[2019] UKSC 51
On appeal from: [2018] EWCA Crim 2843
JUDGMENT
R v Reeves Taylor (Appellant)
|
before
Lady Hale, President Lord Reed, Deputy President Lord Wilson Lord Hodge Lord Lloyd-Jones
|
JUDGMENT GIVEN ON |
|
|
13 November 2019 |
|
|
Heard on 24 and 25 June 2019 |
Appellant |
|
Respondent |
Steven Powles QC |
|
David Perry QC |
Tatyana Eatwell |
|
Paul Rogers |
Margherita Cornaglia |
|
Kathryn Howarth |
|
|
Emilie Pottle |
(Instructed by Bark & Co) |
|
(Instructed by CPS Counter Terrorism Division (Westminster)) |
|
|
Intervener (Redress) |
|
|
Sudhanshu Swaroop QC |
|
|
John Bethell |
|
|
(Instructed by Hogan Lovells International LLP) |
LORD LLOYD-JONES: (with whom Lady Hale, Lord Wilson and Lord Hodge agree)
1. The appellant, Reeves Taylor, who was arrested in the United Kingdom on 1 June 2017, is charged with one count of conspiracy to commit torture (count 1) and seven counts of torture (counts 2-8). The substantive offence alleged in each case is that of torture contrary to section 134, Criminal Justice Act 1988 (“CJA”). The charges relate to events in Liberia in 1990, in the early stages of the first Liberian civil war, when an armed group, the National Patriotic Front of Liberia (“NPFL”), sought to take control of the country and to depose the then President, Samuel Doe. The leader of the NPFL was Charles Taylor REDACTED. The NPFL eventually succeeded in taking control of Liberia and Charles Taylor became President in 1997.
“What is the correct interpretation of the term ‘person acting in an official capacity’ in section 134(1) of the Criminal Justice Act 1988; in particular does it include someone who acts otherwise than in a private and individual capacity for or on behalf of an organisation or body which exercises or purports to exercise the functions of government over the civilian population in the territory which it controls and in which the relevant conduct occurs?”
The prosecution case
7. In a memorandum served by the prosecution after the hearing before the Court of Appeal, the prosecution’s expert clarifies that his use of the term “control” refers to military rather than administrative control over the area. He states that the NPFL offensives in early 1990 caused the Armed Forces of Liberia (“AFL”) to withdraw from nearly all areas of Nimba County and consolidate their forces in military bases located in strategic towns. This withdrawal created a situation in which NPFL forces had freedom of movement throughout the County. As a result, the NPFL was the de facto military authority in the area. Such military control is said to be very different from administrative control. He states that before June 1990 the NPFL did not have a sustained presence in much of Nimba County. It did not assign officials to oversee towns or deploy forces to provide security. NPFL forces passed through towns and villages on an ad hoc basis; there was no sustained or coordinated occupation. Much of the population lived in a ‘no man’s land’, areas without any consistent administrative authority, but with the occasional presence of NPFL fighters.
The proceedings
The UN Convention against Torture and its implementation
14. Section 134 CJA provides in relevant part:
“134. Torture
(1) A public official or person acting in an official capacity, whatever his nationality, commits the offence of torture if in the United Kingdom or elsewhere he intentionally inflicts severe pain or suffering on another in the performance or purported performance of his official duties.
(2) A person not falling within subsection (1) above commits the offence of torture, whatever his nationality, if -
(a) in the United Kingdom or elsewhere he intentionally inflicts severe pain or suffering on another at the instigation or with the consent or acquiescence -
(i) of a public official; or
(ii) of a person acting in an official capacity; and
(b) the official or other person is performing or purporting to perform his official duties when he instigates the commission of the offence or consents to or acquiesces in it.
…
(6) A person who commits the offence of torture shall be liable on conviction on indictment to imprisonment for life.”
2. This article is without prejudice to any international instrument or national legislation which does or may contain provisions of wider application.”
Under UNCAT each State Party is required to take effective legislative, administrative, judicial or other measures to prevent acts of torture in any territory under its jurisdiction (article 2). No State Party shall expel, return (“refouler”) or extradite a person to another State where there are substantial grounds for believing that he would be in danger of being subjected to torture (article 3). Each State Party is required to ensure that all acts of torture are offences under its criminal law (article 4) and to establish its jurisdiction over such offences in cases where the alleged offender is present in any territory under its jurisdiction and it does not extradite him (article 5). In such cases each State Party is obliged, if it does not extradite the alleged offender, to submit the case to its competent authorities for the purpose of prosecution (article 7). Each State Party also undertakes to prevent in any territory under its jurisdiction other acts of cruel, inhuman or degrading treatment or punishment which do not amount to torture as defined in article 1, when such acts are committed by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity (article 16).
Torture in international humanitarian law
(1) Article 3, common to each of the four Geneva Conventions of 1949, prohibits torture in non-international armed conflicts and establishes protections for persons who do not or who no longer take an active part in hostilities. In Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States of America) (Merits, Judgment) ICJ Rep 1986, 14 at 113-4, para 218, the International Court of Justice held that Common Article 3 establishes minimum guarantees that apply in all armed conflict.
(2) The statutes of the ad hoc international tribunals for the former Yugoslavia (“ICTY”) and Rwanda (“ICTR”) conferred jurisdiction to try offences of torture committed during armed conflict without defining the offence. The Tribunals produced their own definitions, based heavily on UNCAT.
(3) Under the Rome Statute of the International Criminal Court (“ICC”), 17 July 1988, torture is capable of constituting (1) a crime against humanity when committed as part of a widespread or systematic attack directed against any civilian population (article 7(1)(f)); (2) a war crime when committed in an armed conflict whether international or not of an international character (articles 8(2)(a)(ii) and 8(2)(c)(i)).
20. Torture as a crime against humanity and torture as a war crime, as defined in the Statute of the ICC, are both offences contrary to UK law by virtue of sections 50 and 51, International Criminal Court Act 2001. As a result, torture committed in certain circumstances may be prosecuted here as a war crime or a crime against humanity. However, the alleged conduct which gives rise to the current charges against the appellant could not be prosecuted in the United Kingdom on either of these bases, even if the elements of these offences were otherwise established, because section 65A of the International Criminal Court Act 2001, inserted by section 70 of the Coroners and Justice Act 2009, which deals with retrospective application, provides that the relevant sections apply to acts committed on or after 1 January 1991, which is later than the date on which the instant offences are alleged to have been committed.
