Last week’s decision of the UK Supreme Court in the R v TRA (Appellant) case provides an important confirmation that armed group members can be prosecuted under s134 of the Criminal Justice Act 1988. The decision should be welcomed for providing authoritative guidance on how Article 1 of the UN Convention against Torture should be interpreted, when applied to prosecutions at national level. Specifically, the judgment addresses the interpretation of the phrase ‘public official or other person acting in an official capacity’, finding that the words ‘other person acting in an official capacity’ can be interpreted to include members of armed groups which exercise governmental control over civilian population in a territory over which they control. It distinguishes these kinds of groups from armed groups whose activities are ‘purely military’. This judgment is to be welcomed as it confirms that members of non-State armed groups can be prosecuted for acts amounting to torture. It is also to be welcomed because the interpretation of Article 1 has long been discussed in academic writings (see Gaeta, Clapham & Gaeta, Fortin, Rodenhauser), and partly pertains to the larger question of when and whether armed non State actors are bound by human rights obligations.
It seems that the majority was mainly persuaded by (i) the ordinary meaning of Article 1 and (ii) the purpose of the Convention to establish a regime for international regulation of ‘official torture’, as opposed to private acts of individuals. According to the court, torture perpetrated on behalf of a de facto governmental authority is clearly a matter of proper concern to the international community and within the rationale of the Convention’s regime. The arguments that lead the Court to this conclusion are too detailed and varied to review in their entirety, but I want to address the following three aspects of the judgment: (i) the Supreme Court’s handling of the ordinary meaning of Article 1 (ii) its interpretation of the practice of the Committee Against Torture and (iii) the consequences of the judgment on the relationship between IHL and IHRL.
Ordinary Meaning of Article 1
In several different places in the judgment, the Court made clear that it was not convinced by the appellant’s argument that only persons acting for or on behalf of a State can perpetrate torture. Reviewing the ‘ordinary meaning’ of Article 1, the Court found that by including the two phrases (i) public official and (ii) other person acting in an official capacity, the drafters intended to exclude private conduct, rather than completely exclude non-State activity. In the court’s view, the words ‘person acting in an official capacity’ could on their ‘ordinary meaning’ include someone who holds an official position or acts in an official capacity in an entity exercising governmental control over a civilian population in a territory over which it holds de facto control. In explaining this point, the Supreme Court noted that while in stable States, official conduct is usually performed by State actors, in non-stable situations (e.g. in times of armed conflict) a range of entities may perform governmental or administrative functions. On this point, I wholly agree with the court, as research has long shown that in situations of armed conflict, armed non State actors very often perform functions of government to the exclusion of the de jure government (see here for a review of political science and anthropology literature on this phenomenon).
Practice of Committee Against Torture on the Public Official Requirement
Probably the most interesting part of the judgment (although not necessarily the most determinative) is where the Supreme Court takes a view on the manner in which the Committee Against Torture itself has dealt with the interpretation of Article 1, in its General Comments and in its decisions, pursuant to its individual complaints mechanism. In particular, the Supreme Court engages with the line of contradictory case law that has emerged out of cases where complainants have framed their complaint under Article 3 of the Convention which contains an obligation not to expel, return, or extradite a person to another State where there are substantial grounds for believing that she would be in danger of being subjected to torture. These decisions have long intrigued scholars, who have in particular been fascinated by the Elmi v Australia 1999 case, trying to explain why it has (until recently) been such an anomaly in the Committee’s line of case law on this issue (see McCorquodale and La Forgia, Sivakumaran, Fortin, Rodenhauser).
In the Elmi v Australiacase the Committee held that the return of the applicant – a Somali national – would constitute a violation under Article 3, because his forced return to Somalia would put him at risk of torture at the hands of a non-State actor, the Hawiye clan. In finding that the members of the Hawiye clan fell within the phrase ‘public officials or other persons acting in an official capacity’, the CAT explained its decision on the basis that (i) Somalia had been without central government for a number of years (ii) the non-State actor in question had set up quasi-governmental institutions and (iii) those factions exercise certain prerogatives that are comparable to those usually exercised by legitimate governments.
