Can Private Actors Torture?

1. The international repression of torture may be connected to different rationales. In the decision in R v. Reeves Taylor (Appellant), the Supreme Court, while interpreting the meaning of ‘a person acting in an official capacity’ in section 134(1) CJA, examined two possible ratios, i.e. state-nexus and de facto governmental authority-nexus. What is, in your view, the essence and the foundation of the criminalization of torture in international law?

was already rooted in the 1975 UN Declaration against Torture, 5 and later became the main rationale of the UNCAT. Other ideas were also present, such as, for instance, preventive visits to places of detention as expressed in the proposal of the Costa Rica Protocol, 6 which later became the Optional Protocol to the Convention Against Torture. 7 The proposal advanced the idea of prevention of torture by means of unannounced visits to places of detention. This proposal was then postponed until after the adoption of the UNCAT, in order to avoid delaying the UN approval of the other treaty.
In my view, the UNCAT has today three main contents and rationales. First, fighting impunity. This objective creates, for the first time, an obligation of states to make torture a crime in their domestic law, ex Article 4 UNCAT. Secondly, prevention. This is enshrined, e.g. in Articles 10 and 11 UNCAT, on training of law enforcement officials and revision of the methods of interrogation. Lastly, on the rehabilitation of the victims of torture. Article 14 UNCAT, in particular, refers not only to compensation but also physical, mental, legal or social forms of rehabilitation. Thus, these three rationales (fight against impunity, prevention and rehabilitation) constitute the three main foundational components of the UNCAT.
Considering Article 4 UNCAT, which requires states to make torture a crime, it was important to provide a definition of the meaning of torture. This was the main reason why torture was defined in the Convention. This aspect is even more relevant since Article 4 UNCAT enshrines an obligation to criminalize which does not apply to other forms of cruel, inhuman or degrading treatment or punishment, referred in Article 16 UNCAT. In this respect, the representatives of Latin American countries were very strong in affirming that, in their view, torture was a pure state crime. Torture had to be connected to the military, the police, its intelligence services. They were not willing to open the definition to, for example, in the 1970s, the Sendero Luminoso, in Peru, or the Montoneros, in Argentina. The Latin Americans' wish was quite firm: torture was an exclusive state crime.
On a side note, I was very much involved in drafting the Convention on Enforced Disappearances (CED) in the 2000s, 8 where we had exactly the same discussion. At first, as an independent expert, I wrote a report to the Commission of Human Rights, where I explained the existing legal framework and what were the different available possibilities to deal with enforced disappearances at the time. 9 Were they enough? Or did we need a special Soon, I found myself in the company not only of the Russian Federation, but also of the representatives of Algeria as well as of many other Arab governments. They all agreed that the definition of 'enforced disappearance' had to be expanded beyond the state-crime. On the contrary, the Western and Latin American countries' representatives were arguing against this position. This is why this division led to a compromise solution: the definition of enforced disappearances, in Article 2 CED, is state-related, whereas Article 3 CED opens up also to non-state groups. Moreover, that was still a time when we did not speak too much about human rights obligations of non-state actors. There was still this old idea that human rights treaties are instruments between states and only states have obligations for themselves. Only later we developed the idea that there are obligations to 'respect', 'protect' and 'fulfill' human rights. In particular, the 'obligation to protect' means that the state has to take positive action to protect the individual against human rights violations by other individuals.
To conclude, these are the main reasons why, in my view, Article 1 of the UNCAT was drafted in such a narrow way. And whenever I had to consider or use the definition in Article 1, either as former UN Special Rapporteur on Torture and Other Cruel, Inhuman and Degrading Treatment or Punishment, or also as an academic, I always tried to interpret it as broadly as possible by means of the acquiescence element. So, for instance -if we take the case of female genital mutilation -I always considered that, even though it is committed by private actors, if the state is taking no action whatsoever, then it is the state that violates its obligation by means of acquiescence.
implications. When we discuss torture and its possible extension to various actors, what is/are the rule(s) banning torture that is/are concerned by the rationale? And to what extent should states that sign and ratify the UNCAT introduce in their legislation the definition of torture enshrined in its Article 1?
I would like to start by focusing on the second part of the question. Many states did not implement the definition of Article 1 UNCAT into their domestic criminal codes. This led to major problems, mostly considering the variety of definitions enshrined in national codes, which did not include certain elements which are instead recognized in the torture definition under Article 1. This is the reason why the Committee Against Torture decided to approach this issue by advising states to simply implement and apply the definition that we have in Article 1 UNCAT. 11 In this way, states would have been on 'the safe side', since such definition really covers all the aspects of torture, such as the mental and physical aspect. By way of example, as regards the severe' nature of the act, the US tried to redefine it as 'extremely severe pain'. 12 The approach adopted by the Committee allowed to simply recall directly the text of Article 1, which instead reads 'the deliberate infliction of severe pain or suffering' -the assessment of when the level of severity is reached being a question of interpretation for courts. More and more states have moved toward this direction and I believe it is a good and positive practice.
