IHL Does Not Authorise Detention in NIAC: A Response to Murray

by Kevin Jon Heller

Over the past couple of years, a number of scholars — including me — have debated whether IHL implicitly authorises detention in non-international armed conflict (NIAC.) The latest important intervention in the debate comes courtesy of Daragh Murray in the Leiden Journal of International Law. As the article’s abstract makes clear, Murray is firmly in the “IHL authorises” camp:

On the basis of current understandings of international law – and the prohibition of arbitrary detention in particular – it is concluded that international humanitarian law must be interpreted as establishing implicit detention authority, in order to ensure the continued regulation of armed groups.

I disagree that IHL cannot regulate non-state actor (NSA) detention in NIAC unless it authorises that detention, for reasons I will explain in this post. Before we get to Murray’s argument, however, it is important to remind ourselves of what is at stake in the debate. Put simply, if Murray is right and IHL authorises NSAs to detain, two significant consequences follow: (1) states have no right to prosecute NSAs who detain government soldiers, even if such detention would qualify as kidnapping or wrongful imprisonment under domestic criminal law; and (2) NSAs have the right to detain government soldiers for as long as they pose a “security threat” to the NSA — ie, essentially forever. In other words, FARC could detain a Colombian soldier for five decades and Colombia couldn’t prosecute the commander responsible for that detention as long as FARC complied with NIAC’s procedural restrictions on detention.

Now let’s turn to Murray’s argument. Here are the critical paragraphs in the article:

[I]nternational law cannot regulate activity that is subject to an absolute prohibition. For example, instances of torture cannot be regulated as torture is subject to an absolute prohibition. The same is true with respect to armed group detention in non-international armed conflict: the absolute prohibition of arbitrary detention precludes the possibility of regulating arbitrary detention (p. 9)

Two possibilities are open: either international humanitarian law establishes an implicit legal basis for detention, or it does not and the authority to detain must be established elsewhere. If international humanitarian law does not establish an implicit legal basis for detention then all instances of detention by armed groups will necessarily violate the prohibition of arbitrary detention as a legal basis for armed group detention does not exist under domestic law or elsewhere in international law. Yet, to interpret Common Article 3 and Article 5 Additional Protocol II in this way is to conclude that states have developed international treaty law to regulate detention operations by armed groups, despite the fact that all instances of armed group detention are illegal. This interpretation is incapable of giving effect to states’ intentions, and to the object and purpose of the provisions themselves. As discussed above, states cannot regulate that which is absolutely prohibited, and so the only means by which Common Article 3 and Article 5 Additional Protocol II can regulate detention by armed groups is if these provisions establish an implicit legal basis for that detention  (p. 14)

The first thing to note is that the torture analogy is misplaced. International law does indeed absolutely prohibit torture. But it does not absolutely prohibit detention — not even in NIAC. On the contrary, a state is free to detain as long as it adopts the necessary domestic legislation. It is even free to domestically authorise an NSA to detain, as well. (Which is not absurd. A state may well conclude that an NSA is more likely to treat captured government soldiers humanely if it does not prohibit the very act of detention.) So what Murray is actually arguing is that because most states choose not to authorise NSAs to detain, international humanitarian law (IHL) necessarily authorises it for them so they can regulate that detention. That’s a very puzzling claim, given that states are the authors of IHL.

The fundamental problem with Murray’s position, however, is that it is simply not the case that IHL can’t regulate a practice that international law absolutely prohibits. I will discuss in a minute the situation regarding detention in NIAC, in which the regulation and the prohibition come from different legal regimes — regulation from IHL, prohibition from international human rights law (IHRL). But before doing so, it is worth noting that Murray’s argument does not work even when the regulation and the prohibition come from the same legal regime — a situation in which you would think Murray’s argument would be even stronger.

Consider, for example, domestic criminal law. A number of American states have enacted so-called “crack taxes” — laws that require individuals who sell illegal narcotics to pay taxes on those narcotics. What are those laws, if not attempts by American states to regulate an activity that it absolutely prohibited? “Don’t sell illegal drugs, but if you do sell illegal drugs, pay your taxes on them.” No one, I think, would argue that crack taxes are valid only if the state is willing to decriminalise the sale of narcotics. But that is what Murray’s argument requires, if we apply it to domestic law — and I see no reason why why international and domestic law should differ in terms of the “regulation requires authorisation” rule.

Even if there is a reason to treat international law and domestic law differently in terms of Murray’s rule, IHL itself both prohibits and regulates certain activities — even in NIAC. Consider Art. 13(2) of the Second Additional Protocol:

The civilian population as such, as well as individual civilians, shall not be the object of attack. Acts or threats of violence the primary purpose of which is to spread terror among the civilian population are prohibited.

Art. 13(2) alone prohibits and regulates. Sentence 1 contains an absolute prohibition — thou shalt not use violence against civilians. Sentence 2 then add an additional regulation — if you use violence against civilians, thou shalt not do so in order to terrorise them. By Murray’s logic, Sentence 1 and Sentence 2 cannot coexist: if states want to regulate the use of violence against civilians (to prohibit using it to spread terror), it must first authorise using violence against civilians. But that makes no sense. And it makes no sense because Art. 13 actually contains two distinct prohibitions. If you use violence against civilians for reasons other than spreading terror, you violate only IHL’s prohibition on using violence against civilians. But if you use violence against civilians for the purpose of spreading terror, you violate both IHL’s prohibition on using violence against civilians and IHL’s prohibition on using violence against civilians to spread terror. It’s not an either/or.

