CHALLENGING THE USE OF EXTERNAL SOURCES BY THE INTER-AMERICAN COURT OF HUMAN RIGHTS

Abstract

This article challenges the justification usually offered by the Inter-American Court of Human Rights for its broad use of external sources when engaging in evolutive interpretation of the American Convention on Human Rights (ACHR). It analyses the Court's jurisprudence concerning international humanitarian law, the rights of the child, and lesbian, gay, bisexual, transexual and intersex (LGBTI) rights, in addition to drawing on interviews conducted with lawyers of the Court. It argues that the discursive strategy used by the Court to justify its ‘import’ of external sources fails to provide a complete normative justification and remains open to the charge of ‘cherry-picking’. The article recommends that the Court tailors its discursive strategy to the specific type of external sources used and suggests that more attention be paid to searching for internationalized consensus when determining the relevance of non-binding sources to evolutive interpretation of the ACHR.

I. INTRODUCTION

The American Convention on Human Rights (ACHR) was adopted in 1969 in order to consolidate, ‘within the framework of democratic institutions, a system of personal liberty and social justice based on respect for the essential rights’ of people in the American continent.¹ The Convention contains 82 provisions, 22 of which safeguard civil and political rights, and one establishes the obligation of States parties to adopt measures to achieve progressively ‘the full realization of the rights implicit in the economic, social, educational, scientific and cultural standards set forth in the Charter of the Organization of American States’ (OAS).² Since its adoption, the ACHR has evolved and expanded to accommodate contemporary contexts and new ways of living. This expansion has not only taken place through the adoption of additional protocols but also through the interpretive practice of the Inter-American Court of Human Rights (IACtHR).

The adoption of amendments and protocols requires protracted negotiations that are frequently time consuming. To depend on such formal modifications to address emergent issues can result in legal vacuums for possibly extended periods of time. Accordingly, the IACtHR promotes the evolutive, or dynamic, interpretation of the ACHR. This is a method of treaty interpretation that seeks to broaden the content and scope of human rights treaties to encompass new social realities and conditions.³ It is based on the understanding that such treaties are living instruments, which must continuously and effectively safeguard the rights and freedoms of individuals over time. In practice, by applying this method human rights courts are able to address such newly emerging issues within the framework of existing provisions.

Whilst human rights courts retain their relevance by updating human rights law to reflect contemporary challenges in this way,⁴ scholars have pointed to tensions between judicial lawmaking and the legitimacy of international courts.⁵ In this context, von Bogdandy and Venzke argue that ‘[o]ne of the first and foremost elements that contribute to the democratic legitimation of judicial lawmaking is nested in the established forms of legal argument—in the discursive treatment of the legal material’.⁶ Thus developing clear and persuasive arguments to support their judgments is crucial if courts are to maintain and strengthen their legitimacy. In turn, studying the discursive strategy of tribunals permits scholars to understand better the court's normative legitimacy and its strategy for democratic legitimation. As Bankowski, MacCormick, Summers and Wroblewski explain, a court's discursive strategies reveal ‘an effort at self-conscious public justification’ and unveil what the court regards as ‘satisfactory and publicly acknowledgeable grounds for decision making’.⁷

As a general rule, the IACtHR engages in evolutive interpretation of the ACHR based on a comparative study of international instruments, including rules and soft-law instruments adopted outside of the Inter-American human rights system (hereinafter, external sources). The IACtHR has often justified relying on external sources by noting that the ACHR is part of the corpus juris of international human rights law and stressing that this ‘outward-looking’ interpretative exercise is guided by Article 31 of the 1969 Vienna Convention on the Law of Treaties (VCLT) and Article 29 of the ACHR.

This article challenges this justification and undertakes an in-depth analysis of how and to what extent the Court's broad use of external sources does in fact constitute an interpretative method based on Article 31 of the 1969 VCLT and Article 29 of the ACHR. In other words, it examines the extent to which the IACtHR's public justification for its reliance on external sources is sufficient to provide a legal basis for the tribunal's broad reliance on them. By showing both the potential and the limitations of the Court's justification, this article aims to identify what types of external referencing, if any, are or are not normatively justifiable on this basis.

This article is based on an analysis of the Court's jurisprudence concerning the evolutive interpretation of the ACHR in three fields of law: international humanitarian law (IHL); the rights of the child; and lesbian, gay, bisexual, transexual and intersex (LGBTI) rights.⁸ All these areas provide examples of evolutive interpretation and involve recourse to diverse external sources as interpretative guidance. Cases concerning IHL illustrate the Court's reliance on sources other than human rights. Cases involving the rights of the child constitute an excellent case study of the Court's reliance on external rules binding upon the respondent State. The Convention on the Rights of the Child (CRC) has been ratified almost universally by United Nations (UN) Member States and by all the States accepting the jurisdiction of the IACtHR.⁹ Lastly, cases concerning the rights of LGBTI persons concern questions on which consensus is generally perceived to be still emerging in the region. In addition, the Court's advisory opinion on the topic of same-sex marriage and gender identity prompted a significant public debate across OAS States and led to questions concerning the Court's legitimacy.¹⁰ The latter debate highlights the importance of this study, as will be discussed in Section II. A few cases that do not fit into these three fields of law—notably Artavia Murillo et al (“In Vitro Fertilization”) v Costa Rica, Hacienda Brasil Verde Workers v Brasil and Claude Reyes et al v Chile—are also considered in this article because they provide a unique illustration of external referencing and offer a detailed discussion of legal hermeneutics.¹¹

Additional research sources drawn on in this article are interviews conducted with lawyers of the IACtHR (hereinafter participants A, B, C and D). The interviews were a supplementary means for identifying the role of external sources in the Court's jurisprudence and for clarifying the Court's broad reference to the VCLT as a justification for its interpretative methodology. As indicated by the International Law Association (ILA), the Court's reference to the VCLT when discussing treaty interpretation ‘appears in a very formal way, as a sort of necessary and obliged international classical mantra’.¹² The interviews were thus valuable for understanding the factors that the Court's Secretariat might consider when examining how the Convention has evolved and what discursive strategy might be adopted to justify these interpretative choices. The interviews pointed to specific links between the Court's external referencing and the VCLT's rules of interpretation, which are examined in this article.

The analysis of the case studies, together with the interviews, shows that three types of external sources have guided the Court's evolutive interpretation of the ACHR: rules binding upon the respondent State; hard law not binding upon the respondent State; and soft-law instruments (see Section III). In all three cases, the Court has sought to justify its use of external sources on the basis of general references to the VCLT and Article 29 of the ACHR (see Section IV). To test the validity of this, the article examines in detail the extent to which the Court's use of external sources falls within the limits of these rules of interpretation (see Sections IV, V and VI).

This article concludes that the Court's approach is insufficient for establishing a normative basis for its extensive reliance on external sources and thus it will continue to be susceptible to charges of selective justification, or ‘cherry-picking’. It recommends that more attention be paid by the Court to a search for internationalized consensus when relying on non-binding sources. In addition, it suggests that the Court tailors its justifications to the specific type of external sources used and that, similarly, scholarly work should tailor its criticism to the specific types of external referencing that fall outside of the Court's normative justifications.

By identifying shortcomings in the normative justifications offered, this article aims to raise greater awareness of both the potential and the shortcomings of the Court's approach. It also recommends how these shortcomings can be overcome, thus contributing to judicial legitimacy. Scholars such as Neuman and Lixinski have made a valuable contribution in identifying the expansion of the ACHR on the basis of ‘foreign’ sources and discussing the impact of this on the Inter-American human rights system, engaging with the notions of compliance and (sociological) legitimacy.¹³ However, the examination of the legal argumentation used by the IACtHR to strengthen its normative legitimacy, carried out in Section IV of this article, must be integrated into this more general debate. De Pauw and Killander have dedicated a significant part of their studies to the Court's usual justifications,¹⁴ and consideration of their work will be particularly valuable for this purpose.

Section II highlights the importance of this discussion by examining the perils presented by unclear interpretative methodologies for judicial legitimacy. Section III presents the typology of external sources identified by the case studies, whilst Section IV introduces the justifications offered by the IACtHR for its reliance on these sources. Section V then examines the extent to which the Court's reliance on external rules for interpretative guidance is supported by the VCLT. Section VI adopts the same approach regarding the Court's reliance on soft law. Finally, Section VII provides an overview of the article's main findings.

