From Norm Violations to Norm Development: Deviance, International Institutions, and the Torture Prohibition

Abstract

How do violations affect international norms? This article demonstrates that violations develop norms by analyzing how international institutions determine the meaning of deviant behavior and the breached norm. Decisions by courts, ad hoc tribunals, commissions of inquiry, and expert committees influence formal and informal lawmaking and drive the contested and often ambiguous development of international norms. To illustrate the impact of these norm applications and lawmaking efforts, the article compares two institutions with different mandates to oversee the international torture prohibition. In the 1960s and 1970s, the European human rights institutions defined torture for human rights law and found that Greece and the United Kingdom had violated the torture prohibition, but created ambiguity regarding the threshold of torture. In 1984, the UN Convention against Torture (CAT) adopted this definition, which was informed by earlier norm violations. In the 1990s and early 2000s, the UN Committee against Torture (CmAT) applied the torture prohibition to interrogation techniques used by Israel and the United States in counterterrorism operations. CmAT’s decisions that both countries had deviated from the norm led to General Comment No. 2 on CAT, which reaffirmed and specified the absolute and non-derogable nature of the torture prohibition.

¿Qué efecto tiene el quebrantamiento de las normas internacionales sobre estas? Este artículo demuestra que los quebrantamientos desarrollan normas a través de un análisis de cómo las instituciones internacionales determinan el significado de este comportamiento desconforme y de la norma quebrantada. Las decisiones de los tribunales, los tribunales ad hoc, las comisiones de investigación y los comités de expertos influyen en la legislación, tanto formal como informal, e impulsan el desarrollo controvertido y a menudo ambiguo de las normas internacionales. Este artículo compara, con el fin de ilustrar el impacto de estas aplicaciones de normas y esfuerzos legislativos, dos instituciones con diferentes mandatos para supervisar la prohibición internacional de la tortura. Durante las décadas de 1960 y 1970, las instituciones europeas de derechos humanos definieron la tortura dentro de las leyes de derechos humanos y determinaron que tanto Grecia como el Reino Unido habían violado la prohibición de la tortura, pero creaban ambigüedad con respecto al alcance de la tortura. En 1984, las Convenciones de las Naciones Unidas contra la Tortura (CAT, por sus siglas en inglés) adoptaron esta definición, que se basó en anteriores violaciones de las normas. Durante la década de 1990 y principios de la década de los 2000, el Comité de la ONU contra la Tortura (CmAT, por sus siglas en inglés) aplicó la prohibición de la tortura a las técnicas de interrogatorio utilizadas por Israel y los Estados Unidos en operaciones antiterroristas. Las decisiones de la CmAT con relación a que ambos países se habían desviado de la norma dieron lugar a la observación general n.° 2 acerca de la CAT, en la que se reafirmaba y especificaba el carácter absoluto e inderogable de la prohibición de la tortura.

Quel est l'effet des violations sur les normes internationales ? Cet article démontre que les violations engendrent des normes en analysant comment les institutions internationales déterminent la signification d'un comportement déviant et d'une violation de norme. Les décisions des tribunaux, des tribunaux ad hoc, des commissions d'enquête et des comités d'experts ont une incidence sur la création de lois formelles et informelles, et favorisent l’élaboration contestée, et souvent ambiguë, de normes internationales. Pour illustrer l'effet de l'application de ces normes et de ces efforts de création de lois, l'article compare deux institutions dotées de mandats différents quand il s'agit de surveiller l'interdiction de la torture à l’échelle internationale. Dans les années 1960 et 1970, les institutions des droits de l'Homme de l'Europe ont défini la torture pour les lois relatives aux droits de l'Homme. Elles ont ainsi conclu que la Grèce et le Royaume-Uni avaient violé l'interdiction de la torture, tout en introduisant une ambiguïté quant aux critères de qualification de torture. En 1984, les Conventions des Nations unies contre la torture (CCT) ont adopté cette définition, renseignée par de précédentes violations des normes. Dans les années 1990 et au début des années 2000, le Comité des Nations unies contre la torture (CmCT) a appliqué l'interdiction de la torture aux techniques d'interrogatoire employées par Israël et les États-Unis dans le cadre d'opérations anti-terroristes. Comme le CmCT a établi que les deux pays n'avaient pas respecté la norme, le Commentaire général n° 2 relatif aux CCT est venu renforcer et préciser la nature absolue et non susceptible de dérogation de l'interdiction de la torture.

Introduction

The international torture prohibition is an absolute and non-derogable norm, codified in several human rights treaties. Historically, the norm has been repeatedly violated—including by democracies (Rejali 2007; Barnes 2017). This raises a central question for international relations (IR) theories: How do violations affect international norms? Because the torture prohibition “is a norm that accepts no deviations” (McKeown 2009, 15), it has been hypothesized that violations weaken it. Legal scholars emphasize that persistent torture practices call into question the norm’s status in customary international law (Parry 2010, 16–17). Following allegations of torture against the United States (US) during the “war on terror”, concerns about the erosion of the torture prohibition grew (McKeown 2009; Brunnée and Toope 2010, 266, 270; Kutz 2014). In retrospect, however, others argued that rather than undermining the norm, it was actually strengthened (Percy and Sandholtz 2022, 937; Lesch and Zimmermann 2023, 115). Building on the now widely shared assumption that violations do not per se weaken norms, explanations for the prohibition’s robustness have focused on contestation types (Liese 2009; Schmidt and Sikkink 2019; Stimmer 2019; Zimmermann et al. 2023, 39–40), strategies to deny deviance (Birdsall 2016), and pushback from international society (Keating 2014). This article argues that norm violations themselves develop norms and provides an explanation for the key role of international institutions in this process.

In international norm disputes, states, non-state actors, and institutions struggle over the meaning of norms and their violations. The article provides a refined understanding of how international institutions, while not fully independent from their interactions with states, contribute to the development of international norms. I argue that norm violations are neither fixed nor immediately observable but result from institutions like courts, ad hoc tribunals, commissions of inquiry, and expert committees applying norms to concrete cases and labeling potential infringements as “deviance”. They record their assessment that a particular practice constitutes deviance in informal and formal instruments, thereby significantly shaping the meaning of international norms. Through norm applications by international institutions, norm violations can drive norm development.

To conceptualize the link between norm violations and norm development, I develop a two-pronged model that draws on insights from legal theory, the sociology of deviance, and organization studies. Norm-applying institutions engage in fact-finding to establish knowledge about alleged violations and decide whether certain actions amount to deviance. This influences formal and informal lawmaking: Case-specific decisions on deviant behavior contribute to an evolving body of case law that incrementally shapes the meaning of international norms. By deciding on novel cases or overturning previous interpretations, precedents can be set. By generalizing from specific decisions on deviance, new or adapted normative understandings, mediated through the (non-)recognition by states, become enshrined in treaties, general comments, and similar documents, changing the meaning of international norms—albeit not necessarily in a progressive and consistent manner.

I analyze two international institutions with different mandates for overseeing the torture prohibition. Building on recent scholarship that examines how human rights courts and treaty bodies develop the concept of torture (Yildiz 2020, 2023; Davidson 2022), I compare how and to what effect the judicial bodies of the European Convention on Human Rights (ECHR)¹ and the quasi-judicial United Nations (UN) Committee against Torture (CmAT) react to alleged norm violations. Comparing norm applications and lawmaking before and after the 1984 UN Convention against Torture (CAT)² puts recent challenges to the torture prohibition into perspective and helps explain how violations affect norm development.