The submissions of the parties
The approach to interpretation of section 134 CJA and article 1, UNCAT
23. Section 134 CJA was intended to give effect to UNCAT in domestic law. As a result, the words “person acting in an official capacity” must bear the same meaning in section 134 as in article 1, UNCAT. (See R v Bow Street Metropolitan Stipendiary Magistrate, Ex p Pinochet Ugarte (No 3) (“Pinochet No 3”) [2000] 1 AC 147 per Lord Browne-Wilkinson at p 200A-B.) The principles of international law governing the interpretation of treaties are to be found in articles 31 and 32, Vienna Convention on the Law of Treaties, 23 May 1969, (1980) (Cm 7964), 1155 UNTS 331.
“Article 31. General Rule of Interpretation
1. A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.
2. The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:
(a) Any agreement relating to the treaty which was made between all the parties in connexion with the conclusion of the treaty;
(b) Any instrument which was made by one or more parties in connexion with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.
3. There shall be taken into account, together with the context:
(a) Any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions;
(b) Any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;
(c) Any relevant rules of international law applicable in the relations between the parties.
4. A special meaning shall be given to a term if it is established that the parties so intended.
Article 32. Supplementary Means of Interpretation
Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:
(a) Leaves the meaning ambiguous or obscure; or
(b) Leads to a result which is manifestly absurd or unreasonable.”
Ordinary meaning
Object and purpose
26. It is well established that a treaty should be interpreted in accordance with the ordinary meaning to be given to its terms in their context and in the light of its object and purpose (Vienna Convention on the Law of Treaties, article 31(1)). It is necessary, however, to sound a cautionary note at this point. While the object of UNCAT was undoubtedly, in one sense, to end impunity for perpetrators of what might be termed “official torture” it does not follow that the reading of the Convention which would best avoid impunity must be adopted in all circumstances. It is, rather, necessary to give effect to the words used in the light of the object and purpose of the scheme created by the State parties to the Convention. Similarly, the mere fact that a particular reading may be seen as a desirable development of the law is not of itself a valid reason for adopting it. It is not for national courts engaged in interpreting a treaty to seek to force the pace of the development of international law, however tempting that may be. There is an analogy to be drawn here with the consideration by national courts of potential rules of customary international law, addressed by the House of Lords in Jones v Ministry of the Interior of the Kingdom of Saudi Arabia (Secretary of State for Constitutional Affairs intervening) (“Jones v Saudi Arabia”) [2006] UKHL 26; [2007] 1 AC 270 where Lord Hoffmann observed (at para 63):
“It is not for a national court to ‘develop’ international law by unilaterally adopting a version of that law which, however desirable, forward-looking and reflective of values it may be, is simply not accepted by other states.”
Lord Bingham noted, similarly, (at para 22) that one swallow does not make a rule of international law.
“The Torture Convention was agreed not in order to create an international crime which had not previously existed but to provide an international system under which the international criminal - the torturer - could find no safe haven.” (See also Lord Hutton at pp 260F-261B)
As Judge Crawford points out, UNCAT, in common with similar treaties relating, for example, to the unlawful seizure of aircraft or the taking of hostages, does not impose criminal responsibility directly upon individuals, but rather requires contracting States to prevent and punish the conduct in question.
“The enforcement of such norms occurs at the domestic rather than the international level, as the treaties envisage punishment only by domestic courts. In addition to obliging states parties to criminalize certain conduct, such treaties generally require them to prosecute or extradite accused persons to other states parties that are willing to prosecute them (aut dedere aut iudicare). … While the enforcement of these norms is dependent on domestic legal systems either prosecuting or extraditing accused persons, various treaty bodies - such as the Committee against Torture - often play an important role in monitoring the implementation of the treaty norms at the domestic level. (James Crawford, Brownlie’s Principles of Public International Law, 9th ed (2019), p 663.)”
It can be seen therefore that the object of UNCAT has been to make torture as defined in article 1 a criminal offence of universal jurisdiction enforceable by domestic courts and, by virtue of very extensive State participation in the Convention, to establish a machinery capable of reducing the likelihood of perpetrators of “official torture” escaping justice before national courts.
Travaux preparatoires
“… in particular, it should be made clear that the term ‘public official’ contained in paragraph 1 refers not only to persons who, regardless of their legal status, have been assigned public authority by State organs on a permanent basis or in an individual case, but also to persons who, in certain regions or under particular conditions, actually hold and exercise authority over others and whose authority is comparable to governmental authority or - be it only temporarily - has replaced government authority or whose authority has been derived from the aforementioned persons.” (E/CN.4/1314/Add 2, para 2)
Clearly, the German proposal was not implemented in terms. (See Boulesbaa, The UN Convention on Torture and the Prospects for Enforcement, (Martinus Nijhoff 1999), pp 27-28; Wendland, A Handbook on State Obligations under the UN Convention against Torture, Association for the Prevention of Torture, (2002), p 29.) Nevertheless, it may have influenced the expansion of the concept of “public official”. Nowak and McArthur summarise the matter as follows:
“116. Severe pain or suffering only counts as torture in the understanding of the Convention if it is ‘inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity’. The formulation in the 1975 Declaration and the original Swedish draft (‘by or at the instigation of a public official’) reflects the traditional view that States can only be held accountable for human rights violations committed by State actors. Since the main purpose of the Convention was to require States parties to use domestic criminal law for the purpose of punishing perpetrators of torture, several governments, such as France, Barbados, Panama and Spain, advocated an extension of the definition covering also private individuals. Germany did not go as far but wished to include also non-State actors who exercise authority over others and whose authority is comparable to government authority. Since other governments, including the United States, United Kingdom, Morocco and Austria, insisted on a traditional State-centred definition, the Working Group finally agreed on a US compromise proposal which extended State responsibility to the consent or acquiescence of a public official. Since the delegations could not agree on a definition of the term ‘public official’, the Austrian proposal to add the phrase ‘or other person acting in an official capacity’ was adopted.” (footnotes excluded)
On this basis Mr Powles submits that the compromise was the inclusion of non-State actors who act with the consent or acquiescence of a public official and not the expansion of the definition of “person acting in an official capacity” beyond persons or entities who in fact act for or on behalf of the State.