Although this finding was initially taken as a sign that Article 1 could extend to pain and suffering inflicted by officials of armed groups, the decision was not followed in some of the Committee’s later case law (S.V. v Canada, 2001, M.P.S. v Australia, 2002 and H.M.H.I. v Australia, 2002). A comparison of these cases with the Elmi v Australia case suggests that in these years the Committee was only prepared to recognize that the complainant was at risk of torture under Article 1, if there was no central government. It was not enough for the armed group in question to hold considerable control over a part of the country and to have set up quasi-governmental institutions. The Supreme Court found the basis on which the Committee distinguished its finding in these later cases from its earlier position in Elmi v Australia unconvincing, as a matter of law. It said that if armed groups exercising de facto authority are capable of falling within the definition of torture in article 1 at all, then that should be the case regardless of whether there exists a central government (para 52). It is noteworthy that the Supreme Court’s finding on this point – although not explained very thoroughly in law – is in accordance with the findings of recent longer academic studies which have found authority for the notion that armed groups are bound by human rights law, especially when they control territory and exercise functions of government (Fortin, Murray).
The Supreme Court further justified its rejection of the H.M.H.I. v Australia line of case law by pointing to the more recent decision of SS v The Netherlands, 2003 in which the complainant argued that he would be in danger of being tortured by the LTTE if he returned to Sri Lanka. In this decision, the Committee indicated that a State has an obligation not to expel a person where they risk pain or suffering inflicted by a non-governmental entity which ‘occupies and exercises quasi-governmental authority over the territory to which the complainant returned’. The Court’s attention was also drawn to a General Comment from 2018 (see here) where the committee seemingly also recognizes that pain and suffering inflicted by armed non -state actors may also constitute torture under Article 1. Recognising ‘manifest inconsistencies’ within this line of authority, the Supreme Court found it to provide some support for the view that the conduct of non-State actors exercising de facto authority over territory which they occupy can fall within article 1 of UNCAT (para 52).
Relationship between Torture in International Humanitarian Law and in International Human Rights Law
On the basis of the Court’s findings on Article 1 of the Convention, it is interesting to re-consider the difference between the definition of torture in international humanitarian law (IHL) and the definition of torture under international human rights law (IHRL). At the beginning of the judgment, the Supreme Court confirms that the ‘vital distinction’ between torture under the Convention and torture under IHL is that the former is limited to cases where ‘pain and suffering is inflicted by or at the instigation of or with the consent of a public official or other person acting in an official capacity’ – whereas there is no such ‘public official’ requirement in IHL (para 19).
The R v TRA judgment by no means dispenses with the Convention requirement that the perpetrator of torture needs to be a person acting in an official capacity. It simply confirms that more types of actors can fall into the ‘public official’ category, than only State officials. In that sense, it confirms that in instances where an armed group is fighting a State – and it exercises governmental control over civilian population in a territory over which it holds de facto control – officials from both fighting parties are able to be prosecuted for torture as a war crime and under the Convention.
Notwithstanding the Court’s finding, the kinds of individuals that can commit torture under IHL remain broader than under IHRL, largely because IHL does not only bind persons who are public officials or who act in an official capacity. The Akayesu Appeals Chamber examined this question back in 2001, finding that while in actuality the people carrying out violations of international humanitarian law are in many cases public officials or persons acting in an official capacity, the fact that they have this role or status is not a condition precedent for their prosecution under IHL (para 427-444).
Another difference between the application of IHL and IHRL arises from the fact that the Supreme Court explicitly states that Article 1 will only apply to officials of armed non State actors in circumstances where the armed non State actor in question exercises governmental functions. As armed groups do not need to exercise governmental functions for the threshold of common Article 3 of the Geneva Convention to be met, this is not a condition for the prosecution of torture as a war crime.
While this judgment may be greeted as a triumph for human rights advocates, a perusal of twitter shows that it will not be universally positively received. Certainly, I can see that there may be some virtue in questioning the timing of the development of the law vis-a-vis the commission of the alleged crime (see dissenting judgment of Lord Reed) – as it does seem as if the Committee Against Torture has ‘changed tack’ on this issue in recent years or certainly been inconsistent. But I do not think that there is cause to worry that this judgment serves to legitimate rebel groups in the international arena. The idea that armed groups and their members have obligations under international law has long been recognised under international law. This fact does not take them any closer to Statehood or mean that they automatically become the beneficiary of rights under international law. It simply provides a means to hold them to account for their behaviour.