In the 1960s and 1970s, the European Commission on Human Rights (ECommHR)³ and the European Court of Human Rights (ECtHR) ruled that Greece and the United Kingdom had deviated from the prohibition of torture and ill-treatment but created ambiguity regarding the threshold of torture. The definition of torture developed by the ECommHR and the decisions of both ECHR bodies later shaped CAT. In the 1990s and early 2000s, CmAT found Israel and the United States in violation of the torture prohibition. Then, in 2008, in response to disputes with both States’ parties, CmAT adopted General Comment No. 2 on CAT to clarify its applicatory scope. The analysis of these cases demonstrates that norm-applying institutions influenced the development of the torture prohibition more than contesting and powerful states did. Without the institutional norm applications, the definition and interpretation of the torture prohibition would be different.

The theoretical model and the case studies contribute to constructivist and rationalist approaches to norm violations. Constructivists focus on the discourse about violations to determine whether there is a shared understanding of the validity of norms (Wiener 2018; Deitelhoff and Zimmermann 2020) and how international society defines its boundaries (Adler-Nissen 2014; Zarakol 2014). Rationalists focus on the (precedent-setting) effects of violations, which are likely to weaken norms in the absence of formal enforcement mechanisms (Panke and Petersohn 2012; Verdier and Voeten 2015). This article builds on these theories and develops them in two ways: Norm researchers have long acknowledged the inherent link between norms and deviance. However, by focusing on state responses to alleged violations, they have treated international institutions primarily as sites of norm contestation. To highlight institutional agency, I shed light on the other side of norm disputes by unpacking how international institutions label certain acts or actors as “deviant” before states accept, contest, or reject their decisions. Building on insights from organization studies, the article traces how international institutions “fix meanings in ways that orient action and establish boundaries for acceptable action” (Barnett and Finnemore 2004, 32). In so doing, it helps to explain the normative power of quasi-judicial institutions, thereby contributing to empirical research on the operation of legal norms and institutions (Finnemore and Toope 2001).

The remainder of this article is organized as follows. I begin by mapping IR scholarship on norm violations. Next, I define my key concepts, develop the two-pronged model of norm application and lawmaking, and introduce the case studies and methods. The case studies analyze how the ECHR institutions and CmAT applied the torture prohibition in four cases and thereby influenced lawmaking. Finally, I discuss the comparative findings and implications for further research.

Norm Violations and International Norm Dynamics

The role of norm violations is at the center of several overlapping IR theories.⁴ For norm researchers, norms indicate what constitutes appropriate and deviant behavior (Finnemore and Sikkink 1998, 891–92; Jurkovich 2020, 695). Norms are “counterfactually valid” despite their violation (Kratochwil and Ruggie 1986, 767). Contestation scholars even argue that “Norms often only become visible when they are violated” (Deitelhoff and Zimmermann 2020, 53). Similarly, research on stigmatization emphasizes that “deviant actors help to clarify norms” (Adler-Nissen 2014, 144). For judicialization scholars, norm violations are the key concern of adjudicatory bodies (Alter, Hafner-Burton, and Helfer 2019, 451). All of these approaches study “how the community assesses the violation and responds to it” (Kratochwil and Ruggie 1986, 767). Based on this widely shared assertion, scholarship has taken three directions, focusing on norm compliance, challenges, and change.

First, scholars have studied the promotion of compliance. Constructivists have demonstrated how non-governmental organizations (NGOs) trigger socialization processes (Finnemore and Sikkink 1998; Keck and Sikkink 1998), sometimes even acting as norm enforcers (Eilstrup-Sangiovanni and Sharman 2022). In contrast to a progressive socialization narrative, others point to the ambivalent links between stigmatization and compliance (Zarakol 2014, 315–17). Rational institutionalists study the politics and effects of naming and shaming norm violations (Lebovic and Voeten 2006; Hafner-Burton 2008) and the role of international courts in inducing compliance (Zangl 2008; Hillebrecht 2014; Staden 2018). More recently, this scholarship has begun to study the effects of quasi-judicial institutions (Carraro 2019; Creamer and Simmons, 2019, 2020; Ullmann and Staden 2023). This research focuses mainly on mechanisms that promote compliance and is primarily interested in state behavior rather than norm development.

Second, norm violations are a trigger for norm challenges. Contestation scholars have developed typologies to study the discourse of norm challengers to justify norm violations (Sandholtz 2007, 14–17; Wiener 2018, 38–42; Stimmer 2019, 272; Deitelhoff and Zimmermann 2020, 54–56; Lesch and Marxsen 2023, 29–34). The backlash literature identifies patterns in challenges to international courts (Soley and Steininger 2018, 240–41; Stiansen and Voeten 2020, 773–75; Kucik and Puig 2022, 2–3). Scholars of deviance distinguish different types of norm violations (Evers 2017, 789–91) and study how actors “manage” their stigma (Adler-Nissen 2014, 153–55; Saha 2022, 5–7). As a discursive phenomenon, however, norm challenges differ from norm violations in practice.⁵ Less attention has been paid to how such norm violations affect norm development (see also Evers 2017, 788).

Third, scholars study how violations affect norm change. Norm researchers have demonstrated how disapproval and stigmatization of violations trigger norm emergence (Finnemore and Sikkink 1998, 892; Rosert 2019, 1108–10). State actions and arguments about alleged violations are a constant driver of norm change (Sandholtz 2007). Recent studies show how deviators even act as norm entrepreneurs (Smetana and Onderco 2019; Wunderlich 2020). In the absence of enforcement, in contrast, rationalists view violations as an indicator for norm decay (Panke and Petersohn 2012, 722–23; Verdier and Voeten 2015, 12). From a slightly different angle, constructivists study violations as an indicator of norm weakening when they are not condemned (Deitelhoff and Zimmermann 2019, 6–7). However, this scholarship brackets the question of how norm violations are determined in the first place and affect the meaning of norms (see also Pratt 2020, 77; Kinsella and Mantilla 2020, 651)—especially when it comes to the role of international institutions (Yildiz 2020, 41). This is surprising because constructivist research highlights the power of international organizations in creating, applying, and diffusing norms (Barnett and Finnemore 2004, 31). Although judicialization scholarship makes legal institutions its object of study, it is mainly interested in their relations with states (Abbott et al. 2000; Alter, Hafner-Burton, and Helfer 2019; Stiansen and Voeten 2020). In contrast to judicial lawmaking scholarship (Venzke 2012), it has paid less attention to the processes of applying and interpreting norms (see also Stappert 2020, 37). Moreover, this research has focused primarily on judicial institutions like courts—less on quasi-judicial bodies (Alter, Hafner-Burton, and Helfer 2019, 453; but see recently Zvobgo and Graham 2020; Davidson 2022; Reiners 2022; Lesch and Reiners 2023).

In short, most of this literature studies norm violations in close relation to states—either as norm violators or as challengers. In doing so, it often assumes that we know a norm violation when we see it. But norms do not “interpret, or apply themselves; doing so requires agents and agency” (Putnam 2020, 32). While acknowledging the central role of states in norm change, this article focuses on the role of international institutions in this process. The next section advances this literature by showing how deviance not only triggers lawmaking but also shapes its content through the activities of international institutions.