31. It seems clear from this account by Nowak and McArthur that, in the result, the intention was at least to exclude from the definition the conduct of private individuals acting in a private capacity. The question is whether it was intended to go further and to include within the definition acts of de facto authorities exercising governmental functions. On behalf of the appellant, Mr Powles draws particular attention to the reference by Nowak and McArthur to insistence “on a traditional State-centred definition” and submits that it supports the view that conduct in an official capacity must be attributable to the State. It is not entirely clear, however, what is meant by this term. A later passage in Nowak and McArthur, on which the prosecution relies, suggests that it may be intended simply to exclude perpetrators acting entirely in a personal capacity.
“118. The term ‘other person acting in an official capacity’ goes, however, clearly beyond State officials. It was inserted on the proposal of Austria in order to meet the concerns of the Federal Republic of Germany that certain non-State actors whose authority is comparable to governmental authority should also be held accountable. These de facto authorities seem to be similar to those ‘political organizations’ which, according to article 7(2)(i) ICC Statute, can be held accountable for the crime of enforced disappearance before the ICC. One might think of rebel, guerrilla or insurgent groups who exercise de facto authority in certain regions or of warring factions in so-called ‘failing States’.
119. In the case of Elmi v Australia, the Committee had to decide whether the forced return of a Somali national belonging to the Shikal clan to Somalia, where he was at a substantial risk of being subjected to torture by the ruling Hawiye clan, constituted a violation of the prohibition of refoulement pursuant to article 3. The Committee found a violation of article 3 and explicitly rejected the argument of the Australian Government that the acts of torture the applicant feared he would be subjected to in Somalia would not fall within the definition of torture set out in article 1: …” (original emphasis, footnotes omitted)
Article 7(2)(i), ICC Statute, to which Nowak and McArthur refer, provides:
“‘Enforced disappearance of persons’ means the arrest, detention or abduction of persons by, or with the authorization, support or acquiescence of, a State or a political organization, followed by a refusal to acknowledge that deprivation of freedom or to give information on the fate or whereabouts so those persons, with the intention of removing them from the protection of the law for a prolonged period of time.”
“There were different opinions on the question as to whether or not the definition of torture in the convention should be limited to acts of public officials. It was pointed out by many States that the purpose of the convention was to provide protection against acts committed on behalf of, or at least tolerated by, the public authorities, whereas the State could normally be expected to take action according to its criminal law against private persons having committed acts of torture against other persons. However, France considered that the definition of the act of torture should be a definition of the intrinsic nature of the act of torture itself, irrespective of the status of the perpetrator.
Although there was little support for the French view on this matter, most States agreed that the convention should not only be applicable to acts committed by public officials, but also to acts for which the public authorities could otherwise be considered to have some responsibility.” (Burgers and Danelius (above) at p 45, original emphasis)
In a further passage, Burgers and Danelius state:
“In principle, the common element of the purposes referred to in the definition should rather be understood to be the existence of some - even remote - connection with the interests or policies of the State and its organs. It is important to note, in this context, that the primary objective of the Convention is to eliminate torture committed by or under the responsibility of public officials for purposes connected with their public functions. Precisely because the public interest is sometimes seen in such cases as a justification, the authorities may be reluctant to suppress these practices. The provisions of the Convention are intended to ensure that torture does not occur in such cases or that, if it occurs, action is taken against the offender.” (Burgers and Danelius (above), at pp 118-119)
While these passages may be read as providing some support for the appellant’s case, it is important to bear in mind that neither was addressing the specific question which arises in these proceedings. The first was addressing the distinct questions of whether torture under the Convention should include private acts of torture and whether it should extend beyond acts committed by public officials. The second was addressing the requirement that torture should be committed for a specific purpose connected to the actor’s public function. As a result, these passages cast little light on the meaning of the words “acting in an official capacity”.
“Most speakers were in favour of the principle of universal jurisdiction, holding it to be essential in securing the effectiveness of the Convention. Territorial jurisdiction would not suffice to punish torture effectively as a State policy, under the definition of article 1.” (E/CN.4/1983/63 at para 21)
Subsequent practice
Kooijmans Report, 19 February 1986
“Article 1, para 1, of the Convention reads as follows ‘… when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity’. The Convention was again following the Declaration of 1975, but developing it by adding the phrases ‘or with the consent or acquiescence of’ and ‘or other person acting in an official capacity’. Consequently, State responsibility is apparent even when the authorities resort to the use of private gangs or paramilitary groups in order to inflict ‘severe pain or suffering’ with the intention and purposes already mentioned. However, private acts of brutality - even the possible sadistic tendencies of particular security officials - should not imply State responsibility, since these would usually be ordinary criminal offences under national law.” (at para 38)
In this passage Mr Kooijmans is focussing on the circumstances in which a State may be responsible for acts of torture. He emphasises the distinction between official and private acts and the extension of the State’s responsibility in cases of consent or acquiescence within article 1. Contrary to the appellant’s submission, this passage does not support the proposition that a “state-nexus requirement” is inherent in the term “official capacity”, such that the term only applies to persons acting on behalf of the State. The relevance of state responsibility to the present issue is considered below.
Committee Against Torture
“… [T]he committee is not an exclusively legal and not an adjudicative body; its power under article 19 is to make general comments; the committee did not, in making this recommendation, advance any analysis or interpretation of article 14 of the Convention; and it was no more than a recommendation. Whatever its value in influencing the trend of international thinking, the legal authority of this recommendation is slight.”
General Comments of the CAT
“The Convention imposes obligations on States parties and not on individuals. States bear international responsibility for the acts and omissions of their officials and others, including agents, private contractors and others acting in an official capacity or acting on behalf of the State, in conjunction with the State, under its direction or control, or otherwise under colour of law.” (at para 15)
It is, of course, correct that UNCAT is binding on the Contracting States in international law and that it does not itself impose obligations on individuals. Rather, it imposes on each State party an obligation to create and enforce in its domestic law an offence which conforms with the definition in article 1. However, it is not possible to derive from this the conclusion that torture within article 1 is limited to conduct attributable to the State as suggested by the appellant. Furthermore, this passage is not necessarily intended to be an exclusive description of the scope of article 1.