II. THE LEGITIMACY OF COURTS: AN ISSUE OF SUBSTANTIVE OUTCOMES ALONE?

Neuman criticizes the IACtHR's extensive use of external sources and argues that it ‘has become too divorced from the consensual aspect of a regional human rights convention in its interpretative practice’.¹⁵ Such criticism is often presented alongside a comparison with the approach of the European Court of Human Rights (ECtHR), which relies on consensus discerned through comparative law techniques. Scholars have often identified this difference as the root cause of what they contend is a threat to the legitimacy of the IACtHR. In turn, the IACtHR appears to rebut this criticism by constantly referring to the VCLT. Its interpretative methods, it argues, are a direct application of the VCLT and Article 29 of the ACHR.

Given the challenges of democratic legitimation of judicial lawmaking, von Bogdandy and Venzke recommend a discursive strategy that strives to provide a justification along the lines of Articles 31 and 32 of the VCLT for what they call ‘creative lawmaking’.¹⁶ Formally, the IACtHR's approach appears to match this recommendation. Therefore, its discursive strategy could be expected to help counter criticism against its being engaged in ‘creative lawmaking’ and, as a result, strengthen the Court's legitimacy. Yet such an outcome depends on there being a causal link between interpretative methodology and judicial legitimacy. In other words, does methodology matter when considering the legitimacy of courts and, if yes, to what extent and to whom?

When analysing the relationship between judges and their audiences, Baum argues that, whilst judges are influenced by the opinions of others, ‘others’ are not always the general public. His main argument is that judges are likely to take more account of certain sets of people, such as members of the legal profession, certain political and ideological groups, and the media. Lawyers are in ‘an especially good position to evaluate judges’ work’.¹⁷ Judges themselves might have an identification with particular political and ideological groups.¹⁸ Finally, ‘the mass media and the legally oriented news media may be important as visible reviewers of [judges’] performance’.¹⁹ The common thread between these three groups is that they all belong to the social, economic and political elites of the State. Baum contends that the collective views of these three groups often matter the most to judges. In other words, the most salient audiences for judges, at a personal level, might be what Bourdieu's relational sociology identifies as agents having most effect on the decisions taken in this field of power.²⁰

Baum's theory gains force when one recalls that, compared to the number of cases decided by courts, relatively few are widely discussed by the general public.²¹ In turn, certain elite groups—particularly those in the legal profession—are more likely to evaluate judges’ work on a regular basis and pay attention to the decision-making process and the procedural tools used by courts to reach their decisions. For a legal audience, interpretative methods and their consonance with sound theories and norms of legal hermeneutics are important.

The IACtHR is aware of the relevance of hermeneutics to a legal audience, and it speaks directly to the latter in its case law. When engaging in evolutive interpretation, the Court consistently refers to the VCLT and Article 29 of the ACHR. A discussion of the law of treaties is directly relevant to a legal audience, and it is reasonable to assume that the sections of the Court's judgments dedicated to hermeneutics are targeted at this audience. A notable example of this practice is the inclusion of a section called ‘Interpretative Criteria’ in the Advisory Opinion on Gender Identity, and Equality and Non-Discrimination of Same-Sex Couples,²² in which the Court engaged with a politically charged topic that was intensely debated among the States in the region. The opening section of its analysis, before any discussion of the substantive rights safeguarded in the ACHR, focused on the VCLT and Article 29 of the ACHR. This was an innovative structural approach when compared to the previous case law of the Court concerning the evolutive interpretation of the Convention.

When the Court discusses rules of treaty interpretation before engaging in evolutive interpretation, it is addressing its own normative legitimacy, that is, ‘the right to rule according to predefined standards’.²³ This article argues that this direct dialogue with the rules of treaty interpretation might, under certain circumstances, surpass the limits of normative legitimacy and so potentially affect the Court's sociological legitimacy, that is, the extent to which the tribunal is perceived as legitimate.²⁴

In certain cases, such as when the IACtHR analyses the evolution of particularly contentious or contested rights, there will be an ever-higher degree of scrutiny of the Court's decisions by legal experts. Irrespective of whether or not these experts agree with the outcome of a dispute, a sound interpretative methodology may engender goodwill towards the Court (diffuse support) even in the absence of support for the specific outcome of a case (specific support).²⁵ Similarly, the absence of a sound interpretative methodology might significantly weaken the immediate support given by legal experts to the Court's decision and, in the long term, weaken their support of it. The absence of a sound interpretative methodology in key cases may be extensively discussed by legal experts in media outlets, possibly resulting in—or being an accelerating factor for—the erosion of the very legitimacy of the tribunal in the eyes of the population in general. In other words, a sound interpretative methodology might indirectly impact the sociological legitimacy of courts.

In conclusion, if courts are to interpret norms in an evolutive manner, they are likely to engage with topics that are politically controversial. Yet, if it is the intent of the Court to fill legal gaps while guarding its legitimacy, it is important that it employs a sound interpretative methodology. The latter has the potential to impact the support the Court receives from legal experts and, in salient cases, to accelerate changes in the level of support it receives from the mass media and general public.

III. CASE STUDIES: IDENTIFYING A TRIPARTITE TYPOLOGY OF EXTERNAL SOURCES

To understand the IACtHR's use of external sources and the justifications provided by the tribunal for its doing so, the Court's jurisprudence concerning the rights of the child, IHL and LGBTI rights has been examined. This analysis reveals that the wide range of external sources used by the IACtHR falls into three categories: (i) norms binding upon the respondent State; (ii) norms not binding upon the respondent State; and (iii) soft law.

External sources binding upon the respondent State have been central to the Court's case law concerning the rights of the child. Article 19 of the ACHR guarantees to every minor the right to measures of protection required by their condition as a child, without explaining what would constitute such special measures. Accordingly, the IACtHR has consistently looked to the CRC for guidance when interpreting the words ‘measures of protection’.²⁶ One example is the case of “Street Children” (Villagrán-Morales et al) v Guatemala, concerning the abduction, torture and murder of children living on the streets of Guatemala City. The IACtHR used the CRC as a source to read positive obligations, including socio-economic obligations, into Article 19 of the ACHR to protect abandoned or exploited children.²⁷ Another example is the case of Gelman v Uruguay, concerning the abduction and enforced disappearance of a mother in the advanced stages of her pregnancy and the decision of Uruguay to give away her newborn child forcefully. The Court relied on Article 6 of the CRC, together with Articles 4(1) and 19 of the ACHR, to read into the ACHR the human right to identity.²⁸

The IACtHR has made explicit reference to the ratification of the CRC (regionally and worldwide). The wide acceptance of this instrument has enabled the Court to identify an international consensus on the principles that it sets.²⁹ In addition, the Court has relied not only on the CRC itself but also on the work of the UN Committee on the Rights of the Child.³⁰

The Court's case law on IHL has also been influenced by reliance on external rules binding upon the respondent State, particularly the 1949 Geneva Conventions, their Additional Protocol II (AP II) and customary IHL, all of which have been important sources used by the Court for the purposes of the evolutive interpretation of the ACHR's obligations in times of armed conflict.³¹ However, the Court's evolutive interpretation of the ACHR has not only been guided by hard law. Soft-law instruments have also been major sources. Among the three fields of law covered by this article, the most notable examples of the relevance of these instruments to the evolutive interpretation of the ACHR come from the Court's case law on LGBTI rights. Atala Riffo and Daughters v Chile is one such example, concerning a mother's loss of custody of her two daughters because of her sexual orientation. In its judgment, the IACtHR relied heavily on soft law to find, for the first time in its jurisprudence, that sexual orientation is a category entitled to protection from discrimination.³² Extensive references to soft-law instruments are also found in the Advisory Opinion on Gender Identity, and Equality and Non-Discrimination of Same-Sex Couples,³³ in which the Court justified their use with reference to the VCLT and Article 29 of the ACHR.