Deviance, Norm Applications, and Lawmaking

To conceptualize the impact of violations on norm development, I draw on legal theory, the sociology of deviance, and organization studies. Based on a dynamic definition of deviance and a constructivist understanding of norm-applying institutions, I introduce a two-pronged model to analyze how norm violations are determined in institutionalized norm applications and how the resulting findings of deviance influence lawmaking.

Norms are inherently linked to deviance. Prohibition norms in particular gain meaning by critiquing behaviors they seek to prohibit (Möllers 2020, 90). A norm indicates that certain behaviors ought and ought not to be enacted (Finnemore and Sikkink 1998, 891; Winston 2018, 640; Jurkovich 2020, 695–96). The critique of targeted killings—though not universal—is based on the shared understanding that national sovereignty ought to protect certain individuals from assassination (Keating 2022, 2). The practice of targeted killing does not automatically invalidate norms against it (see Kratochwil and Ruggie 1986, 767). Taking this theoretical argument further, Möllers (2020, 61) even argues that “Without the possibility of norm violation, there is no normativity.” This builds on Durkheim (1982, 79), who argued that “normal” and “pathological social facts” should be analyzed as two sides of the same coin: Deviance “is not a property inherent in certain forms of behavior; it is a property conferred upon these forms by the audiences which directly or indirectly witness them” (Erikson 1962, 308). That is, a particular act is not a crime in itself: Social reprobation makes it a crime (Durkheim 1969, 81–82). The sociology of deviance foregrounds the role of norm-violating behavior in norm development.

Based on these assumptions, I define deviance as the outcome of a process in which certain actions are determined as breaching a norm and the wrongdoer is labeled. This dynamic concept of deviance is akin to pragmatist theories that suggest focusing less on whether or not actors follow prescriptions and more on the social interactions that negotiate “what counts as conforming to them” and what does not (Pratt 2020, 77; see also Lesch 2021, 616). In the targeted killing case, similar to the debates over autonomous weapons, this has often been difficult. Several state, non-state, and institutional actors struggle to determine this practice as norm violation, hindering the clarification of the scope of anti-assassination norms and the adoption of new norms to regulate emerging technologies (Rosert and Sauer 2021, 15–18; Keating 2022, 6–8). Determinations of deviance would specify, which practices are prohibited. This, in turn, would render existing norms less ambiguous and help to develop new rules against the use of force by drones and autonomous weapons.

Who determines deviance in world politics? Multiple actors determine and label deviance through various practices, including diplomatic practices of exclusion (Adler-Nissen 2014), NGO naming and shaming (Keck and Sikkink 1998; Eilstrup-Sangiovanni and Sharman 2022), and institutional pronouncements on non-compliance, outcasting norm violators (Hathaway and Shapiro 2011, 305–10; Hirsch 2015, 170–71). The different actors and modes of determining deviance can be illustrated in the context of the Russian war of aggression against Ukraine: Peace rallies across the globe and NGO reports have condemned the war; many states have sanctioned Russia, invoking core norms of the international order; the UN General Assembly (UNGA) has denounced the invasion as an illegal aggression; the Human Rights Council has dispatched a fact-finding mission to investigate violations of human rights and humanitarian law; and the International Court of Justice (ICJ) has issued the only legally binding decision do date ordering Russia to cease its military activities. It remains to be seen in which direction—if any—these determinations of deviance, which are not supported by all states and are disputed by some, will develop the international prohibition on the use of force. Yet this example shows that multiple and diverse actors are engaged in disputes over what constitutes deviance.

Deviance is the outcome of intermeshed social, political, and legal processes, triggered by alleged norm violations. In this article, I focus on “norm-applying institutions” defined as international bodies tasked to oversee treaties by ascertaining and applying norms (see Abbott et al. 2000, 415). They determine what violations are and fix what norms mean (see Barnett and Finnemore 2004, 32). My definition is broader than the usual definition of international courts and includes norm applications and lawmaking practices that do not depend on formal authority, binding decisions, and coercive enforcement (see also Alter, Hafner-Burton, and Helfer 2019, 453; Finnemore and Toope 2001, 747). Many institutions, like human rights treaty bodies or compliance panels in development institutions, monitor, review, and otherwise act as custodians of international norms (Creamer and Simmons 2019, 2020; Çalı, Costello, and Cunningham 2020,; Zvobgo and Graham 2020 ). They determine deviance even without formal adjudication.

In these institutions, lawyers and other international experts, bureaucrats, government officials, and NGO staff apply norms by establishing the facts of a case and determining whether or not those facts amount to deviance. Their agency depends on the willingness of states to activate adjudication bodies, cooperate in the process, and recognize their decisions (Alter, Hafner-Burton, and Helfer 2019, 454–56). Friendly settlements, which are increasingly used by the ECtHR, are a way to suppress norm applications and determinations of deviance, thus hindering norm development (Fikfak 2022, 946). Judges and experts in these institutions are not independent in enacting (quasi-) judicial practices that go beyond a purely formalistic norm application as they are embedded in professional fields and political structures. Judges in human rights courts and experts in treaty bodies are influenced in their decision-making by professional backgrounds, networks, and relations to States parties (Stiansen and Voeten 2020, 773–75; Caserta and Madsen 2022, 938–39; Reiners 2022, 58–62). The determination of deviance and its effects on norm development are entrenched in the environment of international institutions.

The practice of norm-applying institutions affects specific norm disputes and the broader normative structure in two ways (Figure 1). First, norm-applying institutions respond to allegations that a state has acted in violation of international norms. They gather information and evaluate the facts in light of these norms. This “norm application” results in institutional statements that label deviators. Second, norm applications link norm violations with formal and informal “lawmaking”. In case-specific decisions and general legal instruments, norm violations contribute to the development of international norms. Lawmaking can reaffirm or alter norms that are tested by future actions and allegations of norm violations as norm development evolves in ongoing cycles (Sandholtz 2007, 9–11). The next two sections further unpack both dimensions.

Figure 1.

Norm Development through Norm Violations.

Norm Application

International institutions use norms to evaluate an action or a situation to determine whether standards of appropriateness are met or whether certain actions deviate from the norm (Kratochwil 1989, 42; Möllers 2020, 109). Norm applications are based on a dual, interwoven “empirical” and “normative” process (Creamer and Simmons 2019, 1053). Empirically, norm-applying institutions engage in fact-finding to establish knowledge about cases of alleged norm violations. Normatively, they determine whether the empirical findings amount to deviance by interpreting and linking the relevant norms to the facts (Table 1).

Table 1.

Norm Applications: Fact-Finding and Determining Deviance

	Fact-finding		Determining deviance
Process	Empirical		Normative
Practices	On-site, Off-site	Collecting facts, hearing witnesses	Abstract	Interpreting norms, invoking precedents
	Adversary, Expert review	Validating evidence	Concrete	Linking facts to norms, evaluating meaning

To establish the facts of a case, norm-applying institutions gather information and give it meaning—thus creating knowledge about alleged norm violations (Barnett and Finnemore 2004, 29–30). They do this on-site by visiting the places where alleged norm violations occurred, e.g., inspecting detention centers or manufacturing sites for illegal weapons. Off-site fact-finding at institutional premises works differently: While many courts operate in an adversarial setting, allowing all parties to produce evidence and witnesses and to cross-examine them (Devaney 2016, 12–13), monitoring bodies typically rely on review-based fact-finding in which experts evaluate government and NGO reports (Viljoen 2004). In both cases, norm-applying institutions validate information in light of competing assessments to establish credible facts as the basis for determining deviance.

