About the author(s):
Hannes Jöbstl is an Associate Legal Officer at the International Court of Justice (ICJ) in The Hague and a DPhil candidate in the field of public international law at the University of Oxford. His research focuses on penal proceedings conducted by non-State armed groups and is supported by the Arts and Humanities Research Council (AHRC).
Article 3 common to the 1949 Geneva Conventions (CA 3) prohibits “[T]he passing of sentences and the carrying out of executions without previous judgment pronounced by a regularly constituted court, affording all the judicial guarantees which are recognized as indispensable by civilized peoples”. Unfortunately, the 1949 Conventions, including their travaux preparatoires, are silent on what is meant by that expression. While some commentators have taken the view that “regular constitution” simply incorporates the condition that a court must meet the requirements of “independence and impartiality”, René Provost concludes in his book that the idea of “regular constitution” in fact incorporates the human rights law concept that a court must be “established by law” (pp. 195-205, 215-216). In this blog post, I explain why this conclusion is correct and why this interpretation will nevertheless remain controversial.
- Regular Constitution as “Established by Law”
In his discussion of the term “regularly constituted”, Provost places a lot of emphasis on the infamous Hamdan case before the US Supreme Court, in which the majority linked the idea of a regularly constituted court to the concept of a court established by congressional statutes. While the Hamdan case is an important (domestic) precedent, it is important to note that there is also practice by international courts and quasi-judicial bodies that confirms this interpretation. In the Simi? case, for example, an ICTY Trial Chamber elaborated on the fair trial rights enshrined in CA 3 and held that they include “the right to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law” (para. 678). Similarly, the Inter-American Commission for Human Rights (IACommHR), concluded that “international humanitarian law mandates compliance with minimum due process and fair trial protections in any criminal proceeding, including most fundamentally the right to trial by an impartial, independent and regularly constituted court” (para. 258). Both the IACommHR and the ICTY Trial Chamber thus understood regular constitution as a requirement that is firmly enshrined in IHL, and that is distinct from (and not equal to) independence and impartiality. Indeed, as Provost demonstrates (see p. 205), while mutually reinforcing, these three concepts remain separate and cannot simply be equated. A predefined composition and procedure do not only limit the potential of undue external influence over the proceedings (and thereby safeguard the court’s independence), they are also crucial for the exercise of rights and means of defence. The “established by law” requirement thus protects aspects of a fair trial that are not adequately captured by the concepts of independence and impartiality alone.
In addition, even if we assume that the term of “regularly constituted” under IHL does not correspond with the “established by law” requirement under IHRL, any such court would nevertheless have to provide “all the judicial guarantees recognised as indispensable”. If we accept that this phrase is meant to incorporate customary human rights law on due process (as has been argued here and here), it would be difficult to conclude that the “established by law” requirement has not found its way into CA 3 via such implicit renvoi. The ECtHR, for example, has described the requirement of a court to be “established by law” as a “fundamental expression of the rule of law” (para. 64). The ILC, in its commentary on the 1996 Draft Code of Crimes Against Peace and Security of Mankind, noted that “the essential requirement for purposes of the judicial guarantees required for a fair trial is that the court be ‘duly established by law’”. Moreover, the ICTY Appeals Chamber described the ‘established by law’ criterion as a general principle of law applicable to the “administration of criminal justice in a municipal setting” (para. 42). As early as 1989, Theodor Meron thus identified the right to a tribunal ‘established by law’ as part of the customary law of due process and understood this criterion to be reflected in CA 3 (p. 96).
- A Problematic Approach to the Relationship Between AP II and CA 3
In light of all these precedents, why is there still so much controversy as to what precisely is the standard expected from a court operated by a Party to a non-international armed conflict (NIAC), including armed groups? The answer is threefold.
The first reason is a misguided approach to the relationship between CA 3 and the 1977 Additional Protocol II (AP II). The formulation adopted in Article 6(2) AP II differs in significant ways from CA 3 by referring to “a court offering the essential guarantees of independence and impartiality”, omitting any mention of the regular constitution of the tribunal. It is therefore sometimes invoked, for example by the Stockholm District Court in the Sakhnah case, to demonstrate that international law no longer requires a court to be “regularly constituted” but merely to offer essential due process guarantees such as independence and impartiality. However, as noted by Provost (pp. 203-204) and others, this interpretation of CA 3 via Article 6 AP II is highly problematic. First, Article 6 AP II does in fact nothing to “develop or supplement” CA 3 as specified in the Protocol’s own preamble, but rather loosens its provisions. Moreover, CA 3 “retains an autonomous existence”, meaning that its applicability is neither limited nor affected by the material field of application of the Protocol. Relying on AP II to narrow the level of protection provided by CA 3 thus seems to be an argument that is difficult to sustain.