Finally, the case studies show that hard-law documents not binding upon the respondent State have been used by the IACtHR for the purposes of evolutive interpretation.³⁴ However, empirical analysis suggests that such sources might play a secondary role when compared to the other two types of external sources in the particular fields of law examined in this article.

As a rule, the Court provides a single and overarching justification for its reliance on all three types of external sources, namely, that the method is in conformity with Article 29 of the ACHR and with Article 31 of the VCLT. The VCLT is not mentioned in the ACHR's text, but it is a primary element in the Court's discourse when justifying its interpretative methodology. This constant and explicit reference to interpretative methodology appears to represent a persuasion strategy and a tool used by the Court to highlight its normative legitimacy. Despite this strategy, the Court is often criticized for its reliance on external referencing. As such, the sections below will challenge the Court's justification in light of the law of treaties and theories on treaty interpretation.

IV. THE COURT'S USUAL JUSTIFICATION

Courts may adopt different interpretative approaches to determine whether a human rights treaty has evolved at a certain point in time and the extent to which it has done so. The approach that is generally associated in scholarly work with the IACtHR is the ‘universalist’ approach. This derives from the theory of the coexistence and coordination of human rights mechanisms, a theory that is discussed by many scholars, including Cançado Trindade. Trindade was a judge at the IACtHR from 1995 to 2006, served as the Court's Vice-President from 1998 to 1999 and as President from 1999 to 2003. Accordingly, it is unsurprising that the practice of the Inter-American Court is deeply influenced by this theory. Indeed, it is in the context of universality and the unity of human rights that the universalist approach emerges.

This approach involves integrating different instruments that protect human rights into the process of treaty interpretation. When relying on universality, consensus is no longer the basis for identifying an evolution in human rights treaties and is thus not a requirement for reliance on external sources. Instead, it is the perceived conceptual unity of human rights that justifies the technique of external referencing. This allows for a broad range of instruments to be used for interpretative guidance and considers them sufficient to justify the evolutive interpretation adopted.

The universalist approach is often associated with the IACtHR, whereas the regional consensus approach is often associated with the ECtHR. However, contrary to what is generally posited by scholars, the empirical research and interviews conducted in this study indicate that regional consensus does play a role at the IACtHR.³⁵ What is particular to the Inter-American System is not the alleged irrelevance of regional consensus but rather the Court's relativization of the role and weight of consensus in the context of evolutive interpretation. For the IACtHR, identifying consensus is not a conditio sine qua non for evolutive interpretation of the ACHR. When a legal gap cannot be filled by consensus via comparative law but can be filled through reliance on the universalist approach, it is the latter that will prevail. More often than not, this relativization of consensus occurs in the Inter-American System of Human Rights, resulting in a broad use of external sources. When used by the IACtHR, consensus is generally a factor which strengthens a finding that is also supported by the universalist approach. These two approaches may also be in dialogue with each other in the form of an ‘internationalized consensus’, a concept that will be explained in Section V.

In any case, when relying on external sources, the IACtHR offers a three-pillared justification: first, it highlights that the sources used integrate the corpus juris of international human rights law, and it makes general and vague references to Article 29 of the ACHR and the VCLT; secondly, it explains its reliance on external sources on the basis of the pro persona principle; thirdly, it argues that human rights instruments are considered to be part of the context of the ACHR and, accordingly, their use is supported by Article 31 of the VCLT. These three pillars are often used simultaneously by the Court.

This section will focus on the two justifications that have been most clearly and specifically delimited by the Court itself, these being pillars two and three. Because the first pillar is vague and is generally repeated ‘as a sort of necessary and obliged international classical mantra’,³⁶ hypotheses have been identified that could link specific elements of section 3 (‘Interpretation of Treaties’) of the VCLT and Article 29 of the ACHR with the Court's use of external sources. These hypotheses will be tested in Sections V and VI below.

A. The Pro Persona Justification

The pro persona principle is codified in Article 29(b) of the ACHR. This principle establishes that, when two instruments which are applicable to the same State regulate the same right, the one that is the most favourable to the alleged victim prevails.³⁷ Furthermore, when a given provision can be interpreted in more than one way, the interpretation most favourable to the individual must be the one adopted. It is this second dimension that allows for external referencing to be used in the dynamic interpretation of the ACHR.

In 1997 and 2000, the Inter-American Commission on Human Rights (IACommHR) argued that the pro persona principle allowed the Commission and the Court to hold a State responsible for violations of norms adopted outside the Inter-American System of Human Rights. More specifically, it argued for the direct applicability of IHL.³⁸ However, the IACtHR disagreed with the Commission's conclusions and held that external sources may only be used ‘as elements for the interpretation of the American Convention’.³⁹ Thus the pro persona principle is limited to being used in treaty interpretation and cannot expand the jurisdiction of the IACtHR beyond Inter-American instruments.

The wording of Article 29(b) of the ACHR and the case law of the IACtHR⁴⁰ indicate that it is the binding nature of the external rule before the respondent State that is relevant for reliance on external rules based on the pro persona principle. Accordingly, when a State ratifies a more beneficial treaty or even adopts a domestic law that affords greater protection to a certain right, the obligations applicable to that State which derive from the ACHR expand in light of Article 29 of the ACHR.⁴¹ This evolution of the ACHR does not bind all States parties to the Convention, but only the State that bound itself to a higher standard of protection.⁴² One notable example is provided by the Hacienda Brasil Verde Workers v Brasil case. This case concerned forced labour and debt servitude in Hacienda Brasil Verde, located in Pará, Brazil.⁴³ Brazil argued that its domestic legislation set a higher standard of protection than the ACHR and, accordingly, that acts committed in Pará could amount to forced labour in Brazilian law but not in international law.⁴⁴ The Court rejected the claim that the practice in question could not amount to a violation of the ACHR by noting, first, that the prohibition against forced labour established under Brazilian domestic law should not be characterized as ‘too wide’ or distinct from the one derived from the ACHR. Secondly, the Court held that when a given State adopts a norm that sets a higher standard of protection than that set out in the American Convention, the Court will apply the norm more favourable to the alleged victim.

The Court's conclusion that it may take into consideration rules adopted by the respondent State that set a higher standard of protection to a right safeguarded in the ACHR is consonant with Article 29 of the ACHR.⁴⁵ Article 29(b) expressly states that the ACHR shall not be interpreted as restricting rights enshrined in the domestic laws of the State. Furthermore, to argue otherwise would equate to claiming that the ACHR authorizes States to settle for a minimum threshold of human rights protection. Such an interpretation would undermine the effectiveness of the ACHR and the realization of its object and purpose. Accordingly, reliance on more beneficial rules applicable to the respondent State for the purposes of interpretative guidance is not only in accordance with Article 29 of the ACHR but also Article 31(1) of the VCLT, which determines that a treaty shall be interpreted in light of its object and purpose. Yet, as explained above, this justification only covers rules binding upon the respondent State.

B. Human Rights Law as Part of the Context of the ACHR

The Court also suggests that the use of external sources is justifiable because human rights law as a whole should inform the interpretation of the ACHR. It is interesting to recall how this justification arose. The Court has repeatedly stated that its broad use of external sources is consistent with the VCLT, but it did not provide a more detailed explanation until the Advisory Opinion on the Right to Consular Information. In this Advisory Opinion, it explained that ‘the interpretation of a treaty must take into account not only the agreements and instruments related to the treaty … …, but also the system of which it is part’.⁴⁶ In Serrano-Cruz Sisters v El Salvador, the Court repeated the dictum from the Advisory Opinion on the Right to Consular Information, but substituted the word ‘system’ for the word ‘context’, linking the latter to Article 31(3) of the VCLT.⁴⁷ It went even further in the 2012 case of “In Vitro Fertilization” (IVF) v Costa Rica, referring to human rights law as an element of the context of the ACHR.⁴⁸

To assess the limits of this justification, it is first necessary to define what constitutes the ‘context’ of a treaty. Article 31(2) of the VCLT provides that the context comprises: (i) a treaty's text, preamble and annexes; and (ii) agreements or instruments made in connection to the conclusion of the treaty and related to it—in simple terms, unilateral documents that have a connection to the treaty that is being interpreted.