The institutions decide whether these facts constitute evidence of a deviation from a particular norm by engaging with the norm in the abstract and with the situation in the concrete. First, they interpret treaty norms and assess existing case law (Stappert 2020, 47–49). When necessary, they deal with potential ambiguities that general norms naturally leave and that are not always resolved in interpretations. Second, institutional actors link norms and their interpretations to specific situations and established facts. They evaluate whether a particular action deviates from or conforms with a norm and indicate what the norm means in specific situations (Alter 2014, 9). This process is not only about subsuming factual findings under normative interpretations. It is also about making decisions in controversies about competing evaluations (Kratochwil 1989, 227). Determining deviance ascribes meaning to the specific act and the applied norm (Möllers 2020, 109–10). For example, describing a state’s use of force as illegal aggression rather than self-defense affirms, specifies, or changes the meaning of the norm that prohibits the use of force in that and future cases. In so doing, norm applications can also create new ambiguities and even backtrack from earlier clarifications.

In short, in this contested process, actors like judges, experts, and state representatives—while embedded in social, political, and professional structures beyond the courtroom or committee panel—argue over their recollections of the facts and how to interpret a norm. Finally, norm-applying institutions publicize the outcomes of these processes, explicitly labeling and outcasting deviants, which affects their status and institutional membership (Hathaway and Shapiro 2011, 305–10; Hirsch 2015, 170–71).

Lawmaking

Fact-finding and determining deviance lead to a decision about whether a norm has been violated. These decisions can take different (usually written) forms, all of which I understand as contributing to lawmaking defined as “an activity that produces outputs that can be argued for as […] law” (Reiners 2022, 3, footnote 5). I distinguish between case-specific and general forms of lawmaking that record deviance in either formal or informal instruments (Table 2).⁶

Table 2.

Formal and Informal Lawmaking

	Formal	Informal
Specific	Judgments, advisory opinions	Observations, reports
General	Treaties	Standard-setting, general comments

Specific lawmaking includes all decisions made by norm-applying institutions in cases of alleged norm violations. Formal outputs of norm applications can be found in the judgments of courts. Advisory opinions are another example of specific, formal lawmaking by international institutions, such as the ICJ on the right to self-determination (Sparks 2023, Chapters 5–6). Decisions on norm violations are also recorded in informal outputs like observations and recommendations by quasi-judicial bodies such as the human rights treaty bodies (Çalı, Costello, and Cunningham 2020; Creamer and Simmons 2019, 2020), as well as World Bank inspection panels and OECD review processes (e.g., Zvobgo and Graham 2020). While they differ procedurally and legally, the formal and informal outputs manifest the meaning of a norm. In so doing, they add to the evolving case law and jurisprudence of norm-applying institutions, which provide reference points for future norm applications that can invoke these decisions—thus incrementally developing the norm.

When decisions on norm violations create new understandings of a norm or apply it to new situations, they can become important precedents as institutions cite these decisions (Lupu and Voeten 2012) or states invoke them to justify their actions (Sandholtz 2007, 16). At the same time, precedents are often a major concern for states. The Appellate Body of the World Trade Organization (WTO), for example, has increasingly relied on precedents in its rulings. While some states agree with these precedents, others challenge them as going too far beyond the original treaty (Kucik and Puig 2022, 7), underlining the limits of institutional lawmaking. Following Stappert (2020, 45), I assume that the basic logic of precedent applies to formal and informal settings like the human rights treaty bodies.

The central mechanism of general lawmaking is the formal adoption of international treaties through which states signal their consent and bind themselves to these norms. Yet this has never been the only source of international law, and recent research highlights the growing influence of informal lawmaking (Pauwelyn, Wessel, and Wouters 2013; Krisch 2014;,Rogers 2020). International organizations, secretariats, and expert committees increasingly make law—sometimes even in direct opposition to formal processes (Venzke 2012). For the purpose of this article, “general comments” by UN treaty bodies are a prime example of general informal lawmaking (McCall-Smith 2016), for example, in making water a human right (Reiners 2022). OECD standard-setting in development assistance is another example of a general, informal instrument.

The empirical part of this article applies this model and shows that norm violations and applications influenced formal and informal lawmaking to prohibit torture: CAT and General Comment no. 2 to CAT. The case studies reconstruct how, in the 1960s and 1970s, ECHR institutions defined torture and determined Greek and British interrogation practices as torture, and how, in the 1990s and early 2000s, CmAT established that similar Israeli and US practices constituted torture. Greek and British norm violations shaped the understanding of torture adopted by the drafters of the CAT. Israeli and US deviations were key for CmAT in drafting its second general comment.

Case Studies and Method

The torture prohibition is deeply rooted in human rights law with several institutions monitoring compliance and implementing the norm (Nowak, Birk, and Monina 2019, 2). I analyze how norm violations have driven the development of the torture prohibition by comparing norm applications and lawmaking in the ECHR institutions and CmAT (Table 3). These institutions oversee the torture prohibition but differ in design and practice. The ECHR institutions issue legally binding decisions through adversarial proceedings. As an informal body, CmAT does not pronounce legally binding decisions; its self-reporting mechanisms are based on expert reviews. The case studies demonstrate how the model works by analyzing how the meaning of violations, conveyed through fact-finding and determinations of deviance, affects norm development.

Table 3.

Comparing ECHR Institutions and CmAT

Institution	ECHR institutions	Committee against Torture (CmAT)
Legal nature	Court, judicial	Expert committee, quasi-judicial
Treaty law	ECHR	CAT
Mechanisms	Inter-state complaints, individual petitions	Self-reporting, inter-state complaints, individual communications, inquiries
Procedure	Adversary	Expert review
Outcome	Binding judgments	Non-binding decisions and observations
Norm applications	Greek case, Ireland v. United Kingdom (1969–1978)	Israeli and US reviews (1994–2006)

The ECHR institutions addressed allegations of torture in Greece and Northern Ireland in inter-state complaints between 1969 and 1978. CmAT dealt with allegations of torture against Israel and the United States in the self-reporting procedure between 1994 and 2006. The focus on these specific cases in the context of counterterrorism limits the jurisprudential terrain somewhat.⁷ However, the comparison of judicial and quasi-judicial institutions provides explanatory insights into whether legally binding judgments in inter-state complaints differ from informal observations in expert reviews; whether the practices and outcomes of norm applications and lawmaking affect norm development in different or similar ways; and how this, in turn, affects recognition by states. Finally, the analysis of cases before and after the adoption of CAT in 1984—a watershed moment for the prohibition—adds a longitudinal perspective that puts into perspective the debates about norm erosion in the early 2000s.

The case studies trace the process leading from an alleged norm violation to a new legal instrument to identify the mechanisms at work between norm violation, application, and lawmaking (see Meegdenburg 2023). Each case study begins with a brief introduction to the institutional setting. The case studies then summarize the actions that triggered allegations of norm violations and analyze how international institutions conducted their fact-finding and determined and labeled deviance. Next, I turn to lawmaking efforts that shaped the torture prohibition to trace how norm violations procedurally and substantively influenced new formal and informal instruments. The case studies provide empirical evidence for the claim that the outcomes of norm applications are integrated into new legal instruments—either during the drafting process or in the instrument itself. The in-depth analysis is based on archival materials, the official travaux préparatoires, meeting minutes, expert interviews, participant observations at a CmAT session in 2016, and secondary literature.