“… States parties should refrain from deporting individuals to another State where there are substantial grounds for believing that they would be in danger of being subjected to torture or other ill-treatment at the hands of non-State entities, including groups that are unlawfully exercising actions that inflict severe pain or suffering for the purposes prohibited by the Convention, and over which the receiving State has no or only partial de facto control, or whose acts it is unable to prevent or whose impunity it is unable to counter.” (at para 30)
This passage accurately describes the non-refoulment obligation. However, it does not address the question whether the conduct of such non-State entities might constitute torture within article 1 if they are quasi-governmental entities performing governmental functions. More generally, the General Comments do not provide any support for the reading of article 1 for which the appellant contends.
“18. The Committee has made clear that where State authorities or others acting in an official capacity or under colour of law, know or have reasonable grounds to believe that acts of torture or ill-treatment are being committed by non-State officials or private actors and they fail to exercise due diligence to prevent, investigate, prosecute and punish such non-State officials or private actors consistently with the Convention, the State bears responsibility and its officials should be considered as authors, complicit or otherwise responsible under the Convention for consenting to or acquiescing in such impermissible acts. Since the failure of the State to exercise due diligence to intervene to stop, sanction and provide remedies to victims of torture facilitates and enables non-State actors to commit acts impermissible under the Convention with impunity, the State’s indifference or inaction provides a form of encouragement and/or de facto permission. The Committee has applied this principle to States parties’ failure to prevent and protect victims from gender-based violence, such as rape, domestic violence, female genital mutilation, and trafficking.”
In this regard the prosecution further relies on passages in the CAT’s initial report on Iraq, 7 September 2015 (CAT/C/IRQ/CO/1, paras 11-12) and its second periodic report on Afghanistan, 12 July 2017 (CAT/C/AFG/2, para 7). On this basis it submits that it is envisaged that it is not only State agents who should be prosecuted and punished pursuant to the State parties’ obligations under the Convention but also perpetrators from non-State party groups or organisations operating on the territory of the State party.
Decisions under article 22(7), UNCAT
48. A series of decisions of the CAT under article 22(7) UNCAT addresses the obligation on State Parties under article 3 not to expel, return or extradite a person to another State where there are substantial grounds for believing that he would be in danger of being subjected to torture. In these decisions the statements by the CAT in relation to the present issue are inconsistent. In SV v Canada (15 May 2001; Communication No 49/1996; UN Doc CAT/C/26/D/49/1996 (2001)) the authors complained that they were at risk of torture by the Sri Lankan authorities, but also complained that they were at risk of torture by the Liberation Tigers of Tamil Eelam (“LTTE”) a rebel organisation which, the decision recorded, had in 1990 taken control of the Tamil region. The CAT, having referred to the definition in article 1, rejected the latter complaint.
“The Committee considers that the issue of whether the State party has an obligation to refrain from expelling a person who might risk pain or suffering inflicted by a non-governmental entity, without the consent or acquiescence of the Government, falls outside the scope of article 3 of the Convention. Consequently, the issue, on which the authors base part of their claim that they would suffer torture by LTTE or other non-governmental entities on return to Sri Lanka, cannot be considered by the Committee.” (para 9.5)
Similarly, in GRB v Sweden (19 June 1998; CAT/C/20/D/83/1997 at para 6.5) (where the author complained that if returned to Peru she would be at risk of torture both by the State authorities and by Sendero Luminoso) and in MPS v Australia (30 April 2002; CAT/C/28/D/138/1999 at para 7.4) (concerning the risk of torture by the LTTE in Sri Lanka) the CAT repeated this conclusion in almost identical terms.
49. By contrast, in Elmi v Australia (14 May 1999; CAT/C/22D/120/1998) the author, a Somali national of the Shikal clan, claimed that his forced return to Somalia would constitute a violation of article 3 because he was a risk of torture at the hands of the Hawiye clan. The CAT concluded:
“6.5 The Committee does not share the State party’s view that the Convention is not applicable in the present case since, according to the State party, the acts of torture the author fears he would be subjected to in Somalia would not fall within the definition of torture set out in article 1 … The Committee notes that for a number of years Somalia has been without a central government, that the international community negotiates with the warring factions and that some of the factions operating in Mogadishu have set up quasi-governmental institutions and are negotiating the establishment of a common administration. It follows then that, de facto, those factions exercise certain prerogatives that are comparable to those normally exercised by legitimate governments. Accordingly, the members of those factions can fall, for the purposes of the application of the Convention, within the phrase ‘public officials or other persons acting in an official capacity’ contained in article 1.
…
6.7 The Committee further notes, on the basis of the information before it, that the area of Mogadishu where the Shikal mainly reside, and where the author is likely to reside if he ever reaches Mogadishu, is under the effective control of the Hawiye clan, which has established quasi-governmental institutions and provides a number of public services.”
“… with three years elapsing since the Elmi decision, Somalia currently possesses a State authority in the form of the Transitional National Government, which has relations with the international community in its capacity as central government, though some doubts may exist as to the reach of its territorial authority and its permanence. Accordingly, the Committee does not consider this case to fall within the exceptional situation in Elmi and takes the view that acts of such entities as are now in Somalia commonly fall outside the scope of article 3 of the Convention.” (at para 6.4)
“… the issue whether the State party has an obligation to refrain from expelling a person who might risk pain or suffering inflicted by a non-governmental entity, without the consent or acquiescence of the Government, falls outside the scope of article 3 of the Convention, unless the non-governmental entity occupies and exercises quasi-governmental authority over the territory to which the complainant would be returned.” (at para 6.4)
52. The basis on which the CAT sought to distinguish Elmi in HMHI is, with respect, unconvincing. If acts by rebel 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. Furthermore, the suggestion on behalf of the appellant that once there was a central government, notwithstanding doubts as to the reach of its territorial authority and permanence, the risk of torture fell outside the scope of the Convention, cannot be accepted. The decision in Elmi did not turn on the notion that refoulement would violate the Convention because there was no effective government in Somalia to protect individuals from non-State actors. Rather, the decision in Elmi made clear that the Hawiye were a quasi-governmental institution performing functions comparable to those normally performed by legitimate governments and that it was that de facto status which brought its conduct within the scope of article 1. By contrast with HMHI, the ruling on this point in SS makes eminent sense. It reaffirms that acts of a de facto authority are capable of falling within the definition in article 1 and it does so in terms which free it from the unprincipled restriction apparently imposed in HMHI. Despite its manifest inconsistencies, therefore, this line of authority does 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.