Human rights treaties, customary international law and documents of international organizations—external sources often used by the IACtHR—cannot be classified as unilateral documents. In addition, they have not been made ‘in connection with the conclusion’⁴⁹ of the ACHR and do not directly relate to the Convention. Therefore, they do not fall within Article 31(2)(a) and (b) of the VCLT. The question then becomes whether they could be considered to form part of the text of the treaty, its preamble and annexes, under Article 31(2).

The ACHR's Preamble makes a brief reference to external sources of international human rights law. In its third paragraph, the Preamble remarks that the principles safeguarded therein have been set forth in the Universal Declaration of Human Rights (UDHR) and ‘have been reaffirmed and refined in other international instruments, worldwide as well as regional in scope’.⁵⁰ While this reference is rather vague, it is clarified by the text of the ACHR.

For the purposes of Article 31(2) of the VCLT, the text of a treaty provision is comprised of: (i) the text of the specific article that is being interpreted; and (ii) other articles, sections or chapters of the treaty in question. The World Trade Organization (WTO) has called the former the ‘immediate context’ and the latter the ‘broader context’.⁵¹ The ‘immediate context’ of an article in the ACHR encompasses the terms used in the sentence and paragraph under examination, as well as all other paragraphs that comprise the article in question. The ‘broader context’ of a provision is the entirety of the text of the treaty. As explained by Villiger, taking the broader context of a provision into consideration is logical, because ‘[t]reaty terms are not drafted in isolation, and their meaning can only be determined by considering the entire treaty text’.⁵² Accordingly, the interpretation of any article of the ACHR must take into consideration the other articles that integrate it, one of which is Article 29(b), (c) and (d), which sets guidelines for the interpretation of the ACHR and makes express reference to external sources.

Article 29(b) establishes that the proper interpretation of the ACHR shall take into consideration norms more beneficial to the alleged victim enshrined in the domestic laws of the State and in other treaties ratified by it. Article 29(c) provides the same in relation to other rights and guarantees that are inherent to the human personality or derived from representative democracy. It is important to recall that inherent rights and guarantees might be safeguarded under treaties to which the respondent is not a party, as well as in customary rules and even rules of jus cogens.⁵³

Accordingly, through cross-referencing, Article 29(b) and (c)—which are themselves within the broader context of every article of the ACHR—expands the context of the ACHR to encompass the domestic laws of the States parties, other treaties ratified by them, and rights and guarantees that are inherent to the human person or derived from representative democracy. Article 29(b) does not refer to treaties binding upon all States parties, but rather to treaties to which ‘one of the said states is a party’.⁵⁴ The practice of the IACtHR has clarified that this refers to the respondent State. Furthermore, neither Article 29(b) nor (c) makes any reference to the principal purpose of external sources. The practice of the Court shows that it is not necessary that they be human rights rules as such,⁵⁵ but only that they concern human rights.⁵⁶

Finally, Article 29(d) provides that the ACHR shall not be interpreted in a manner that excludes or limits ‘the effect that the American Declaration of the Rights and Duties of Man or other international acts of the same nature may have’.⁵⁷ The American Declaration was approved by the OAS a few months before the UDHR, thus constituting the first human rights instrument of this kind. It is not an international treaty but a soft-law instrument—more specifically, a declaration of international law. For an instrument to have the same nature as the American Declaration, it must be an international declaration and thus only the latter can be the object of the cross-referencing technique used in Article 29(d) of the ACHR.

Including in the Preamble and in Article 29 an express reference to fundamental rights and guarantees found in other instruments suggests there was an intention that these external sources be applied in the context of the Convention. As the IACtHR explained in its Advisory Opinion on “Other Treaties” Subject to the Consultative Jurisdiction of the Court, ‘[a] certain tendency to integrate the regional and universal systems for the protection of human rights can be perceived in the Convention’.⁵⁸ The Court has identified this tendency both in the Preamble of the ACHR and in several of its provisions, in particular Article 29, which ‘clearly indicates an intention not to restrict the protection of human rights to determinations that depend on the source of the obligations’.⁵⁹

In summary, the text and the Preamble of the ACHR do not integrate international human rights law as a whole into the broader context of the American Convention. Rather, they integrate certain external rules that concern the protection of human rights into the broader context of the Convention. These are: (i) treaties ratified by the respondent State; (ii) rules (be they conventional or not) that safeguard rights and guarantees that are inherent to the human person or derived from representative democracy; and (iii) international declarations. When determining whether a given external rule or declaration can be integrated into the ACHR context by means of cross-reference to Articles 29(b), (c) and (d), it is necessary to consider whether the rule in question: (i) concerns the protection of human rights; (ii) is capable of assisting the interpreter of the ACHR; and (iii) is either enshrined in a treaty ratified by the respondent State, is inherent to the human person or derived from representative democracy, or, alternatively, integrates an international declaration. If such conditions are met, these sources become part of the broader context of the ACHR, and their use for interpretative purposes is in conformity with Article 31(1) and (2) of the VCLT.

It could be argued that Article 29(c) of the ACHR also provides a legal basis for the use of any soft-law instrument. After all, while soft-law instruments do not generate rights or obligations, they might nonetheless specify the scope of legally binding instruments that protect inherent rights and guarantees of representative democracy.⁶⁰ However, Article 29(c) clearly refers to rights and guarantees, and not to standards or guidelines. The fact that a soft-law instrument clarifies the content of legally binding rights is not sufficient to include it in the broader context of the ACHR. Only an express reference to those instruments in the provisions of the ACHR would be capable of doing so. Therefore, while soft-law instruments other than international declarations have an undeniable legal relevance, the legal basis for their use cannot rely on Article 29(c) of the ACHR.

C. Conclusions on the Limits of Pillars Two and Three of the Court's Usual Justification

To conclude, this section demonstrated that the Court's usual justification for drawing on external sources is sufficient to validate reliance on: (i) external rules binding upon the respondent State; (ii) other rules that safeguard rights or guarantees that are inherent to the human personality or derived from representative democracy; and (iii) international declarations. However, it does not constitute an all-encompassing justification for each and every type of external source that the Court uses. Having demarcated the limits of the Court's specific explanations and demonstrated that one single justification is not sufficient to validate reliance on all types of external sources, this article will now challenge the Court's more general argument that its use of external sources is in conformity with the VCLT. Section V will analyse the extent to which Article 31 of the VCLT could justify the Court's reliance on external rules, and Section VI considers the Court's use of non-binding sources. Taken together, Sections V and VI consider the extent to which reference to external sources can be justified on the basis of the VCLT.

V. RELIANCE ON EXTERNAL RULES IN LIGHT OF THE VCLT

The IACtHR has relied on a broad range of rules adopted outside the Inter-American human rights system. This includes both customary international law and other treaties, whether or not applicable to the respondent State. Examples include: the International Covenant on Civil and Political Rights (ICCPR);⁶¹ the UN Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment;⁶² the CRC;⁶³ AP II;⁶⁴ and customary rules of IHL.⁶⁵

This section will consider two potential legal bases for reliance on external rules. First, it will discuss Article 31(3)(c) and its reference to relevant rules of international law applicable to the parties. Secondly, it will discuss whether the use of treaties not ratified by the respondent State can be justified if there is an established international consensus concerning the rights in question.

A. External Rules Applicable to the Respondent State and Article 31(3)(c) of the VCLT

Under Article 31(3)(c) of the VCLT, other rules shall be taken into consideration during the process of treaty interpretation, provided they are: (i) rules of international law;⁶⁶ (ii) relevant to the interpretation of the ACHR—a requirement that includes, but goes beyond, norms that have as their primary object human rights; and (iii) applicable to the parties. Only if all of these requirements are met can Article 31(3)(c) of the VCLT support the use of the external source in question. The most controversial requirement is the third, that the rule must be ‘applicable in the relation between the parties’. Accordingly, Sections V.A.1 and 2 will now examine this criterion.