ECHR Institutions

Adopted in 1950 under the auspices of the Council of Europe (CoE), the ECHR established a tripartite architecture to oversee compliance with its norms—including the absolute and non-derogable torture prohibition (Mavronicola 2021, 16–17). The ECommHR was tasked with receiving applications, establishing facts, and reporting on norm violations. Initially, the CoE Committee of Ministers made the final decisions. Then, in 1959, the ECtHR was established to render legally binding decisions and order remedies in cases referred by States parties or ECommHR. Cases could be brought to the ECommHR—until its dissolution in 1998, a pre-trial chamber of the ECtHR that was also tasked with mediating friendly settlements—through individual and inter-state applications (Bates 2010, 120–24).

Adjudicating Interstate Complaints

In April 1967, the infamous “Greek colonels” overthrew the democratic government, claiming to fight communism (Bates 2010, 264). Reports of torture soon began to circulate. In September 1967, Denmark, Norway, Sweden, and the Netherlands filed interstate complaints against Greece under the ECHR, which became known as the Greek case (CoE 1972, 5–6). In 1969, the ECommHR found Greece in violation of the ECHR torture prohibition. In 1971, as the Northern Ireland conflict escalated, the British were accused of torturing detainees with the “five techniques”: wall-standing, hooding, white noise, sleep, and food deprivation (Rejali 2007, 363). British inquiries into the torture allegations produced findings that refuted them and were ambivalent about their legality under international law (Foley 2021, 112–13). Ireland was dissatisfied and, in December 1971, filed an interstate complaint against the United Kingdom at the ECommHR (Dickson 2010, 61). In October 1975, Ireland declined the proposal for a friendly settlement (Dickson 2010, 35, n 67). The ECommHR issued its decision in 1976 followed by an ECtHR judgment of 1978, both of which found the United Kingdom in violation of ECHR Article 3 on torture and inhumane treatment but differed in their interpretation of torture.

Fact-Finding

The ECommHR used an adversarial procedure—hearing both applicant and respondent government representatives—to establish the facts of the case. In the Greek case, the commissioners received memoranda and countermemos from Scandinavian and Greek representatives, heard witnesses from both sides, and visited Greece (Becket 1970, 96–104). They collected information in a representative sample of 213 submissions. Based on this material, ECommHR found that Greece practiced falanga (beating on soles), electric shocks, beatings, mutilations, and various forms of sensory deprivation (see also Rejali 2007, 276). In Ireland v. United Kingdom, the commissioners received oral and written submissions from the Irish and British governments and heard 100 witnesses speak to the allegations (CoE 1977, 522, 528). Striving to maintain a careful balance between the disputing parties, they heard witnesses produced and cross-examined by both sides. It found that the five techniques had indeed been used. As British commissions of inquiry had made similar assessments, the United Kingdom did not contest this finding. In both cases, fact-finding was based on written submissions, witnesses, and, in the Greek case, visits to places of detention. ECommHR validated the findings as credible facts in adversarial proceedings.

Determining Deviance

ECommHR then evaluated whether these facts constituted evidence of violating the torture prohibition. They began by interpreting the norm from scratch because key terms were not defined in European and international human rights law. The commissioners defined “torture” and “inhumane treatment” in the Greek case:

The notion of inhuman treatment covers at least such treatment as deliberately causes severe suffering, mental or physical, which, in the particular situation, is unjustifiable. The word ‘torture’ is often used to describe inhuman treatment, which has a purpose, such as the obtaining of information or confession, or the infliction of punishment, and is generally an aggravated form of inhuman treatment. (CoE 1972, 186).

This definition distinguishes torture—inhumane treatment that causes severe suffering—from inhumane treatment by its deliberateness and aggravation. In the Greek case, the ECommHR found evidence of an administrative practice: persistent acts of torture tolerated by the state (CoE 1972, 501). By thoroughly examining the torture allegations, it was able to “make a decision based on extensive evidence” (Becket 1970, 110). ECommHR established that the Greek practices crossed the threshold of torture and violated the country’s obligations under the ECHR. In Ireland v. United Kingdom, the commissioners based its interpretation on the Greek case. Addressing ambiguities in the Greek decision, it emphasized the non-derogable nature of the prohibition, which does not allow for exceptions even in cases of emergency (CoE 1977, 752). Based on two representative cases, they were convinced that the United Kingdom had used the five techniques to cause severe stress and suffering to obtain information—which constitutes ill-treatment and meets the threshold of torture (CoE 1977, 792–94). It found that the United Kingdom had violated Article 3 of the ECHR.

Ireland then referred the case to the ECtHR, which also found the United Kingdom in violation of the ECHR but differed in its interpretation of torture: A majority of judges found that the five techniques, while ill-treatment, did not amount to torture. The ECtHR (1978, para. 167) held that torture is limited to “a special stigma to deliberate inhuman treatment causing very serious and cruel suffering.” The special stigma notion has no roots in human rights law and jurisprudence; it was “invented” by the ECtHR to reserve the concept of torture to “particularly brutal” practices that leave visible marks (Farrell 2022, 4). The ECtHR cited the recently adopted 1975 UNGA Declaration on Torture, which states that torture is “aggravated and deliberate” inhumane treatment.⁸ Demonstrating the close links between norm violations and lawmaking (see below), the UNGA Declaration based its definition on the wording used in the Greek case, emphasizing the degree of suffering inflicted by each technique rather than its purpose.

The dissenting judges Franz Matscher and Dimitris Evrigenis took issue with the interpretation that reopened the uncontested ECommHR finding, arguing that modern torture also targets the psychological level, which should not be excluded from the ECHR definition (ECtHR 1978, 123–27; see also Mavronicola 2021, 65). These different positions in Ireland v. United Kingdom illustrate how norm applications are open to negotiation, which is not without social, political, and legal interactions beyond the courtroom, as the ECtHR decision and the dissenting opinions—including the opposing views by the Irish and British judges—show. Deviance is not predetermined, and its meaning does not necessarily evolve progressively.

Greece and the United Kingdom reacted differently to the decisions. Greece denounced the ECHR and withdrew from the CoE before being expelled (Becket 1970, 107). The United Kingdom did not contest the ECommHR report. As Commissioner Jochen Frowein noted, British recognition was partly based on satisfaction with the procedures of the case. The United Kingdom delegation had been skeptical as to whether “this Commission of continental lawyers” could handle their case (quoted in Wolfrum and Deutsch 2009, 23). Unsurprisingly, the United Kingdom accepted the ECtHR judgment, which acquitted it from the charge of torture. While the United Kingdom subsequently changed its discourse on torture (Foley 2021), the ECtHR decision had permissive effects as evidenced by the United Kingdom’s use of the five techniques in the early 2000s (Farrell 2022, 11–13).