Context of international law
Relevance of State responsibility
Relevance of recognition of States and Governments
56. The appellant’s suggested reading of article 1 gives rise to a number of difficulties and anomalies concerning issues of recognition of States and governments. First, before the Court of Appeal the appellant submitted that the term “person acting in an official capacity” is limited to a person acting for or on behalf of a government authority of a recognised State, a submission which was rejected by the Court of Appeal. At the hearing before the Supreme Court that submission was no longer maintained. For present purposes it is not necessary to embark on a consideration of the relevance, if any, of recognition to statehood in international law. It is sufficient to refer to the following observation of Chief Judge Newman in the US Court of Appeals, Second Circuit in Kadic v Karadic 70 F 3d 232 (2d Cir 1995), at 245 with which I respectfully agree.
“The customary international law of human rights, such as the proscription of official torture, applies to states without distinction between recognized and unrecognized states … It would be anomalous indeed if non-recognition by the United States, which typically reflects disfavour with a foreign regime - sometimes due to human rights abuses - had the perverse effect of shielding officials of the unrecognized regime from liability for those violations of international law norms that apply only to state actors.”
UK authorities
“The definition of torture, both in the Convention and section 134, is in my opinion entirely inconsistent with the existence of a plea of immunity ratione materiae. The offence can be committed only by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. The official or governmental nature of the act, which forms the basis of the immunity, is an essential ingredient of the offence. No rational system of criminal justice can allow an immunity which is coextensive with the offence.” (at p 277D-E)
So much is uncontroversial. What is more problematical is what constitutes acting in an official capacity, a matter on which, with one exception, the UK authorities provide little assistance.
61. The appellant relies on a line of authority concerning the relevant definition of torture for the purposes of immigration and asylum detention policy which, it is submitted, reveals a clear understanding that “person acting in an official capacity” in article 1 UNCAT refers to persons acting on behalf of the State. In R (EO) v Secretary of State for the Home Department [2013] EWHC 1236 (Admin), ACD 116 Burnett J concluded that “torture” in rule 35(3), Detention Centre Rules 2001 (SI 2001/238) (“DCR”) and in certain policy documents had a broader meaning than the UNCAT definition. It was not confined to acts of public officials or other persons acting in an official capacity or in which they were complicit or acquiesced. Subsequent to EO, the Secretary of State introduced statutory guidance entitled “Adults at Risk in Immigration Detention” (“AARSG”). The definition of “torture” for the purposes of rule 35 of DCR and the AARSG is set out in Detention Services Order, DSO 09/2016 which sets out the definition in article 1 UNCAT but adds:
“It includes such acts carried out by terrorist groups exploiting instability or civil war to hold territory.”
In R (Medical Justice) v Secretary of State for the Home Department [2017] EWHC 2461 (Admin); [2017] 4 WLR 198, Ouseley J set out the history of these provisions and noted that:
“The reference to acts carried out by terrorist groups is not part of the UNCAT definition, but was added following discussions between the SSHD and an NGO, Freedom from Torture; it was suggested by Sir Keir Starmer MP.” (at para 33)
This, it is said on behalf of the appellant, demonstrates that the courts and the parties to these cases, including the Secretary of State, clearly understood that “person acting in an official capacity” for the purposes of article 1, UNCAT applies only to a person acting for or on behalf of a State. While this might, at first sight, appear to support the appellant’s case, I am unable to attach any great weight to it. The precise question in issue here was not under consideration. Moreover, in the particular context of immigration detention it was clearly desirable to include such an express provision for the benefit of persons who, because of their history, should not be detained.
62. We were informed that this is only the third occasion on which a prosecution has been brought in the United Kingdom pursuant to section 134, CJA. In R v Lama [2014] EWCA Crim 1729; [2017] QB 1171 the present issue did not arise. However, in my view considerable assistance is to be found in the first instance decision of Treacy J in R v Zardad, Case No T2203 7676, 7 April 2004. Zardad was charged with a conspiracy to torture in Afghanistan in circumstances where the substantive charge would have been that contrary to section 134, CJA. The case concerned the period between 1992 and 1996 when the Hezb-I-Islami faction was in control of Laghman Province. During that period Zardad was a chief commander of Hezb-I-Islami and the military controller of the area of Sarobi. Zardad maintained that he was not a public official since there was a recognised government in Afghanistan at the relevant time and the group to which he belonged was not a part of that government and was actively opposed to it. The prosecution maintained that he was either a public official de jure or a person acting in an official capacity de facto.
“It seems to me that what needs to be looked at is the reality of any particular situation. Is there sufficient evidence that Hezb-I-Islami had a sufficient degree of organisation, a sufficient degree of actual control of an area and that it exercised the type of functions which a government or governmental organisation would exercise? It seems to me that I have to take care not to impose Western ideas of an appropriate structure for government, but to be sensitive to the fact that in countries such as Afghanistan different types of structure may exist, but which may legitimately come within the ambit of an authority which wields power sufficient to constitute an official body.” (at para 33)
He rejected a submission, based on Elmi, that the mere fact that there is a central government in existence precludes there being a de facto authority of which a person might be a public official or on whose behalf a person might act in a similar capacity. He considered that the words “person acting in a public capacity” included those acting for an entity which had acquired de facto effective control over an area of a country and was exercising governmental or quasi-governmental functions in that area. In his view, there was material on which a jury could conclude that Zardad was such a de facto public official in an area totally controlled by his organisation which exercised, with a degree of permanence, functions which would be functions of a state authority (at paras 34-38).