1. A temporal requirement: applicable when?

A question that arises from the word ‘applicable’ is whether Article 31(3)(c) refers to rules in force at the time of the adoption of the treaty or to rules in force at the time of interpretation. According to Villinger, the applicable rules are those in force at the time of the interpretation.⁶⁷ Villinger's understanding is based on the drafting history of the VCLT. He notes that the 1964 Draft Articles on the Law of Treaties make express reference to ‘the general rules of international law in force at the time of its conclusion’.⁶⁸ Yet in 1966 the International Law Commission (ILC) decided to delete those words.⁶⁹ In Villinger's opinion, this indicates that this requirement was dropped and thus was not incorporated into the Article. Judge Robinson applied the same legal reasoning in his declaration in the International Criminal Tribunal for the former Yugoslavia's (ICTY) Furundžija case.⁷⁰ In light of the ILC's commentaries to the Draft Articles on the Law of Treaties, he asserted that the modification was made ‘so as to take account of “the effect of an evolution of the law on an interpretation of legal terms in a treaty”’.⁷¹ He concluded that ‘the relevant rule of international law need not have been in force at the time of the conclusion of the treaty being interpreted; it need only be in force at the time of the interpretation of the treaty’.⁷²

A different theory was put forward by Yasseen. In the absence of express reference in Article 31(3)(c) to the inter-temporal nature of external rules, Yasseen argues that one should look for guidance in general public international law. In international law, rules of jus cogens limit the freedom of States in treaty-making. In accordance with Article 64 of the VCLT, a treaty becomes void if it is contrary to a peremptory rule of international law, even if such a rule emerged after the adoption of the treaty. Jus cogens norms thus have what this article calls an ‘atemporal applicability’ as relevant rules for treaty interpretation. While jus cogens norms have this ‘atemporal applicability’, the same does not apply to ordinary norms. With regard to the latter, Yasseen argues that one should look at the treaty itself and at the intention of the parties. If the text of the treaty does not regulate this issue, its object and purpose might.

Yasseen's approach seems to accord with the ILC's commentaries to the Draft Articles on the Law of Treaties. According to the ILC, ‘the relevance of rules of international law for the interpretation of treaties in any given case was dependent on the intentions of the parties’, and the ‘correct application of the temporal element would normally be indicated by interpretation of the term in good faith’.⁷³ Whether an interpretation is conducted in good faith must be assessed by considering the object and purpose of a treaty.⁷⁴

An interpretation in good faith and in the light of the object and purpose of the ACHR indicates that its parties did not intend to create a stagnant instrument, the interpretation of which remains indifferent to the evolution of international law. Quite the contrary, they created a dynamic instrument, which aims to protect the fundamental rights and freedoms of the human person effectively, from the date of its signing in 1969 to the present, reflecting modern-day issues. Therefore, the fact that a rule emerged after 1969 should not be perceived as an obstacle for it to be used in the process of interpretation of the ACHR.

2. Applicable to whom?

A final question with regards to the ‘applicable’ requirement is whether the words ‘applicable between the parties’ refer to the parties to a contentious case or to all States parties to the treaty in question. In other words, if an external treaty is ratified by the State party to a case before the IACtHR but not by all States parties to the ACHR, can it be used by the Court when interpreting the ACHR?

In “Juvenile Reeducation Institute” v Paraguay, a case concerning the living conditions of children detained in a juvenile re-education institute in Paraguay, the Court relied on the CRC and the Protocol of San Salvador to interpret the content and scope of the words ‘measures of protection required by … [a child's] condition as a minor’ in Article 19 of the ACHR.⁷⁵ According to the Court, ‘[t]hese instruments and the American Convention are part of a very comprehensive international corpus juris for the protection of children that the Court must honor’.⁷⁶ Accordingly, the IACtHR took them into consideration when interpreting Article 19 of the ACHR and, as a result, found that States must take measures to protect children's economic, social and cultural rights.⁷⁷ However, the Court first noted that Paraguay had ratified both the CRC and the Protocol of San Salvador.⁷⁸ Only then did it consider these two instruments when interpreting Article 19 of the ACHR.

The Court's reliance on the CRC in “Juvenile Reeducation Institute” v Paraguay is clearly justified under Article 31(3)(c) of the VCLT. Not only is the CRC ratified by Paraguay, but it is ratified by—and thus applicable to—all the States that have accepted the jurisdiction of the IACtHR. However, the IACtHR does not always draw on external rules that have been ratified by all States subject to its jurisdiction. In Santo Domingo Massacre v Colombia, concerning the bombardment on the village of Santo Domingo perpetrated by the Colombian Air Force during an armed conflict between the State authorities in Colombia and the Colombia Revolutionary Armed Forces (FARC), the Court relied on IHL.⁷⁹ When determining whether Colombia had complied with Article 21 of the ACHR (the right to property), the IACtHR found it ‘useful and appropriate to interpret the scope of Article 21 of the American Convention’⁸⁰ using AP II and Rule 7 of the Customary Rules of IHL, which prohibit attacks directed against civilian objects and looting.⁸¹ AP II has been ratified by all but one State that accepts the jurisdiction of the IACtHR, this being Mexico, which has neither ratified nor signed the Protocol. This means that that instrument is not applicable to all States parties to the ACHR. The question then becomes whether this single case of non-ratification prevents the IACtHR from relying on AP II, even though it is applicable to all other States parties to the ACHR.

Yasseen argues that only rules that are common to (all) the parties of the treaty can have a ‘direct juridical effect upon said treaty’.⁸² The same understanding was upheld by the WTO Panel Report to the dispute Measures Affecting the Approval and Marketing of Biotech Products.⁸³ However, in the 2006 Report of the ILC Study Group finalized by Koskenniemi, the WTO Panel Report was (rightly) criticized by the ILC itself.⁸⁴ In the words of the ILC Study Group, ‘[b]earing in mind the unlikeliness of a precise congruence in the membership of most important multilateral conventions, it would become unlikely that any use of conventional international law could be made in the interpretation of such conventions’.⁸⁵ Upholding such a restrictive view would have ‘the ironic effect that the more the membership of a multilateral treaty such as the WTO covered agreements expanded, the more those treaties would be cut off from the rest of international law’.⁸⁶ The practical result would thus be ‘the isolation of multilateral agreements as “islands” permitting no references inter se in their application’, in a manner that seems ‘contrary to the legislative ethos behind most of multilateral treaty-making and, presumably, with the intent of most treaty-makers’.⁸⁷ Additionally, reference to important treaties, ‘which represent the most important elaboration of the content of international law on a specialist subject matter’ and which enjoy an almost universal acceptance, would be precluded.⁸⁸

Transposing this reasoning to the Inter-American context, adopting the restrictive understanding of Article 31(3)(c) put forward by the WTO Panel would mean, for example, that the IACtHR would be prevented from relying on the UN Convention Against Torture, because Suriname, a State Party to the ACHR which has accepted the jurisdiction of the Court, is not a party to this UN treaty.⁸⁹ In the opinion of the ILC Study Group, a better solution is to allow reference to be made to external treaties that are binding upon the parties in dispute.⁹⁰ However, the ILC Study Group calls for caution to be paid on the use of external sources not ratified by all parties to a treaty if that treaty establishes obligations erga omnes partes, that is, if it protects collective interests instead of being limited to promoting individual and reciprocal interests. For treaties containing obligations erga omnes partes, such as human rights conventions, the ILC Study Group affirms that it might be useful to ‘take into account the extent to which that other treaty relied upon can be said to have been “implicitly” accepted or at least tolerated by the other parties’, that is, that it can be reasonably considered to express the common understanding of all parties.⁹¹

Young believes that such an implicit acceptance would take place, for example, when States have signed, but not ratified, a given treaty.⁹² This argument can be illustrated by applying it to a hypothetical case. Today, all States parties to the ACHR have ratified the CRC. Yet, what would happen if the United States of America (US), which has not ratified the CRC, decided to become a party to the American Convention? Would the Court now be prevented from using the CRC to interpret the ACHR, despite having consistently relied on it previously? If so, this would mean that an expansion in the number of ratification and in the jurisdiction of the Court could result in a diminution in the level of human rights protection. Such a result clearly undermines the effectiveness of the ACHR.

A different outcome emerges if implicit acceptance by all States parties suffices. For example, since the US has signed the CRC, according to Young this can be considered as tolerating or implicitly accepting this instrument. The requirement of ‘relative consensus’ endorsed by the ILC Study Group would thus be met, and the IACtHR would still be able to apply the CRC when interpreting the ACHR, despite the US not having ratified it.