From ECHR Precedents to CAT

The Greek case and Ireland v. United Kingdom are important instances of case-specific lawmaking that set precedents for the application of the torture prohibition and also shaped formal lawmaking in new international treaties. In the Greek case, the ECommHR articulated the first international definition of torture; it was also the first time that an international institution found a state in violation of the norm (Bates 2010, 266). Ireland v. United Kingdom was the first time that the ECtHR made such a decision (Yildiz 2020, 90). The special stigma notion has continued to shape its jurisprudence (Farrell 2022, 6–9). While the ECtHR 20 years later acknowledged that its concept of torture was open to change when it issued its first decision finding a state (France) guilty of torture (Yildiz 2020, 90), its unsystematic approach to Article 3 has been criticized for undermining its “capacity to guide” (Mavronicola 2021, 49–50). Even beyond the ECHR context, the decision had precedential effects: Decades later, Israel and the United States used the ECtHR’s narrow understanding of torture as a precedent to justify interrogation practices (see also below).

The decisions in the Greek case and Ireland v. United Kingdom also influenced lawmaking at the UN, where concerns about human rights in Chile inspired new efforts to combat torture (Burgers and Danelius 1988, 14–16). In 1975, the UNGA adopted the Declaration on Torture, which relied on the ECommHR decision in the Greek case to define torture (Burgers and Danelius 1988, 115). Article 1 of the Declaration restates the ECommHR definition from the Greek case and is based on its determination of deviance. As noted above, the ECtHR used the definition in the UN Declaration to justify its controversial conclusion that the five techniques did not constitute torture, thus demonstrating the close links between norm violations, applications, and lawmaking.

When the drafters of CAT sought to define torture, the main bone of contention was the second paragraph in the Declaration of 1975 and the notion of “aggravated” inhumane treatment (Burgers and Danelius 1988, 44). With the United States support, the United Kingdom pushed for language consistent with the ECtHR judgment that had acquitted them of torture (Burgers and Danelius 1988, 42–45, 73; Nowak, Birk, and Monina 2019, 31–32, 43). The first Swedish draft of CAT included “aggravation” in the definition of torture (Burgers and Danelius 1988, 203). A revised version bracketed this subclause (Burgers and Danelius 1988, 208), which was omitted in the final text: CAT Article 1 defines torture as

any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity.

This definition echoes the Greek case but avoids the ambiguity about “aggravation” in the second sentence of the UN Declaration on Torture and its interpretation in Ireland v. United Kingdom. The drafters of CAT adopted ECommHR’s interpretation of torture in Ireland v. United Kingdom, which defined a different threshold for torture than the 1978 ECtHR decision, which the United Kingdom and the United States had unsuccessfully advocated in the CAT drafting process (Nowak, Birk, and Monina 2019, 44). Unlike the ECtHR decision, ECommHR considered the five techniques and similar interrogation techniques (including sensory deprivation) to be torture by placing greater emphasis on their purpose. This is also key to the multidimensional CAT definition, which went beyond the focus on the severity of suffering that the ECtHR had advocated in 1978 (Nowak, Birk, and Monina 2019, 45).⁹ The torture definition in CAT translated this decision on norm violations against the United Kingdom into a general, formal instrument of international law. In terms of the norm’s applicatory scope, however, CAT obligations regarding torture in Article 2 and ill-treatment in Article 16 created space to exploit the vague distinction between torture and ill-treatment under CAT (Nowak, Birk, and Monina 2019, 443–46).

Norm violations by Greece and the United Kingdom and their determination as deviance by the ECHR institutions did not only shape law through case-specific decisions and precedents regarding the torture prohibition’s meaning. They were also essential for defining torture during the formal lawmaking process that led to CAT. While norm violations are crucial to norm development, the disputes over the “special stigma” of torture show that it can also stall or even go awry when institutions fail to unequivocally determine deviance.

Committee against Torture

CAT builds on the absolute and non-derogable torture prohibition in the International Covenant on Civil and Political Rights (ICCPR) and develops means to effectively implement this norm (Burgers and Danelius 1988, 1; Nowak, Birk, and Monina 2019, 6–7). CmAT is an expert body of ten elected members that was established in 1989 under CAT. Its central monitoring mechanism—obligatory for all States parties—is the review of self-reports and NGO shadow reports in “constructive dialogues” with state delegations (Creamer and Simmons 2019, 1052). The professionally diverse composition of CmAT shapes its interactions with state delegations (see Reiners 2022, 60). For example, longtime CmAT chair Jens Modvig’s medical background led him to ask different questions than the lawyers on the committee. The outcome of the review is observations and recommendations to improve the implementation of the norm. While non-binding and not always explicitly addressing norm violations (Kelly 2009, 793), even these informal documents raise “the political stakes of ignoring them” (Creamer and Simmons 2020, 17).

Reviewing State and Shadow Reports

In the 1990s, Israel was criticized for interrogation practices, including shabeh, overwhelming restraint, stress positions (forced squatting, sitting, and standing), strenuous physical exercises, white noise, and beatings (Rejali 2007, 354–57). The methods were often intensified through sleep deprivation and shaking (Rejali 2007, 329, 337). In the early 2000s, the United States used similar techniques, including waterboarding, stress positions, sleep and food deprivation, white noise, beatings, and mock burials (US Senate 2014, 32). The methods were accompanied by legal justifications. In Israel, the Landau Report of 1987 included guidelines for “moderate physical pressure”. In the United States, the 2002 Bybee Memorandum made the case for the legality of “enhanced interrogation techniques”. Both countries referred to the ECtHR judgment in Ireland v. United Kingdom to justify their methods (Landau Commission 1987, 69–70; Bybee [2002] 2005, 197–98). This underscores the (unintended) precedential effects of this decision. As before CAT, there was growing uncertainty about the legality of certain interrogation practices. CmAT applied the torture prohibition in reviewing Israeli and US state reports. Despite the informal procedure, both countries dispatched high-level delegations to engage with CmAT (see also Schmidt and Sikkink 2019, 108).

Fact-Finding

In contrast to the adversarial ECHR procedures, CmAT finds facts “indirectly” by having its experts review state reports and NGO “shadow reports” (Viljoen 2004, 59–61). The review is less formalized and relies heavily on secondary sources (Kelly 2009, 789). In the 1990s, Israel appeared before CmAT in 1994 and 1998 for its initial and second periodic reports, and in 1996, after submitting a special report requested by CmAT. It also used reports on the interrogation techniques by Amnesty International and Al-Haq, the Palestinian branch of the International Commission of Jurists (CmAT 1994b, para. 239). In 2006, after submitting its initial report, the United States met with CmAT for the first time. In addition to the US report, CmAT reviewed over 3,000 pages of NGO reports.¹⁰ It heard the Israeli and US positions and questioned the delegations. In meetings with the Israelis, CmAT experts referred to the findings of Amnesty International and Al-Haq, which they described as reputable sources (CmAT 1994a, para. 11). By 1998, Peter Burns, the country rapporteur for Israel, was confident that CmAT had “full knowledge of the facts” about Israeli interrogation techniques like beatings, hooding, sleep deprivation, stress positions, and shaking (CmAT 1998, para. 13). The Landau Report and Israeli Supreme Court rulings provided additional information on interrogations using moderate physical pressure (CmAT 1998, paras. 22–24). With regard to the United States, CmAT used shadow reports by Amnesty International, Human Rights Watch and others as important points of reference (CmAT 2006a, paras. 85, 97, and 102). CmAT members then inquired about the US position on water-boarding, sexual abuse, stress positions, and other techniques (CmAT 2006a, paras. 82 and 97), pointing out that some were explicitly sanctioned in official US documents (CmAT 2006b, para. 43).