I would like to note that some states have also implemented, in their domestic laws, the last sentence of Article 1, which reads: 'it does not include pain or suffering arising only from, inherent in or incidental to lawful sanctions'. This clause has not been very useful and has caused various problems of interpretation: research has indeed shown that the 'lawful sanctions' clause lacks any meaningful scope of application. 13 In particular, a mere repetition of the text of the 'lawful sanctions' clause in domestic criminal law would only open a Pandora's box. For this reason, even if States consider that certain practices, such as certain forms of corporal punishment, would fall under this exception, they would be under an obligation to clearly define them in their respective criminal codes.
I do think that states that are following the definition under Article 1 UNCAT, at least up to the lawful sanctions' clause, simply do better. This however also means that, in their definition of torture, they have the 'state-nexus' very strong, the nexus with 'another person under official capacity' or, lastly, the acquiescence, which ultimately can help broadening the definition of torture.
If crimes against humanity, it reads as follows: '"Torture" means the intentional infliction of severe pain or suffering, whether physical or mental, upon a person in the custody or under the control of the accused; except that torture shall not include pain or suffering arising only from, inherent in or incidental to, lawful sanctions.' The article does not refer at all to the state nexus, but simply says 'the accused'. In addition, the acts fall under the definition of crimes against humanity when they are committed as part of a widespread or systematic attack directed against any civilian population. So, to keep the previous example of the Bosnian Serbs, non-state actors could be considered perpetrators of acts of 'torture' -and this is indeed the basis on which many of them have been convicted.
3. To build on your last consideration, and as a consequence of the rationale(s) discussed, what is then the spectrum of actors, public and/or private, that could fall under the category of 'perpetrators' of torture under different rules? Although human trafficking is not within the scope of our study, when we read that migrants are victims of abuses and torture during their journey to Europe, could we define migrant smugglers as torturers? We could expand the spectrum in the name of the protection of the victim's human dignity but, on the other side, one could also say that it would mean going too far.
I would like to discuss this question from two perspectives, de lege lata and de lege ferenda, i.e. by reflecting on what states should do and on how states and courts shall interpret existing law.
First, in my view, the very narrow definition of torture of the UNCAT is outdated. I always advised those states that have decided to implement the definition under UNCAT to adopt a twofold approach: to follow the definition in Article 1 UNCAT and also to recognize the crime of torture committed by actors beyond 'public official or other person acting in an official capacity'. An additional option could even be to include a different punishment on the basis of the perpetrator's status: the punishment could be higher if the act is carried out in an official governmental function, 14 while different if the perpetrator is a non-state actor -whether an armed group or simply an individual who is, for instance, deliberately inflicting severe pain or suffering on his wife. De lege ferenda, this is what states should do. This is even more so if we consider that a state implements the ICC Statute and it would have torture defined also as a crime against humanity, which can be committed by anybody, as long as it is 'part of a widespread or systematic attack directed against any civilian population, with knowledge of the attack' (Article 7 ICC Statute).
If torture is however not part of a widespread or systematic attack on the civilian population, then, all of a sudden, a private person cannot be held accountable. This contradiction is today present in various states' legislations that have implemented both instruments, i.e. the ICC Statute and the UNCAT, and thus reflect this legal loophole. 15 Therefore, also from this point of view, I think it is good if states include a definition that clearly goes beyond the mere state-authority. Certainly, if a court has to face the question of interpreting what is 'a person acting in an official capacity', it acts within the scope of Article 1 UNCAT if armed groups exercising de facto governmental authority are involved. This is exactly how the British court ruled in the case R v. Reeves Taylor, by interpreting Article 1 of the UNCAT definition, which finds full implementation by way of Section 134 of the 1988 Criminal Justice Act. This approach would help addressing possible delicate matters of accountability in case of acts of torture committed, for instance, in Transnistria, in Nagorno Karabakh, Abkhazia or South Ossetia. Moreover, even though they were not 'state' forces, the Tamil Tigers exercised de facto control of a certain territory. They were non-state actors that had enough power to control and certainly persons 'acting in an official capacity' in this sense. Whether they were military or simply political officials did not really make a difference. Lastly, in Bosnia and Herzegovina, both Ratko Mladi c, the highest military commander, and Radovan Karad zi c, the highest political commander of the Bosnian Serb forces, had effective control over the territory and were held accountable for their acts.