64. Mr Powles submits that Zardad is wrongly decided. However, I find the approach of Treacy J compelling and in conformity with the preponderant weight of material relevant to the interpretation of article 1, UNCAT.
65. Zardad is also instructive as to which features are indicative of governmental activity. There, in support of its contention that Zardad should be treated as a public official on a de facto basis, the prosecution maintained that Zardad was akin to a Military Governor in control of a province and that he was, accordingly, to be regarded as a quasi-official and amenable to the provisions of section 134 CJA. Treacy J drew attention, inter alia, to Zardad’s admission that he was a general within Hekmatyar’s army (Hekmatyar being the leader of the Hezb-I-Islami faction), which controlled the Sarobi area, and to the clear command structure within that force. The judge referred to the fact that prisons within the controlled area were run by Hezb-I-Islami which was the only law enforcement authority in the area and to the role of Hezb-I-Islami and Zardad personally in arresting and imprisoning lawbreakers and in mediating and resolving disputes between individuals. Representatives of international organisations and aid agencies would make representations to Zardad, as opposed to any central government authority, if equipment was seized or delayed at any of the checkpoints for which Zardad’s force was responsible. Those who complained of ill-treatment, torture and hostage taking regarded Zardad and Hekmatyar as the only official authority in the area which was dominated and controlled by them. Against this background Treacy J concluded:
“The material to which I have referred in this judgment leaves it open for a jury to conclude that Mr Zardad was a de facto public official in an area which was totally controlled by Hezb-I-Islami and controlled by them with a degree of permanence. There is no evidence to show that at any material time the central government exercised any governmental function over the area controlled by Hezb-I-Islami. Such evidence as there is tends to show that Hezb-I-Islami had total control of the area in question. There is evidence that the Hezb-I-Islami faction exercised functions which could be functions of a state authority.” (at para 35)
US authorities
“However, torture and summary execution - when not perpetrated in the course of genocide or war crimes - are proscribed by international law only when committed by state officials or under color of law.”
and then recited the definition in article 1, UNCAT. In a passage cited earlier in this judgment at para 56 the Court of Appeals held that the proscription of official torture applied without distinction to both recognised and non-recognised States and continued, at p 245:
“Appellants’ allegations entitle them to prove that Karadzic’s regime satisfies the criteria for a state, for purposes of those international law violations requiring state action. Srpska is alleged to control defined territory, control populations within its power, and to have entered into agreements with other governments. It has a president, a legislature, and its own currency. These circumstances readily appear to satisfy the criteria for a state in all aspects of international law. Moreover, it is likely that the state action concept, where applicable for some violations like ‘official’ torture, requires merely the semblance of official authority. The inquiry, after all, is whether a person purporting to wield official power has exceeded internationally recognized standards of civilized conduct, not whether statehood in all its formal aspects exists.”
“… the beatings carried out by Vuckovic and his accomplices were clearly perpetrated, instigated, and acquiesced in, by persons acting in an official capacity as part of the police or military forces of Republika Srpska.” (at p 1346)
The appellant relies in particular on the following passage:
“… Vuckovic clearly committed abuses against plaintiffs under official authority. In light of the de facto governmental authority of the Republika Srpska, under which Vuckovic served as a soldier, and the control exerted over it by the Serbian government, Vuckovic may be considered also to have been acting under the authority of a ‘foreign nation’.” (at p 1347) (Emphasis added)
In this second passage, the court was considering whether the claim also satisfied the requirement of TVPA that the torture be “under actual or apparent authority, or color of law, of any foreign nation”. This does not detract in any way from the court’s earlier conclusion that the conduct was that of persons acting in an official capacity. On the contrary, the court’s reasoning supports the view that conduct of a de facto governmental authority can constitute official torture within article 1, UNCAT.
“The scope of the Convention is limited to torture ‘inflicted by or at the instigation or with the consent or acquiescence of a public official or other person acting in an official capacity’. Thus, the Convention applies only to torture that occurs in the context of governmental authority, excluding torture that occurs as a wholly private act or, in terms more familiar in US law, it applies to torture inflicted ‘under color of law’.”
The appellant points to the fact that the Court of Appeals then went on to draw attention to the definition of “under color of law” in the different context of 42 USC para 1983 (“[t]he traditional definition of acting under color of state law requires that the defendant … have exercised power ‘possessed by virtue of state law and made possible only because the wrongdoer is clothed with the authority of state law’”: West v Atkins, 487 US 42, 49 (1988) quoting in turn from United States v Classic, 313 US 299, 326 (1941)). Notwithstanding this equiparation of official conduct under UTCA and acting “under color of law” in a different context, the statement by the Senate Executive Committee and its adoption by the Court of Appeals makes clear that the distinction both sought to draw was between “torture that occurs in a the context of governmental authority” and “torture that occurs as a wholly private act”. Furthermore, Belfast was concerned solely with conduct which took place in Liberia after Charles Taylor had established himself as President of that State. Accordingly, it was not directly concerned with the question whether the conduct of a person acting on behalf of a quasi-governmental entity which is in de facto control of territory may give rise to official torture.
Academic commentators
“This premise, …, helps clarify why the Torture Convention sets out the requirement of the involvement of state officials for torture. … The requirement of a ‘state official’ is therefore needed to avoid that under international law a single conduct - although consisting of an infliction of severe mental or physical pain or suffering - be considered criminal when it is carried out by private individuals for private purposes. Such conduct is not of international concern and is therefore not covered by the Convention. In other words, the state official requirement constitutes what one could term the quid pluris, transforming an ‘ordinary’ criminal offence into an international crime. It simply serves the purpose of precluding every single wicked act carried out by private individuals against other private individuals from being elevated to the international level.” (at p 190)
Similarly, Burgers and Danelius (above) observe (at p 1) that UNCAT “does not deal with cases of ill-treatment which occur in an exclusively non-governmental setting. It only relates to practices which occur under some sort of responsibility of public officials or other persons acting in an official capacity.”