Nevertheless, it could be argued that the parties to AP II wanted to clarify the definition of “regularly constituted” irrespective of the type of NIAC. As such, the Protocol could be seen as a subsequent agreement under Article 31(3)(a) of the VCLT and its customary law equivalents. However, one must bear in mind that whereas the 1949 Conventions enjoy universal adherence, a number of States have not yet signed or ratified AP II. An authoritative interpretation established by subsequent agreement normally requires the agreement of all State Parties to the original treaty. Accordingly, the Protocol cannot constitute such agreement. Alternatively, the Protocol could be considered as subsequent practice under Article 31(3)(b) of the VCLT. Subsequent practice does not necessarily require the practice of all States to the original treaty, as long as the inactive States accept that practice. Consequently, two questions arise: First, whether Article 6 AP II established an “agreement” in relation to the interpretation of CA 3. Second, if that is the case, whether those Parties to the 1949 Conventions that did not accede to AP II have accepted this practice.
With regard to the intention of the parties to adopt an agreement on the meaning of CA 3, the travaux (in particular, CDDH/1/SR.33-34) are inconclusive at best, with some statements indicating that Article 6 was reinforcing a rule already enshrined in CA 3 and others that it was meant to introduce a new, independent standard. However, even if such agreement were to be assumed, the practice of the States Parties to AP II alone would not be sufficient to amount to an authoritative interpretation or modification of CA 3. There must be at least “good evidence” that the other Parties to the 1949 Geneva Conventions, which have not acceded to AP II, accept such an interpretation. However, not only is it difficult to find such evidence, there are also examples of State practice clearly contradicting such an interpretation, for example the Hamdan or Simic decision referred to above. In the absence of further conclusive evidence, the default position must therefore be that CA 3 applies unmodified, as specified in AP II’s Preamble. Accordingly, Provost is completely right to conclude that the requirement of regular constitution, in the sense that courts must be ‘established by law’, applies in all NIACs, even when they are governed by AP II (p. 215-16).
- The Rome Statute’s Elements of Crime
The second source of confusion in relation to the interpretation of “regular constitution” are the Elements of Crimes of the Rome Statute. These Elements, which are meant to assist the Court in the interpretation and application of the Statute, specify that the phrase “regularly constituted” is to be interpreted as affording the “essential guarantees of independence and impartiality”. This phrase is thus often invoked (e.g. here, para. 27 and here, p. 413) to support the notion that a “regularly constituted” court is in reality nothing else than an “independent and impartial” court. Provost dismisses this interpretation by focussing on the discrepancy between the non-binding Element of Crime and the authoritative wording of Article 8(2)(c)(iv) of the Statute (p. 205). He also notes that, regrettably in his view, the ICC Pre-Trial Chamber in Al-Hassan endorsed the interpretation included in the Elements of Crimes. Yet, the Chamber also acknowledged that the drafters of the Elements of Crime deliberately “departed” from CA 3 by not including regular constitution as a separate requirement in addition to the requirements of independence and impartiality (para. 377).
I wonder whether this finding does not, ultimately, lead to a very reasonable compromise between accepting that CA 3 requires courts to be established by law and the challenge that this requirement poses to armed groups and individuals involved their courts. The most likely suspects of Article 8(2)(c)(iv), such as judges, prosecutors or police officers, will usually have some influence on whether due process guarantees are being complied with and whether the accused is being afforded the necessary rights and means in defence. However, it is usually not in their power to influence the composition or jurisdictional powers (that is, the “legal basis”) of the court in question. Although war crimes are serious violations of IHL that carry individual criminal responsibility, this does not mean that the scope of these crimes is exactly the same as that of the underlying rule of IHL. Indeed, unlike CA 3 or AP II, the Statute does not impose obligations onto armed groups as a collective. It could therefore be argued that the bar for individual criminal responsibility under Article 8(2)(c)(iv) is simply higher than for an ordinary violation of CA 3. Accordingly, the Elements of Crimes should not be understood as an authoritative interpretation of “regular constitution” under CA 3.
4. Armed Group Courts Established by Law – The Fear of a Normative Overreach?
This leads to the third reason, namely a certain hesitancy in applying the “established by law” requirement to armed groups. If the term “law” is interpreted in the strict sense of a parliamentary statute, many commentators have expressed their doubts over whether an armed group could ever comply with this requirement. As Provost notes in his book, armed groups do not normally have a constitutional structure that replicates that of states, including separate legislative bodies (p. 201). Yet, many critics seem to forget that dropping the “established by law” criterion would equally water down the guarantees that are required from States during NIAC, or, if dropped only in relation to armed groups, erode the principle of belligerent equality. Moreover, it should be born in mind that high compliance is the means to an end, not an end in itself. Indeed, as pointed out by Yuval Shany, “the overriding raison d’être of IHL is not full compliance with its norms but rather the protection of humanitarian values”.
The better view might thus be to accept that armed group courts must be established by law, but to adopt a certain flexibility in relation to what exactly constitutes ‘law’ for that purpose (see also Sivakumaran, p. 306). In his book, Provost takes a similar view by suggesting that the concept of a “regularly constituted court” should be understood as the ‘need for deliberate consideration by an established authority to constitute a court of general jurisdiction that will apply “law” broadly defined’ (p. 202). Indeed, as pointed out by Ezequiel Heffes, this understanding would place the internal judicial regulation of all the parties to a NIAC in an equal position and, in turn, arguably enhance the respect for the rule of law. In conclusion, there is no reason not to expect courts to be “established by law” during armed conflict, irrespective of whether they belong to States or armed groups.