CmAT assessed state and NGO reports on the relevant interrogation techniques. A review based on such documentation resembles an inquisitorial setting, with NGOs presenting evidence to the reviewing experts (on behalf of the plaintiffs as it were) to which the state under review reacts. Israel and the United States challenged that procedure by questioning the validity of NGO reports and demanding to respond to the allegations with their own information (CmAT 1998, para. 2). In contrast to the United Kingdom in the 1970s, neither Israel nor the United States was fully satisfied with this review-based fact-finding.

Determining Deviance

CmAT then evaluated these facts, addressing the specific interrogations techniques and the prohibition’s applicatory scope. The findings on the interrogation techniques were key to CmAT’s evaluation of these facts as evidence of torture. The Israeli delegation was asked whether it considered the practices to be torture or ill-treatment, since it did not contest using “moderate physical pressure”. CmAT disagreed with Israel’s view that it was not torture. Bent Sørensen, the alternate rapporteur for Israel, applied the CAT definition of torture to shaking “inflicted intentionally by public officials with a view to obtaining information”. He concluded that “there was no doubt that those cases fell within the meaning of article 1 of the Convention” (CmAT 1998, para. 25). For him, it was primarily the purpose of the technique, not the intensity of pain, that made it torture. Similarly, CmAT rejected Israel’s reliance on Ireland v. United Kingdom to show that practices involving moderate physical pressure were not torture (CmAT 1994a, para. 34; 1997b, para. 29). The US delegation was questioned about its interpretation of the CAT definition of torture and whether its enhanced interrogation constituted torture (CmAT 2006a, paras. 18, 82, and 99). Nora Sveaass stated that

according to official documents of the United States, certain methods of interrogation being applied covered practices that the Committee defined as acts of torture, such as forced naked exposure, being required to remain in painful position, and the exploitation of phobias in the person being interrogated. (CmAT 2006b, para. 43)

When Israel and the United States sought to exploit ambiguities in the non-derogable prohibition of torture, CmAT responded that CAT applies at all times and in all circumstances—rejecting arguments about its limited applicability in territory under de facto control or in occupied territory (CmAT 2006a, paras. 14–15, 18–19, 1994a, para. 167, 1998, para. 238).

Based on this assessment, CmAT concluded that Israeli and US interrogation techniques violated the torture prohibition. In 1997, CmAT stated that moderate physical pressure constitutes “breaches of article 16 and also constitute torture as defined in article 1 of the convention. This conclusion is particularly evident where such methods of interrogation are used in combination, which appears to be the standard case” (CmAT 1997a, para. 5). Similarly, CmAT found that the United States had authorized interrogation techniques that “led to serious abuses of detainees” under CAT Articles 1 and 16. CmAT rejected the US interpretation of the torture prohibition and concluded that US interrogation techniques clearly violated CAT. In its concluding observations, CmAT called on the United States to

rescind any interrogation technique, including methods involving sexual humiliation, ‘waterboarding’, ‘short shackling’ and using dogs to induce fear, that constitutes torture or cruel, inhuman or degrading treatment or punishment, in all places of detention under its de facto effective control, in order to comply with its obligations under the Convention. (CmAT 2006c, para. 24).

Even as informal decisions about deviance, CmAT’s concluding observations put the torture prohibition into effect. While the United States and Israel attempted to blur the distinction between “what the rules allow and forbid” with respect to torture (Hurd 2017, 125), CmAT established that the torture prohibition had been violated and exhibited US deviance for all to see (similarly, Creamer and Simmons 2019, 1053).

Although Israel and the United States challenged the findings by questioning CmAT’s authority to adopt general conclusions and interpretations, both countries officially denounced their own interrogation techniques. CmAT helped bring “international law to bear on Israeli domestic actions” when the Israeli High Court of Justice ruled in 1999 that moderate physical pressure was illegal (Grosso 2000, 333). By decree, Israeli security agencies were ordered to comply with this judgment (Barnes 2017, 134). Yet its implementation remains controversial, which also shapes the ongoing exchange with CmAT (CmAT 2019, para. 30). The institutional determination of deviance by CmAT also provided an important counterweight to the US challenge, which received little international support (Keating 2014, 80–81; Schmidt and Sikkink 2019, 109–12; Stimmer 2019, 277). In one of its first decisions, the Obama administration rescinded the enhanced interrogation techniques. In 2014, the United States returned to CmAT for its second and third periodic reports, seeking to revoke the deviator label by accepting that enhanced interrogation constituted torture and assuring CmAT that it had ended all related methods (CmAT 2014, paras. 4–5).

General Comment No. 2

The CmAT review process was not just important for the US case. As an example of informal, specific lawmaking it clarified “antitorture norms more generally and what normatively defines torture and cruel, inhumane, and degrading treatment” (Creamer and Simmons 2019, 1053). In rejecting the ECtHR decision, CmAT applied the ECommHR interpretation of the torture prohibition and rebuffed attempts to accommodate counterterrorism practices under the torture prohibition. This is important because some states emulated US arguments to justify torture (Schmidt and Sikkink 2019, 111). By determining and labeling deviance, CmAT helped push back on these challenges. These two norm applications also influenced more general lawmaking in CmAT’s second “general comment” on CAT (CmAT 2008).¹¹

Human rights treaty bodies are created to monitor human rights norms and are not explicitly mandated to engage in lawmaking. However, drafting general comments helps them to develop norms relatively independently (Reiners 2022, 21). They use this informal instrument to enshrine their interpretation of human rights norms and to inform states of their obligations (Creamer and Simmons 2020, 33). CmAT was long reluctant to use general comments (Nowak, Birk, and Monina 2019, 528). The first was adopted in 1998, but several attempts to draft others were stifled. One of the reasons for this reluctance can be found in the professional background of CmAT members as international lawyers from a common law tradition have argued for maintaining a case-specific approach (e.g., CmAT 2002, para. 73). Discussions about commenting on CAT Article 2 only took off in 2006 in the face of US norm violations and ongoing disputes with Israel (Gaer 2008, 195; Rodley 2008, 354; for a detailed reconstruction of the drafting process, see Lesch and Reiners 2023).

Article 2 of CAT codifies state obligations regarding the universal territorial applicability and non-derogability of the norm. Building on the ICCPR, it is one of CAT’s core provisions (Nowak, Birk, and Monina 2019, 72–73). After Israel and the United States attempted to exploit the ambiguity of Article 16 with respect to ill-treatment, CmAT drafted General Comment No. 2 to counter their arguments in a general informal instrument, based on a letter that CmAT had circulated to all States parties to CAT in the wake of September 11, 2001 (CmAT 2006d, para. 26, 2008, para. 6, n1). In the drafting process, CmAT members cited Israeli arguments about moderate physical pressure as an attempt to blur the line between torture and ill-treatment (CmAT 2006d, para. 23). CmAT clarified that the “obligation to prevent ill-treatment in practice overlaps and is largely congruent with the obligation to prevent torture”, including its non-derogability (CmAT 2008, para. 3). General Comment No. 2 builds on and reaffirms CmAT’s position on Israel and the United States, and distinguishes CAT from the ECtHR decision in Ireland v. United Kingdom by treating torture and ill-treatment as equally prohibited—and not “at the top end of a pyramid of pain and suffering” as implied by the 1978 ECtHR decision and subsequent Israeli and US interpretations (Rodley 2008, 356–57).