A further question is what would arise if one day Brazil, a party to the ACHR that consented to the jurisdiction of the IACtHR, decided to denounce the ICCPR? Would the Court be prevented from relying on the ICCPR when interpreting the ACHR, even in cases not involving Brazil and despite the fact that all the remaining States parties to the ACHR have at least signed the Covenant? Some interesting theories have emerged addressing this problem.⁹³ However, for the present purposes, these need not be considered as other legal bases can be provided to support the use of rules of international law that are applicable to some, but not all, the States parties to the ACHR, as will be explained in the next subsection.

B. Interactive Interpretation and ‘Internationalized Consensus’

Consider once more the hypothetical situation in which Brazil, a party to the ACHR which has consented to the jurisdiction of the IACtHR, denounces the ICCPR. If Brazil is no longer bound by this convention, could the IACtHR nonetheless rely on this treaty when interpreting the ACHR in a case against Brazil? The ICCPR is widely accepted internationally. Even if Brazil or another OAS Member State decided to denounce it, the Covenant would continue to be ratified by very many States and contain rules on which there is a near-universal international consensus.

Consensus is not only regional. What is often called a ‘European consensus’ in the context of the European Convention on Human Rights is identified through a comparative analysis of the laws and practices of contracting parties to the treaty in question.⁹⁴ This is merely one type of consensus relevant for treaty interpretation. A second type is consensus identified not on the basis of the domestic practice of the States concerned but rather in light of other instruments of international law, be they international or regional in nature.⁹⁵ It is an ‘internationalized consensus’.⁹⁶

If the international community or, at a minimum, the great majority of States parties to the ACHR, have agreed that the standard of protection of a given human right has been raised, then there is a common understanding that the interpretation of the right is evolving. An interpretation that considers a consensus based on international treaties widely accepted by the regional or international community contributes to the effectiveness of the ACHR and is in accordance with its object and purpose. Furthermore, a consolidated consensus could arguably be regarded as subsequent practice, an element of treaty interpretation under Article 31(3)(b) of the VCLT that will be discussed in Section VI.A.

The IACtHR has already recognized international consensus in its 2002 Advisory Opinion Juridical Condition and Human Rights of the Child, when referring to the principles and institutions set forth in the CRC.⁹⁷ This article argues that a certain degree of consensus within the international community or, at a minimum, among the States parties to the ACHR, should have to be identified by the Court in order for it to rely on a treaty not ratified by the party to a case. After all, were the ACHR to be expanded based on the existence of any human rights treaty whatsoever, the States parties to the ACHR would be subject to greater juridical insecurity. This situation could undermine the Court's legitimacy, potently reduce State consent to its jurisdiction and hence weaken the furtherance of the object and purpose of the ACHR.

To be effective, the ACHR must evolve to respond to novel situations and changing circumstances. The existence of an internationalized consensus on a given issue highlights the emergence of a new social reality and the evolution of international law in a given direction. Insulating the ACHR from those developments would be detrimental to its effectiveness. Hence, in light of the object and purpose of the ACHR (Article 31(1) of the VCLT) and arguably Article 31(3)(b) of the VCLT (see Section VI.A), the IACtHR may expand the content and scope of a provision of the ACHR by taking into consideration instruments that are sufficient to support the existence of an internationalized consensus.

VI. RELIANCE ON SOFT-LAW INSTRUMENTS IN LIGHT OF THE VCLT

The IACtHR has relied on a wide range of non-binding instruments, including European directives and recommendations,⁹⁸ UN declarations and principles,⁹⁹ reports of UN special rapporteurs,¹⁰⁰ concluding observations,¹⁰¹ general comments,¹⁰² recommendations,¹⁰³ communications,¹⁰⁴ and reports¹⁰⁵ and final comments¹⁰⁶ of UN treaty bodies. The 2012 Santo Domingo Massacre v Colombia case mentioned above provides an example of the Court relying on soft-law instruments. In its judgment, the Court expands the content and scope of Article 22 (freedom of movement and residence) to cover the right not to be forcibly displaced within a State party to the Convention.¹⁰⁷ The UN Guiding Principles on Internal Displacement was the only source expressly drawn on by the Court to justify expanding the scope and content of that provision.¹⁰⁸ Similarly, in Atala Riffo and Daughters v Chile, the Court concluded that sexual orientation constitutes a prohibited ground of discrimination under the ACHR after taking into consideration a series of soft-law instruments.¹⁰⁹

Soft-law instruments have played a significant role in updating the text of the ACHR to reflect present-day realities. The Court's case law on LGBTI rights is a paradigmatic example of the contribution of soft law to an evolving interpretation of the ACHR. However, the question that remains is whether the Court's justifications for doing so are normatively acceptable.

A. Soft-Law Instruments as Subsequent Practice

In the Japan – Taxes on Alcoholic Beverages case, the WTO Appellate Body affirmed that ‘the essence of subsequent practice in interpreting a treaty has been recognized as a “concordant, common and consistent” sequence of acts or pronouncements which is sufficient to establish a discernible pattern implying the agreement of the parties regarding its interpretation’.¹¹⁰ In other words, subsequent practice does not refer to an isolated act, but to a sequence of acts that establishes a discernible pattern. This pattern must show the agreement of the States parties to the treaty with regard to its content and scope.

In its 1966 commentaries to the VCLT, the ILC noted that when drafting Article 31(3)(b) it intended to refer to the practice of the parties of the treaty as a whole. Yet this does not mean that each and every party must have engaged in a given practice for it to qualify as subsequent practice. Rather, in the words of the ILC, ‘it suffices that [the State party] should have accepted the practice’.¹¹¹ Tacit acceptance or acquiescence would thus suffice in order to identify a common practice among the parties.¹¹²

To argue for a stringent definition of ‘subsequent practice’ which demands that all parties must have actively engaged in a given practice would not only be contrary to the understanding of the ILC itself, but also be detrimental to the effectiveness of human rights treaties. Such treaties would be undermined if dynamic interpretation could be paralysed by a few groups of States not actively engaging in a practice that is consistent and common among the other parties.

The following subsections will explore four hypotheses concerning how certain soft-law instruments might qualify as subsequent practice. The first is that the practice of treaty bodies is tantamount to the practice of States parties themselves. The second hypothesis was put forward by the ILA, according to which the responses of States to the work of human rights treaty bodies may be regarded as State practice. The third examines resolutions of international organizations and considers whether their adoption is evidence of State practice. The fourth and last hypothesis was raised by Killander, according to whom resolutions may be evidence of an emerging consensus on a given issue.

External referencing to the work of treaty bodies is considered under this section since such documents (including outcomes of individual complaints, analysis of State reports, general comments or recommendations) are generally¹¹³ not binding.¹¹⁴ However, the classification of the work of the UN treaty bodies as soft law is not universally accepted.¹¹⁵ As there is no consensus on their legal nature, more resistance to the referencing of these documents as sources for the evolutive interpretation of the ACHR is to be expected, particularly when described as soft-law instruments. Accordingly, this section analyses the justifications for reliance on such materials even though they are considered as non-binding. Furthermore, the analysis carried out in Sections VI.A.1 and 2 below is equally applicable, whether or not these documents are considered as soft law.

