subjects of the directives and are relevant to their action in the circumstances covered by the directive’.Footnote ¹²⁹ The ‘point and purpose of authorities … is to preempt individual judgment on the merits’.Footnote ¹³⁰

Raz’s conception elides the question of how precisely the authoritative person or body elicits compliance with its determinations, reasoning solely that ‘[a] person can have authority over another only if there are sufficient reasons for the latter to be subject to duties at the say-so of the former’.Footnote ¹³¹ Explanations for what these sufficient reasons are tend to fall into either coercion or accepted legitimacy. Thus, an entity with authority is one that ‘is effective in imposing its will on many over whom it claims authority, perhaps because its claim to legitimacy is recognized by many of its subjects’.Footnote ¹³² Raz does consider the potential non-compliance with an authoritative ruling – in this case, an arbitral award – by a recalcitrant actor. However, he contends solely that compliance should be expected, since the very point of an arbitration is to require the renunciation by each party of the reasons they previously advanced for their actions, once these reasons are rejected by an arbitrator: ‘[i]f the disputants do not then reject these reasons as possible bases for their own action, they defeat the very point and purpose of the arbitration’.Footnote ¹³³

The example inevitably prompts the question: why would a dissatisfied litigant not seek to defeat the very point and purpose of an arbitration she has lost? Raz’s findings stop short of examining the social mechanisms through which an authoritative entity’s authority can be effective, even when (or rather, especially when) a party would prefer not to comply with the authority’s determinations and instead insists on the already-rejected reasons for action. In other words, Raz, whose central concern is with the legitimacy of authority, assumes authority itself, in the sense of the voluntary subordination of the will: ‘Raz’s conception of authority depends for its cogency on the thought that as long as the subject does better by reason overall by obeying certain classes of commands, the subject has a duty to obey every one of the commands: the correct as well as the incorrect’.Footnote ¹³⁴

The concept of authority relevant for present purposes – the authority of ICs – does not necessarily involve a subordination of the will in the sense that an international judicial authority’s pronouncements will be obeyed by others. After all, a key feature of the ‘sovereign equality’Footnote ¹³⁵ of states is the preservation of each state’s ability to remain the key focal point for decision-making even after expressing a commitment to one among the various available courses of action. States are not subject to each other’s authority or compelled to submit to a hierarchically superior authority, retaining the ultimate ability not to follow through on international commitments. Nor does international judicial authority match Raz’s alternative concept of ‘theoretical authority’, which is focused on the heightened persuasive value, a ‘reason for belief’, that authority grants to a person’s or an entity’s advice.Footnote ¹³⁶ A more illuminating parallel is the remark by Theodor Mommsen regarding the auctoritas of the Roman Senate, whose pronouncements he deemed ‘more than advice and less than a command, an advice which one may not safely ignore’.Footnote ¹³⁷ Similarly, international adjudicators hold authority in that their pronouncements may be disobeyed by their addresses but not disregarded without consequences.

This formulation, however, appears to equally beg the question. Why may states ‘not safely ignore’ an IC’s pronouncements? A claim to adjudicative authority is a claim to be able to influence perceptions regarding the content of its normative framework merely by issuing communications regarding norms and facts, without a connection to the coercive apparatus of the state.Footnote ¹³⁸ To affirm the possibility of ICs’ authority is to assert that their mere pronouncements, unaccompanied by the threat of coercion, can produce such a meaningful effect on sovereign states that they cannot be disregarded by their addressees.

Chapter 3 investigates this question, arguing that IC auctoritas derives from two related but distinguishable processes. One involves the adjudicator’s legitimacy or authority as traditionally considered, in the sense that an actor may have a propensity to comply with, or at least believe that she has a ‘rebuttable duty’ to implement, a ruling.Footnote ¹³⁹ This requires, as Laurence Helfer and Anne-Marie Slaughter put it, ‘convincing domestic government institutions … to use their power on [the tribunal’s] behalf’.Footnote ¹⁴⁰ The second process operates at the inter-state or, more broadly, interactional level, as what Thomas Ginsburg and Richard McAdams have termed the ‘expressive’ effect of adjudication.Footnote ¹⁴¹ The common recognition of the distinctive character of judicial pronouncements by the various participants means that, even where a specific actor would rather not recognise the adjudicator’s authority, each actor’s expectation is that other actors’ expectations will be affected by the pronouncement. The result is a self-fulfilling prophecy, in which common recognition of the adjudicator’s special status leads each actor to adjust its perceptions and conduct following the adjudicator’s ruling, not out of individual recognition of the adjudicator’s legitimacy to rule but owing to its expectation that other actors will adjust their perceptions and conduct.Footnote ¹⁴² Thus, an individual actor may disobey the adjudicator’s pronouncements and even declare its intention not to implement them or not to recognise their effects. However, because each of the actors in the international order knows the adjudicator’s pronouncements will anchor the expectations of the others, it is impossible for an individual actor to disregard them without consequence.