General Comment No. 2 also addresses CAT’s territorial scope and non-derogability. It further specifies that the torture prohibition applies in the context of “any threat of terrorist acts or violent crime as well as armed conflict, international or non-international” (CmAT 2008, para. 5). During the drafting process, CmAT members explicitly referred to the dialogue with the United States to clarify that terrorism could not justify an exception (CmAT 2006d, para. 28). General Comment No. 2 states that CAT applies to all territories under the de jure and de facto control of States parties (CmAT 2008, para. 7). CmAT’s determination of deviance in Israeli and US self-reports helped to reinforce the norm by underlining that torture and ill-treatment are prohibited under international law in all circumstances and in all territories under the legal and effective control of States parties. This rejects arguments that implicitly seek exceptions for counterterrorism, in (secret) detention facilities, and in occupied territories.

General Comment No. 2 critically developed the torture prohibition and became a central point of reference in CmAT’s practice.¹² The legally amorphous nature of general comments means that, perhaps more than other lawmaking mechanisms, they depend on recognition (Reiners 2022, 32–33). Many States parties supported CmAT during the drafting process (CmAT 2007, para. 4). The UNGA included General Comment No. 2 in its 2008 resolution on the torture prohibition to emphasize its absoluteness and non-derogability—a sign of broad support by states.¹³ The US State Department, however, has issued a detailed critique of General Comment No. 2, aimed primarily at denying it any legal effects (US State Department 2008). Yet human rights courts cite General Comment No. 2,¹⁴ which underlines that this informal lawmaking has been recognized as part of international human rights law (see also McCall-Smith 2016, 34–45). By establishing that Israeli and U.S. interrogation techniques violated the torture prohibition, CmAT has significantly reinforced the norm, which also shaped informal lawmaking in General Comment No. 2.

Conclusion

Norm violations are a driving force behind the development of international norms. In this article, I have demonstrated how international institutions determine deviance and engage in lawmaking. In the two case studies, norm violations influenced formal and informal lawmaking through norm applications by institutions ranging from courts to expert committees—even despite powerful challenges to the norm.

Once the ECHR institutions had defined torture and determined Greek and British interrogation techniques as “torture”, these violations shaped CAT. However, the ECtHR precedent in Ireland v. United Kingdom continued to influence contestations of the torture prohibition. Although institutionally very different, CmAT advanced norm development by reviewing Israeli and US interrogation techniques and establishing that they deviated from the prohibition. Then, in General Comment No. 2, CmAT took on the role of a lawmaker. It reaffirmed the non-derogability of the torture prohibition in light of Israeli and US deviance and ruled out arguments based on the “bad precedent” of Ireland v. United Kingdom. Norm-applying institutions—even without formal authority—can engage in lawmaking when their pronouncements are judicially or politically recognized. CmAT’s General Comment No. 2 is part of a series of successful lawmaking efforts by several treaty bodies (Reiners 2022, 116). More research is needed on whether and how general comments have been driven by norm violations. This should also extend beyond human rights to study the link between norm violations and other informal lawmaking efforts by compliance panels, commissions of inquiry, peer-review mechanisms, and NGO reporting.

The case studies also shed light on how the design of different institutions affects the way states perceive them and react to their judgments, adding to research on compliance with decisions of expert bodies and courts. More research is needed to identify conditions under which states accept or reject labels of deviance from international institutions. This should link insights on contestation patterns and stigma management with the practice of norm-applying institutions. Although both the Greek and British delegations acknowledged the adversarial procedures—similar to common law court proceedings—only the United Kingdom accepted the verdict. On the other hand, Israel and the United States contested the procedures and decisions by CmAT, whose members review evidence based on self-reports and shadow reports by NGOs—closer to proceedings in a civil law tradition. British recognition of the ECHR’s adversary procedures and Israeli and US critiques of the practice of expert review suggest that states are more likely to recognize proceedings in which they are closely involved. Such research could also add another layer to scholarship on the backlash against international institutions.

Beyond human rights, future studies should examine the informal and confidential approach by the International Committee of the Red Cross to monitoring compliance in armed conflicts and analyze how norm violations have influenced formal and informal lawmaking efforts to improve, for example, the protection of civilians in the Additional Protocols to the Geneva Conventions or the recent Dublin Declaration (Kinsella and Mantilla 2020). In trade law, scholars should study how, for example, disputes over intellectual property rights violations have influenced lawmaking in the 1994 Agreement on Trade-Related Intellectual Property Rights and the 2001 Doha Declaration. These disputes continue to shape debates about the limits of intellectual property rights with regard to compulsory licensing of medicines in health crises (Sell and Prakash 2004, 157–58). Finally, future research should study how institutions that are unable or unwilling to determine deviance affect norm development. This can be seen in the blockade of the WTO Appellate Body (Kucik and Puig 2022) and the silence of the UN Security Council regarding interventions against non-state actors (O'Connell, Tams, and Tladi 2019). Studying norm violations and international institutions in these and other cases will provide better explanations for the puzzle of why norms remain robust and develop despite—or even because—of their violation.

Footnotes

1

European Treaty Series, No. 5.

2

UN Treaty Series, no. 24841, vol. 1465 at 85.

3

ECommHR functioned as a pre-trial chamber of the ECtHR (Bates 2010, 120). It was dismantled in the 1998 reform.

4

For a recent international law approach to study the link between norm violations and the international legal order, see Marxsen (2021).

5

Stimmer and Wisken (2019, 521–22) include behavioral contestation that not necessarily violates norms.

6

Next to norm violations, technological and societal changes, emerging issues, social pressure on states, and agenda setting by transnational coalitions also influence lawmaking (Keck and Sikkink 1998; Sandholtz 2007; Rosert 2019; Kinsella and Mantilla 2020; Reiners 2022).

7

For comprehensive assessments of ECHR jurisprudence, see Yildiz (2020, 2023) and Mavronicola (2021); for an inter-institutional perspective, see Davidson (2022).

8

UNGA Resolution 3452 (XXX), Article 1(2).

9

Recent ECtHR jurisprudence frequently cites CAT, emphasizing “purpose” as a definitional criterion (Mavronicola 2021, 74–75).

10

Interview with CmAT member, May 2016.

11

It was also important to frame domestic violence as torture (Davidson 2022, 214–18).

12

Interview with CmAT member, May 2016.

13

UNGA Resolution 63/166, para. 26.

14

The ECtHR cited General Comment No. 2 in five decisions: 67258/13, 36391/02, 25703/11, 26828/06, and 41261/17; https://hudoc.echr.coe.int (accessed December 2, 2022).

Author Biography

Max Lesch is a Postdoctoral Researcher at the Peace Research Institute Frankfurt (PRIF). His research interests include norm contestation and deviance in human rights and peace and security law, as well as international practice theories, fact-finding missions, and (in)formal lawmaking.

Notes

Author’s note: This paper has benefited greatly from discussions with Rebecca Adler-Nissen, Antonio Arcudi, Nicole Deitelhoff, Klaus Dingwerth, Jennifer Erickson, Frank Gadinger, Nele Kortendiek, Stefan Kroll, Christian Marxsen, Galia Press-Barnathan, Elvira Rosert, Philip Wallmeier, and Lisbeth Zimmermann. I would like to thank the editors of International Studies Quarterly and the three anonymous reviewers for their constructive feedback and helpful comments on earlier versions of this article.