1. The work of human rights treaty bodies as State practice

McGrogan argues that it would not be unfounded to perceive ‘the practice of the treaty bodies as constituting a devolved or delegated form of subsequent practice’,¹¹⁶ if this practice is met with acquiescence from the States parties to the treaty.¹¹⁷ This would indicate tacit endorsement of the practice of the treaty body by the States. McGrogan's approach is based on two factors. First, he considers that the travaux préparatoires of the VCLT indicate that tacit approval of a practice would be sufficient to indicate the existence of a subsequent practice. Secondly, he notes that certain treaties ‘entrust some organs with the competence to detail the content of treaty provisions requiring interpretation’.¹¹⁸

One could go even further than McGrogan and argue that the ratification of a treaty or additional protocol that creates a treaty body and entrusts it with the competence to interpret the terms of the treaty could, by itself, imply tacit approval of the practice of that organ. However, the work of treaty bodies is generally devoid of binding force,¹¹⁹ and their interpretations of the text of the treaty from which they derive their competence are, therefore, generally not binding on States parties. Therefore, it appears to be going too far to consider the practice of these organs as automatically equating to that of States parties and establishing ‘the agreement of the parties regarding its interpretation’.¹²⁰ On the other hand, if the practice of a treaty body is met with acquiescence from States parties, McGrogan argues that a tacit agreement among the parties could be identified and the practice of these organs could then amount to subsequent practice.¹²¹

However, the work of a treaty body may only be considered as subsequent practice as regards the interpretation of the specific treaty in relation to which it has competence. For example, the Committee on the Rights of the Child was established under the CRC, which provided it with the competence to monitor progress achieved by the States parties towards compliance with the Convention. Accordingly, if met with the acquiescence of the States parties to the CRC, the practice of the Committee might be considered as subsequent practice for the purposes of interpreting the provisions of the CRC, and the CRC only. The practice of the Committee is not in any sense subsequent practice regarding the interpretation of other treaties, such as the ACHR. Still, as explained in Section V, the CRC is a source of relevant external rules applicable to the parties of the ACHR, and when the IACtHR relies on an external rule, it is relying on its content and scope, which can only be determined through interpretation. Hence, if the work of the Committee on the Rights of the Child is perceived as subsequent practice for the interpretation of the CRC, the IACtHR could properly rely on the Committee's work for the interpretation of the ACHR. In other words, the work of the Committee becomes relevant for the interpretative exercise conducted by the IACtHR as ‘second-degree’ interpretive guidance. One should note, though, that this theory is controversial, as it is far from settled that Article 31(3) of the VCLT can be interpreted as covering more than plain State practice.¹²²

2. State responses to the work of human rights treaty bodies as State practice

In its report to the Berlin Conference, the ILA affirmed that Article 31 of the VCLT was ‘written as if no monitoring body had been established by a treaty, as if no third-party interests existed, and as if it were only for other States to monitor each other's compliance and to react to non-compliance’.¹²³ Yet human rights treaties are different from regular multilateral treaties, and many of them establish an independent monitoring mechanism. In this context, the ILA considered that ‘relevant subsequent practice might be broader than subsequent State practice and include the considered views of the treaty bodies adopted in the performance of the functions conferred on them by the States parties’.¹²⁴ Alternatively, even if one adopts a traditional approach to the interpretation of human rights treaties, under which only the practice of States could constitute ‘subsequent practice’, the work of treaty bodies would still influence (although not constitute) subsequent practice for the purposes of Article 31(3)(b). After all, many States respond to general comments and recommendations of these organs. A positive response by a State or its acquiescence might constitute subsequent agreement which contributes to the establishment of a common agreement between the parties.¹²⁵ It must be noted that the ILA emphasizes that if the latter approach is preferred, interpreters will ‘depend on the results of a detailed analysis of how States parties [have] responded to that output’.¹²⁶

In this hypothesis put forward by the ILA, it would not be the work of the treaty bodies themselves that would constitute subsequent practice but rather the actions or omissions of the State when responding to them. It is not, though, the responses or the acquiescence of States that the IACtHR mentions during its interpretation of the ACHR; it is the soft-law instruments themselves. Therefore, the theory put forward by the ILA does not provide support for the Court's referencing of external sources.

3. Adoption of a resolution as subsequent State practice

Turning now to the use of non-binding resolutions in treaty interpretation, it may be argued that resolutions of international organizations can be considered as subsequent practice because the act of voting in favour of a resolution and thus contributing to its adoption may itself constitute State practice. Yet this hypothesis faces a central objection: when States adopt a resolution, they could be acting in their capacity as members of an international organization. Accordingly, should the approval of such a resolution be regarded as State practice or as the practice of the international organization in question?

Conforti argues that certain documents approved within an international organization should be regarded as State practice, for example, declarations of principles, such as the UDHR and the Millenium Declaration. In the words of Conforti, ‘[w]ith regard to customary law, the Declarations play a role in the process of its formation, as State practice, as the synthesis of the attitudes of States that adopt them and not as formal acts of the UN’.¹²⁷ Conforti's theory is of interest when considering whether declarations of principles form subsequent practice for the purposes of the treaty interpretation. However, it is an insufficient justification for the purposes of this article because, when interpreting the ACHR, the IACtHR has not limited itself to using declarations of international organizations.¹²⁸

By applying Conforti's understanding to the IACtHR's reliance on non-binding resolutions, this section concludes that the use of a certain type of resolution—namely, those adopting declarations of principles—is normatively legitimate. Alternatively, even if Conforti's approach is not adopted, the Court's reliance on declarations adopted by international organizations would still be legitimate for the reasons outlined in Section IV.B, that is, the cross-referencing technique used in Article 29(d) of the ACHR and so reflecting the broader context of the ACHR.

Treating other types of non-binding resolutions as State practice is more controversial, given that States may be acting in their capacity as members of an international organization when voting in favour of the adoption of a resolution. As explained in Sections VI.A.1 and 2, it is far from generally accepted that subsequent practice includes the adoption of instruments by non-States parties, such as international organizations, for the purposes of the rules of treaty interpretation. Nonetheless, even if non-binding resolutions adopted by international organizations are not subsequent practice for the purposes of Article 31(3)(b) of the VCLT, the individual votes of States could be. If this approach were to be adopted, the IACtHR would need to assess the votes cast by States parties to the ACHR when adopting the resolution in question, in order to identify whether there was an (explicit or tacit) agreement of the parties. The weight of abstentions and votes against a resolution will be lessened when there is an international consensus concerning the newly developed approach to the interpretation of the right in question, as explained in Section VI.A.4 below.

4. Resolutions of international organizations as evidence of consensus

A different hypothesis has been put forward by Killander, who suggests that a resolution could be considered as ‘subsequent practice’ under Article 31(3)(b) of the VCLT because it ‘could illustrate emerging consensus on an issue’.¹²⁹ In connection with this claim, Killander argues for a modern conception of customary international law which places emphasis on the element of opinio juris and provides a more flexible definition of State practice. In his words, ‘opinio juris and verbal State practice can in itself form customary international law’.¹³⁰

Resolutions of international organizations may be regarded as evidence of opinio juris. They represent a common ground reached by a group of States and may indicate new developments in international law. This holds true particularly for resolutions of the General Assembly of the UN, the approval of which means that an agreement (unanimous or majoritarian) has been reached among its 193 Member States. While a single resolution may be insufficient evidence of opinio juris, particularly when it was not adopted unanimously, a series of resolutions can be of greater relevance for the identification of an evolution in opinio juris and the emergence of a customary norm of international law.¹³¹

Others have also argued for a wider definition of State practice and a greater focus on opinio juris when considering important moral issues, such as international human rights law.¹³² For example, Wouters and Ryngaert, whose work has been drawn on by Killander, argue that, in the field of human rights law and IHL, opinio juris should ‘play a more important role than state practice’.¹³³ This does not mean that State practice does not play any role in the formation of customs of human rights law and IHL. Quite the contrary, even Wouters and Ryngaert recognize that ‘the existence of the customary rule in the opinio juris of States should still be confirmed by practice’.¹³⁴ However, they suggest that verbal as well as physical practice is relevant. When physical State practice is inconsistent, verbal State practice gains pre-eminence. Examples of the latter include public statements of State representatives in which they orally affirm or deny certain practices.¹³⁵

By directly referring to the work of Wouters and Ryngaert, Killander seems to build an argument for regarding resolutions of international organizations as opinio juris and thus being of value for the formation of international customary law. Coupled with verbal State practice, such as statements made by States when drafting and adopting resolutions, resolutions of international organizations could thus give rise to rules of customary human rights law. However, even if this is the case, it would be the verbal practice of States that would constitute subsequent practice for the purposes of treaty interpretation rather than the resolutions themselves, unless one took the view that the adoption of resolutions itself constitutes State practice, rather than being only the practice of the international organization. Yet, as explained in Section VI.A.3, there is no consensus on recognizing resolutions of international organizations as State practice.