“Finally, under the UN Torture Convention, the ‘pain and suffering’ that is a necessary ingredient of torture must be inflicted ‘by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity’. The need for this sort of participation of a de jure or de facto state official stems from: i) the fact that in this case torture is punishable under international rules even when it constitutes a single or sporadic episode; and ii) the consequent necessity to distinguish between torture as a common or ‘ordinary’ crime (for example, torture of a former intimate partner by a sadist) and torture as an international crime covered by international rules on human rights.” (at p 133)
“But I would go further than you - one does not even need the involvement of a state official under the Convention. The Convention uses the expression ‘official capacity’ in contrast to ‘state official’. As you know, in the Zardad case in the UK this was interpreted to cover a person working against the state as part of a rebellion. One could not say that the state acquiesced in the torture, and yet the defendant was convicted and sentenced to 20 years. The judge said that in Afghanistan there may be ‘different types of structure’ which may ‘come within the ambit of an authority which wields power sufficient to constitute an official body’. It seems to me that one can have torture by an authority even where that ‘authority’ is fighting against the state. I admit that this is just one case. But as we know it is more likely that prosecutions will be brought for international crimes against non-state actors than against state actors. I have not found many national prosecutions for torture by state actors.” (at p 292, footnotes omitted)
“The committee stressed in the later HMHI v Australia, in which it distinguished Elmi on the facts, that its finding in Elmi as to ‘groups exercising quasi-governmental authority’ was restricted to ‘the exceptional circumstances of state authority that was wholly lacking’. But notwithstanding what the Committee reiterated was ‘the exceptional situation in Elmi’, the Committee’s view is open to doubt. Even more doubtful is the ruling in the English criminal case of R v Zardad (Faryadi) that, even where there exists a government within a state, the expression ‘a public official or other person acting in an official capacity’ in article 1 of the Torture Convention can extend to ‘people who are acting for an entity which has acquired de facto effective control over an area of a country and is exercising governmental or quasi-governmental functions in that area’.” (at para 7.121, footnotes omitted)
Unfortunately, however, in the absence of any further explanation it is difficult to understand the basis of the author’s objection.
Conclusion
75. It is time to draw the threads together.
“‘A person acting in an official capacity’ in section 134(1) of the Criminal Justice Act 1988 includes a person who acts or purports to act, otherwise than in a private and individual capacity, for or on behalf of an organisation or body which exercises, in the territory controlled by that organisation or body and in which the relevant conduct occurs, functions normally exercised by governments over their civilian populations. Furthermore, it covers any such person whether acting in peace time or in a situation of armed conflict.”
78. Thirdly, the exercise of a governmental function must be distinguished from purely military activity not involving any governmental function. I note that, in this regard, Treacy J in Zardad distinguished governmental functions from “the activities of a rebel faction which has not acquired a sufficient degree of control, permanence, authority or organisation to fulfil criteria sufficient for it to be recognised as an authority wielding official or quasi-official powers” (at para 36). However, insurrectional forces engaged in fighting the forces of the central government of a State may nevertheless exercise sufficient governmental authority over territory and persons under their control for acts done on their behalf to be official acts for this purpose. Thus, in Zardad the area controlled by Hezb-I-Islami was controlled essentially by military force but the group also exercised governmental functions. The failure to take account of the distinction between governmental and military activity leads me to the view that the formulation adopted by Sweeney J in the present case - “in situations of armed conflict, … individuals who act in a non-private capacity and as part of an authority-wielding entity” - is too broad. It is also necessary to bear in mind that there are circumstances in which torture might constitute a crime against humanity or a war crime contrary to UK law, whether or not performed by a public official or a person acting in an official capacity. However, for reasons explained at para 20 above, that is not so with regard to the alleged facts in the present case.
79. Fourthly, it is necessary to say something about what may be the indicative features of governmental authority in any particular case. I consider that Treacy J in Zardad correctly identified the required approach when he observed that it is necessary to look at the reality of any particular situation and to consider whether, at the relevant time, the entity in question had a sufficient degree of organisation and actual control over an area and whether it exercised the type of functions which a government or governmental organisation would exercise. This will require examination of evidence as to the position on the ground. In doing so it will be necessary to make allowance for the particular conditions which may make administration difficult and for different views of appropriate structures of government. The question will be whether the entity has established a sufficient degree of control, authority and organisation to become an authority exercising official or quasi-official powers, as opposed to a rebel faction or a mere military force. The one reservation I have about the approach of Treacy J. in Zardad is his view that the entity would be required to establish itself with a degree of permanence. This, it seems to me, is likely to be a flexible concept and the fact that the long-term survival of an entity may be an unlikely prospect should not prevent it from being considered a de facto government provided that it has effectively established itself as such. Furthermore, it is clear that the continued existence of a central government would not prevent an entity exercising the authority described above from being a de facto government in respect of the territory under its control. The application of this approach to the particular facts in Zardad has been considered at para 65 above.
Lord Reed: (dissenting)
“A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.”
Applying that general rule to article 1 of the UN Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (UNCAT), the ordinary meaning of the phrase “a public official or other person acting in an official capacity” does not in my opinion extend to a member of an insurgent group engaged in armed insurrection against the government of the country. That point does not admit of much elaboration. The ordinary meaning of the words “a public official” is reasonably clear, and would not in my opinion apply to such a person. The words “or other person acting in an official capacity” would, in their ordinary meaning, extend to persons who were not public officials but who were acting in a similar capacity, by reason for example of the outsourcing of public functions to private agencies. The core idea seems to me to be that the person in question is acting on behalf of the state. I have difficulty in applying the words “acting in an official capacity” to persons participating in an armed insurrection against the government.
85. The context also includes article 2(1):
“Each State Party shall take effective legislative, administrative, judicial or other measures to prevent acts of torture in any territory under its jurisdiction.”
The territory under the jurisdiction of a state would ordinarily be understood as being the territory over which it has de jure control. If torture carried out by insurgents in territory under their de facto control falls within the scope of UNCAT, it follows that article 2(1) imposes an obligation on states with which they cannot comply: they cannot take “effective” measures in relation to territory which they do not control. UNCAT cannot sensibly be interpreted in a way which would have the effect of imposing an obligation on states with which they cannot comply: lex non cogit ad impossibilia. That strongly suggests that article 2(1) cannot have been intended to apply in those circumstances, which in turn implies that the definition of torture in article 1 cannot have been intended to apply to torture committed by insurgent forces, without the consent or acquiescence of the state in question.