1.4.2 Using Auctoritas: Four Categories of Judicial Remedies

Authors considering adjudicative authority often portray it as a single phenomenon: the ability to issue communications that influence social perception of appropriate or required conduct. Adjudicators, however, act through specific communications, which I will refer to generally as ‘remedies’. Douglas Laycock defines remedies as ‘anything a court can do for a litigant who has been wronged or is about to be wronged’.Footnote ¹⁴³ A metaphor originating in English law, ‘remedies’ is now widely used to designate, in both common law and civil law jurisdictions, ‘[t]he means by which a right is enforced or the violation of a right is prevented, redressed, or compensated’.Footnote ¹⁴⁴

The notion that judicial remedies involve a court ‘doing’ something would seem to put remedies beyond the reach of international adjudicators. If ICs are defined by their lack of connection to a coercive apparatus, they are unable to directly affect controlling reality. They operate entirely through auctoritas. Accordingly, their remedies amount exclusively to communications regarding international law.

The multiple possibilities of these communications, on the one hand, and their lack of connection to coercive apparatuses, on the other, have led some to caution against attempts to develop extensive typologies in this area. As regards the ICJ, Ian Brownlie counselled that ‘there is no great profit in seeking to erect internal partitions within the sphere of judicial remedies’.Footnote ¹⁴⁵ Examining the question in light of the multiplication of ICs and tribunals, Christine Gray found that ‘[t]he diversity of tribunals is clearly reflected in the diversity of their practice on remedies’.Footnote ¹⁴⁶ The ICJ has affirmed a broad freedom to award remedies.Footnote ¹⁴⁷ Providing more detail, the European Court of Human Rights (ECtHR) has repeatedly asserted that its judgments are ‘essentially declaratory in nature’,Footnote ¹⁴⁸ including in this category its legal interpretations,Footnote ¹⁴⁹ findings of violation,Footnote ¹⁵⁰ and determinations of conduct required from wrongdoers.Footnote ¹⁵¹

However, it is possible to classify judicial remedies not on the basis of the practice of the various courts – an exercise in which one inevitably ends up concentrating on what courts themselves are prepared to award as remedies – but on the focus and function of the various potential types of authoritative communication an actor may issue. In Chapters 4 and 5, I offer and develop a fourfold typology of international judicial remedies, capitalised to facilitate identification: Mere Adjudication, Declaration of Breach, Consequential Duties, and Permissible Responses concern the new legal relations that emerge.

This typology is based on two divides. As regards their focus, remedies may focus on the ordinarily applicable rules of international law or on the new legal relations that arise following a violation.Footnote ¹⁵² The focus of Mere Adjudication and Declaration of Breach is the interpretation and application of the normative framework that applies to the parties independently of any alleged breach. Through Mere Adjudication, an adjudicator settles disputes regarding the applicable normative framework: the existence of certain disputed rules, their scope, applicability in a case, and interpretation. A Declaration of Breach settles a dispute regarding the legal characterisation of conduct under the adjudicated rules (as does its negative form – an express declaration that no violation has taken place).

The other two categories of remedies, Consequential Duties and Permissible Responses, focus on the new legal relations that emerge from a breach. Having identified a breach, an adjudicator may establish Consequential Duties. This covers both specifying the conduct required for cessation and determining the reparation that is adequate to offset the injury caused by the violation. Additionally, though less commonly, adjudicators may determine Permissible Responses, which focus not on the obligations that arise for a wrongdoer but on the permissible conduct for other actors in the face of the breach. In awarding the remedy of Permissible Responses, an IC employs its judicial authority to establish that certain measures are warranted (and, in some cases, required) and can (or must) lawfully be adopted by other actors, owing either to the violation or following the wrongdoer’s failure to provide redress for it.

On a different axis, it is possible to classify the four types of remedies according to the primary purpose of the remedy: clarifying obligations or imposing sanctions. A judicial pronouncement is useful for a complainant for two distinct reasons. It may provide authoritative interpretations of rules and clarifications of obligations, directly serving the interests of the complainant. But a complainant resorting to adjudication may have a different primary goal: harming the interests of the wrongdoer. Indeed, in many disputes brought to international adjudication, legal interpretation is incidental to the primary purpose of the complainant, which is to obtain a ruling that harms the interests of the alleged wrongdoer. This can be done either symbolically by the adjudicator itself, by creating for the wrongdoer embarrassment or stigma, or by a pronouncement that enables or enjoins others – those in control of political and economic resources – to adopt measures against the wrongdoer.Footnote ¹⁵³ In different ways, issuing a Declaration of Breach and determining Permissible Responses serve the latter purpose. These remedies trigger international law’s compliance-inducing mechanisms, and in this sense sanction, or enable the sanctioning of, violations.

Gathering on two different axes these two divides, and anticipating the detailed examination of remedies and judicial practice in Chapters 4 and 5, Table 1.2 categorises the various remedies that can be awarded by international adjudicators.