In any case, Killander notes that resolutions ‘could illustrate emerging consensus on an issue’.¹³⁶ The question is, then, whether emerging consensus qualifies as subsequent practice in the sense of Article 31(3)(b) of the VCLT. Scholars such as McGrogan¹³⁷ and Yasseen¹³⁸ believe that Article 31(3)(b) refers to a concordant, common and consistent practice which establishes an agreement between the parties of the treaty. The fact that a consensus has not yet been reached seems to indicate that those requirements are not yet met. Conversely, a consolidated consensus fits well with the concept of subsequent practice for the purposes of the VCLT.¹³⁹

A consolidated regional, international or internationalized consensus may indeed be identified as arising as a result of a series of resolutions of international organizations, and this is reflected in the practice of the IACtHR. In Claude Reyes v Chile, the IACtHR read into Article 13 of the ACHR (freedom of thought and expression) the ‘right of all individuals to request access to State-held information, with the exceptions permitted by the restrictions established in the Convention’.¹⁴⁰ When doing so, the IACtHR said that it was ‘important to emphasize that there is a regional consensus among the States that are members of the Organization of American States … … about the importance of access to public information and the need to protect it’.¹⁴¹ In support of this, it drew attention to a series of resolutions issued by the General Assembly of the OAS on this specific right.¹⁴²

Taking into consideration the importance of international organizations in the international community and in the development of international law, it would be unreasonable to disregard resolutions as indicating that a consensus had been reached on a certain topic. In conclusion, resolutions of international organizations may be perceived as an indication of a consensus.

Yet, even if a resolution evidences a consolidated consensus, could it amount to subsequent practice? As explained in Sections VI.A.1 and 2 above, Article 31(3)(b) refers to ‘any subsequent practice in the application of the treaty which establishes the agreement of the parties’.¹⁴³ A non-traditional interpretation of this text could lead to the conclusion that subsequent practice covers more than pure State practice. The IACtHR is a regional human rights court that is integrated within the OAS, which itself adopted the ACHR. Given this context, the practice of the OAS cannot be irrelevant for interpreting the ACHR. If there is tacit agreement or acquiescence by the States parties to the ACHR, it is possible to consider the practice of the OAS as subsequent practice for the purposes of Article 31(3)(b). However, this is far from being a unanimously held view.

Alternatively, even if one considers that Article 31(3)(b) refers exclusively to State practice, a series of resolutions that indicate the existence of a regional (or even an internationalized) consensus will still be of relevance when interpreting the ACHR in light of its object and purpose. While a single resolution may not, by itself, demonstrate the existence of an international consensus, it can be used alongside other relevant instruments in order to do so. Courts may take into consideration the fact that a significant number of resolutions and/or rules of international law indicate that human rights law is evolving in the same direction and that these instruments, taken together, have been widely adopted by the international community or by the States subject to the Court's jurisdiction. If so, taking these instruments into consideration contributes to the effectiveness of the ACHR and is consonant with its object and purpose, in the sense of Article 31(1) of the VCLT.

In any case, it is generally well accepted that an evolutive interpretation can be supported by the existence of a consensus. As long as this remains the case, the nature of the justification for relying on documents that indicate the existence of such a consensus might be of less significance.

B. The Work of Experts as Subsidiary Evidence of the Evolution of International Law

Legal doctrine is of relevance for determining the rules of international law under Article 38 of the Statute of the International Court of Justice (ICJ).¹⁴⁴ Highly qualified publicists applying Article 31 of the VCLT engage in in-depth analysis that helps clarify the content and scope of international treaties and results in interpretations which can be considered by the IACtHR when it is itself determining their meaning.

The concept of legal doctrine covers both the work of publicists in their individual capacity, as well as the work of groups of experts. ILC members, for example, are eminent publicists and the work of the ILC represents their collective work, which, inter alia, contributes to the progressive development of international law. To consider the work of one expert, in isolation, as legal doctrine, but not the work of a group of experts elected to clarify and develop international law, appears utterly illogical. The same considerations apply to the work of special rapporteurs, who are generally experts in their field, and to guiding principles which following expert consultation have been approved and adopted by international organizations.¹⁴⁵

In conclusion, the works of the ILC and special rapporteurs and the adoption of guiding principles can all be considered as forms of legal doctrine for these purposes. They are relevant to the process of interpretation because they help clarify the provisions which fall to be interpreted. Quoting Roberts, ‘[t]hough not formal sources of law, [writings by publicists] may provide authoritative evidence of the state of the law’.¹⁴⁶ For these reasons, the IACtHR's reliance on such works is fully justified.

Hennebel argues that general comments adopted by the Human Rights Committee could also be qualified as ‘non-binding doctrine serving as guidelines for the Committee and the States parties or as a body of jurisprudence of interpretations with “authoritative and universal character”’.¹⁴⁷ However, this is controversial. Unlike the ILC, the UN treaty bodies are not expressly provided with the competence to develop international law and engage in its codification. Furthermore, their composition is diverse and may include members without legal backgrounds or who are not ‘publicists’. For example, with regard to the UN Committee on the Elimination of Racial Discrimination, Thornberry notes that ‘[m]embers are required to be “experts” but not necessarily “experts on racial discrimination” and the membership continues to include persons with official, particularly foreign policy, connections’.¹⁴⁸ More specifically, he explains that ‘[s]uccessive Committees have had a mixed membership of diplomats, academics, graduates of NGO [non-governmental organization] or activist sectors and national human rights institutions, etc’.¹⁴⁹ Similarly, Byrnes notes that, while nearly all members of the Committee on the Elimination of Discrimination against Women have been actively engaged with the promotion of gender equality, their backgrounds have been diverse, including ‘sociology, medicine, dentistry, international relations, education, political science, psychology, communications, law, and government and foreign service’.¹⁵⁰

That said, the work of the treaty bodies retains a normative value, and it is legitimate for their work to be considered by the IACtHR as it provides guidance concerning the interpretation of other rules which are also applicable to parties to the ACHR, a form of ‘second-degree’ interpretative guidance. The legal value of the work of the human rights treaty bodies has also been recognized by the ICJ, which has said that: ‘Although the Court is in no way obliged, in the exercise of its judicial functions, to model its own interpretation of the Covenant on that of the Committee, it believes that it should ascribe great weight to the interpretation adopted by this independent body that was established specifically to supervise the application of that treaty.’¹⁵¹

VII. CONCLUDING REMARKS

This article shows that the Court's discursive strategy of referring to the VCLT and Article 29 of the ACHR is sufficient to justify its reliance on some external sources when engaging in the evolutive interpretation of the ACHR,¹⁵² but not all. The main conclusions are as follows: first, it fails to provide a comprehensive normative justification for the Court's broad use of external sources and is, therefore, likely to remain susceptible to the charge of ‘cherry-picking’. Different justifications apply to the use of different types of external sources.

Secondly, soft-law instruments undeniably have legal relevance for the evolution of international human rights law and, naturally, for the Court's activities. Yet the fact that they are relevant to treaty interpretation does not mean that they provide sufficient evidence of the extent to which a given provision has evolved over time. Article 31 of the VCLT and Article 29 of the ACHR do not provide a comprehensive justification for the use of each and every soft-law instrument that engages with the corpus juris of international human rights law. Rather, the relevance of each soft-law instrument to the interpretation of the ACHR must be considered separately and conducted in light of Article 31 of the VCLT and Article 29 of the ACHR. This article argues that the Court could benefit from a more direct dialogue with the concept of internationalized consensus in the use of non-binding sources.

Although the Court's approach is sufficient to justify its reliance on certain types of external sources, this article suggests this is not true of all the external sources that it relies on, and this poses a threat to its normative legitimacy. This article does not argue that the Court should refrain from relying on external sources, the use of which cannot be justified under the VCLT, such as soft-law instruments which indicate an emerging, but not yet established, consensus. What it does argue is that the Court's current discursive strategy is insufficient to justify reliance on certain types of external sources to which it does have recourse, and that it does not seem to have an answer to this.

Were the Court to clarify its reasons for relying on the specific external sources it has used and the extent to which these sources have influenced the dynamic interpretation of the ACHR, it would provide more guidance to States parties in predicting the evolution of their obligations. In addition, it would help counter any accusations of ‘cherry-picking’, enhance consistency and predictability, and thus strengthen the juridical scrutiny of the human rights record of the States in the region.