“The problem with which the Convention was meant to deal was that of torture in which the authorities of a country were themselves involved and in respect of which the machinery of investigation and prosecution might therefore not function normally. A typical case is torture inflicted by a policeman or an officer of the investigating or prosecuting authority. But many variations are conceivable. It could be that the torturer is not directly connected with any public authority but that the authorities have hired him to help gather information or have at least accepted or tolerated his act. All such situations where the responsibility of the authorities is somehow engaged are supposed to be covered by the rather wide phrase appearing in article 1: ‘inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity’.”
The two authors were actively involved in the preparation of UNCAT, Herman Burgers as chairman-rapporteur of the Working Group set up to draw up the text of the Convention, and Hans Danelius as the author of the initial draft of the Convention and as an active participant in all sessions of the Working Group.
“The Committee considers that the issue whether the State party has an obligation to refrain from expelling a person who might risk pain or suffering inflicted by a non-governmental entity, without the consent or acquiescence of the Government, falls outside the scope of article 3 of the Convention.”
The Committee went somewhat further in Elmi v Australia, Communication No 120/98, 14 May 1999, para 6.5, but in circumstances where there was no functioning state:
“The Committee notes that for a number of years Somalia has been without a central government, that the international community negotiates with the warring factions and that some of the factions operating in Mogadishu have set up quasi-governmental institutions and are negotiating the establishment of a common administration. It follows then that, de facto, those factions exercise certain prerogatives that are comparable to those normally exercised by legitimate governments. Accordingly, the members of those factions can fall, for the purposes of the application of the Convention, within the phrase ‘public officials or other persons acting in an official capacity’ contained in article 1.”
In SV v Canada, Communication No 49/1996, 15 May 2001, para 9.5, the Committee reiterated its established position:
“The Committee considers that the issue of whether the State party has an obligation to refrain from expelling a person who might risk pain or suffering inflicted by a non-governmental entity, without the consent or acquiescence of the Government, falls outside the scope of article 3 of the Convention.”
That was repeated in MPS v Australia, Communication No 138/1999, 30 April 2002, para 7.4:
“The Committee recalls its previous jurisprudence that the issue whether the state party has an obligation to refrain from expelling a person who might risk pain or suffering inflicted by a non-governmental entity, without the consent or acquiescence of the Government, falls outside the scope of article 3 of the Convention.”
The special nature of the Elmi case was made clear by the Committee in the case of HMHI v Australia, Communication No 177/2001, 1 May 2002, para 6.4:
“The Committee recalls its jurisprudence that the State party’s obligation under article 3 to refrain from forcibly returning a person to another State where there are substantial grounds of a risk of torture, as defined in article 1 of the Convention, which requires actions by ‘a public official or other person acting in an official capacity’. Accordingly, in GRB v Sweden, the Committee considered that allegations of a risk of torture at the hands of Sendero Luminoso, a non-state entity controlling significant portions of Peru, fell outside the scope of article 3 of the Convention. In Elmi v Australia, the Committee considered that, in the exceptional circumstance of State authority that was wholly lacking, acts by groups exercising quasi-governmental authority could fall within the definition of article 1, and thus call for the application of article 3. The Committee considers that, with three years elapsing since the Elmi decision, Somalia currently possesses a State authority in the form of the Transitional National Government, which has relations with the international community in its capacity as central government, though some doubts may exist as to the reach of its territorial authority and its permanence. Accordingly, the Committee does not consider this case to fall within the exceptional situation in Elmi, and takes the view that acts of such entities as are now in Somalia commonly fall outside the scope of article 3 of the Convention.”
“The Committee observes that the issue whether the State party has an obligation to refrain from expelling a person who might risk pain or suffering inflicted by a non-governmental entity, without the consent or acquiescence of the Government, falls outside the scope of article 3 of the Convention, unless the non-governmental entity occupies and exercises quasi-governmental authority over the territory to which the complainant would be returned.”
The Committee’s change of approach was neither acknowledged nor explained. On the contrary, the passages quoted above from Elmi v Australia, SV v Canada and MPS v Australia were cited as supporting this approach. The approach adopted by the Committee more recently in MKM v Australia, Communication No 681/2015, 10 May 2017, was seemingly more orthodox. It referred at para 8.6 to the failure of the state in question to provide protection from torture by non-state actors, and referred in para 8.7 to its General Comment No 2 (2008), discussed below, and to “the failure on the part of a state party to exercise due diligence to intervene and stop the abuses [by non-state actors] that are impermissible under the Convention, for which it may bear responsibility”.
“States bear international responsibility for the acts and omissions of their officials and others, including agents, private contractors, and others acting in official capacity or acting on behalf of the State, in conjunction with the State, under its direction or control, or otherwise under colour of law.”
The only situations in which the actions of non-state actors would be relevant were where the state consented or acquiesced in them, or failed in its duty under article 2 to take effective measures to prevent them. In that regard, the Committee stated at para 18:
“… where State authorities or others acting in official capacity or under colour of law, know or have reasonable grounds to believe that acts of torture or ill-treatment are being committed by non-state officials or private actors and they fail to exercise due diligence to prevent, investigate, prosecute and punish such non-state officials or private actors consistently with the Convention, the State bears responsibility and its officials should be considered as authors, complicit or otherwise responsible under the Convention for consenting to or acquiescing in such impermissible acts. Since the failure of the State to exercise due diligence to intervene to stop, sanction and provide remedies to victims of torture facilitates and enables non-State actors to commit acts impermissible under the Convention with impunity, the State’s indifference or inaction provides a form of encouragement and/or de facto permission.”
Following that approach, UNCAT would not normally apply to the conduct of insurgent forces within territory under their control.
“States parties should refrain from deporting individuals to another State where there are substantial grounds for believing that they would be in danger of being subjected to torture or other ill-treatment at the hands of non-state entities, including groups that are unlawfully exercising actions that inflict severe pain or suffering for purposes prohibited by the Convention, and over which the receiving State has no or only partial de facto control, or whose acts it is unable to prevent or whose impunity it is unable to counter.”
99. For the foregoing reasons, I would have allowed the appeal.