If we want to go further -namely, to apply the category to armed groups such as extremist groups or even organized criminal groups, which would not really exercise de facto control, i.e. governmental functions -then we will get into a slippery ground. This is also the position adopted by the Committee Against Torture. 16 Their status would still be a little bit unclear. If they are actors which fall in the category of paramilitary groups 17 and are acting with some support and with the knowledge of the government, they may be deemed public officials. Differently, if actors are fighting the government and do not have effective control on the territory, as also organized criminal groups may be, then they do not fall under the definition under Article 1 of the Convention. Whether you can say that it is the state to be held accountable because of acquiescence, it would not work if these are groups that are fighting the government. It should however be expected that a state, under the due diligence principle, makes efforts to adopt measures to control such non-state actors, to stop them and to ensure that they do not engage in acts of torture.
On this point, the Supreme Court discussed in detail the indicative features of governmental authority, also referring to the R v Zardad case. 18 In particular, the Court observed that 'the continued existence of a central government would not prevent an entity exercising the authority . . . from being a de facto government in respect of the territory under its control', 19 arguing also taking into account various views of the Committee against Torture, such as Elmi v. Australia. 20 Do you think this can be considered a settled interpretation of Article 1 UNCAT?
The Court here referred to non-refoulement cases of the Committee against Torture. 21 In the Elmi v. Australia case, the question was whether someone can be sent back to Somalia when the person could have been at serious risk of being tortured. The core question was if torture can only be considered as such if it comes from the Somali government and what can be done in the situation when there is no functioning government. The Committee against Torture concluded, in the case, that given that 'for a number of years Somalia has been without a central government', it 'follows . . . that, de facto, those factions exercise certain prerogatives that are comparable to those normally exercised by legitimate governments'. 22 This was a decision of some time ago. In my view, also in light of the very raison d'etre of the principle of non-refoulement, it should not really make a difference whether there is a central government or not. The state of Australia violates the Convention if its own people are torturing and, therefore, it does not really make a difference whether the state is torturing or whether it is handing the person over to another country where there is a serious risk that this person will be tortured. Logically, it really does not make any difference whether this is private torture or state torture because, ultimately, we hold the Australian state accountable.
Moreover, this is also the reason why the European Court of Human Rights has concluded, for instance, that the principle of non-refoulement applies also in cases related to the availability of medical treatments for HIV. By way of illustration, the Court recognized that it applies in the case where a person is HIV positive and in a late stage of the disease and can receive medical treatment in the UK, while not in the country where he could be sent back, thus risking his life. 23 Here, it does not matter whether these medical treatments come from the state or private actors. This extension of the principle is simply relevant in order to hold a government accountable in the case a person can face severe pain or suffering upon return. In addition, a woman may decide to flee her country because of being afraid of female genital mutilation ('FGM'). Again, the principle of non-refoulement should apply if a state knows that if sending this woman back to her country, she is most likely going back again to her community, where she will be subjected to FGM and receive no protection from the government. In this case, in my opinion, I would say that FGM constitute a form of torture, as it implies a direct control over somebody else, it is a deliberate infliction of severe pain and it is discriminatory on the basis of gender. All the definitional criteria are fulfilled. And if such acts are committed by a private actor, then we have a responsibility not to send this woman or girl back to her country. This is why I would really not make any kind of distinction. There is a general principle that criminal law must be extremely clearly defined, because it has important consequences for the persons who are accused and then convicted and sentenced for a crime. This is why, as we discussed before considering the perspectives de lege lata and de lege ferenda, it would be better if states are solving possible issues by defining torture in a much broader way so that everybody can actually be considered as possible perpetrators of acts of torture. As long as we have a criminal code that is simply reflecting Article 1 of the UNCAT, I think that the possibilities for a criminal court are very limited. As I said before, when we discussed what could enter in the definition of what is a 'public' official or a person 'acting in an official capacity', it is difficult to really go beyond the state nexus or the nexus with a de facto governmental authority. This does not regard the non-refoulement cases, since they relate to a different issue, as explained. If we are talking about the person who is accused of having committed or at least condoned torture, there must be a direct responsibility.
In international human rights law, however, I would go very far in clarifying what states' obligations are. I would affirm that the positive obligation to protect can extend to a person who could be victim, i.e. under the effective power, of an organized criminal gang. By way of illustration, trafficking is very much already part of the anti-torture discourse because the way victims of trafficking are treated is very often amounting to torture. Since states are under an obligation to combat trafficking, 24  Differently, as a criminal judge, I would adopt a stricter approach in order to really follow the dictum of criminal law and avoid creating an unclear situation on how broad the scope of the crime of torture can be. Politically speaking, limiting the approach to the state nexus remains wrong and too narrow. We should consider that, in the year 2020, more and more acts of torture are committed by different types of non-state actors, rather than by states. Since I do not believe that we will be able to amend the UNCAT, I see more important to recognize the states' obligation to broaden the scope of the definition of torture in their domestic criminal codes.