And therein lies the fundamental flaw in Murray’s argument. What Murray describes as regulations concerning detention in NIAC are actually better understood as distinct and independent prohibitions of certain kinds of detention practices. Consider, for example, the first provision of Common Article 3 (CA3):

To this end the following acts are and shall remain prohibited at any time and in any place whatsoever with respect to the above-mentioned persons… (a) violence to life and person, in particular murder of all kinds, mutilation, cruel treatment and torture.

Murray describes this provision — and the other provisions in CA3 — as regulating acceptable detention in NIAC (p. 4). But CA3 itself describes those provisions not as regulations but as prohibitions: such acts “are and shall remain prohibited at any time and in any place whatsoever.” So just as IHL does not have to authorise using violence against civilians in order to prohibit using violence against civilians to spread terror, IHL does not have to authorise detaining individuals in order to prohibit treating detainees inhumanely. The two detention “prohibitions” — the act of detention and the inhumane treatment of detainees — are distinct and independent wrongs. So if an NSA wrongly detains but treats detainees humanely, the NSA only violates the “prohibition” of detention. But if an NSA wrongly detains and treats detainees inhumanely, the NSA violates both the “prohibition” of detention and the prohibition of inhuman treatment. It’s not an either/or.

The same analysis could easily be applied to torture, one of Murray’s primary examples of an activity that supposedly cannot be regulated because it is absolutely prohibited. Once we recognise that what Murray calls “regulations” are better understood as prohibitions, we can see that there is actually no reason why states could not both absolutely prohibit torture and “regulate” torture that is committed. States could, for example, require a doctor to be present whenever an individual is tortured. Or (perhaps less fancifully) they could require torture victims not be selected on the basis of characteristics such as race or religion. In fact, CAT arguably already regulates torture, despite CAT itself absolutely prohibiting it. How else should we understand Art. 15, which prohibits a state that tortures from using statements obtained through torture in civil or criminal proceedings? Does Murray believe Art. 15 is invalid?

In the context of detention in NIAC, however, there is an additional wrinkle to the analysis: namely, that IHL does not prohibit NSAs from detaining government soldiers. IHL neither authorises nor prohibits such detention. That is why I put scare quotes around “prohibition” in “prohibition of detention” in the discussion of CA3 above. Instead, the supposed “absolute prohibition” of NSAs detaining in NIAC comes from a different legal regime entirely — IHRL. Murray openly admits as much (p. 6):

All the major human rights law treaties ‘provide that no one shall be deprived of his liberty except on grounds and in accordance with a procedure that are established by law’… It is exceptionally unlikely – if not an absolute impossibility – that a legal basis for detention by an armed group will exist under domestic law. Indeed, armed group detention is typically regarded as a form of unlawful confinement, subject to criminal sanction.

Murray’s real argument, therefore, is that  IHL has to authorise NSAs to detain in NIAC because IHRL will otherwise absolutely prohibit such detention, thereby prohibiting IHL from regulating it. But there are two fundamental problems with this argument.

The first is that it is ahistorical. As Murray acknowledges, IHL first began to regulate NIAC detention in 1949, when the Geneva Conventions were adopted. But in 1949 there was no rule of IHRL that prohibited all arbitrary detention. That IHRL prohibition did not evolve until much later — after the adoption of the ICCPR, according to the Working Group on Arbitrary Detention:

42. The prohibition of arbitrary deprivation of liberty is recognized in all major international and regional instruments for the promotion and protection of human rights. These include articles 9 of the Universal Declaration of Human Rights and the International Covenant on Civil and Political Rights, article 6 of the African Charter of Human and Peoples’ Rights (African Charter), article 7, paragraph 1, of the American Convention on Human Rights (American Convention), article 14 of the Arab Charter on Human Rights (Arab Charter), and article 5, paragraph 1, of the European Convention for the Protection of Human Rights and Fundamental Freedoms.

43. Currently, 167 States have ratified the International Covenant on Civil and Political Rights, and the prohibition of arbitrary deprivation of liberty is widely enshrined in national constitutions and legislation and follows closely the international norms and standards on the subject. This widespread ratification of international treaty law on arbitrary deprivation of liberty, as well as the widespread translation of the prohibition into national laws, constitute a near universal State practice evidencing the customary nature of the arbitrary deprivation of liberty prohibition.

By Murray’s own logic, therefore, IHL did not have to implicitly authorise NSAs to detain in 1949, because international law did not yet absolutely prohibit unauthorised (and thus arbitrary) detention. But that changed sometime between 1949 and 1977, when states “regulated” NIAC detention again by adopting AP II, because international law — via IHRL — suddenly prohibited unauthorised detention of any kind. At that point, IHL did have to implicitly authorise NSAs to detain, because otherwise the absolute illegality of such detention in NIAC would have prohibited states from regulating detention through AP II. So voila! By some unforeseen force — certainly not the actions of states — IHL developed the required implicit authorisation for NSAs to detain. That is a very strange explanation of IHL, to say the least.

The more significant problem, however, remains the one identified above: because wrongful detention and inhumane detention are separate and distinct prohibitions, there is no reason why IHRL can’t absolutely prohibit NSAs from detaining (because neither international law nor domestic law authorises such detention) while IHL regulates NSA detention (by prohibiting NSAs from treating detainees inhumanely). Unauthorised detention violates IHRL (and very likely domestic law), but not IHL; inhumane detention violates IHL (and likely IHRL). It’s not an either/or.

Finally, let me end by noting that Murray’s argument, if correct, is not limited to detention by NSAs. Everything he says applies equally to how NSAs use force during NIAC. IHL does not authorise NSAs to use force against anyone, not even against enemy combatants — they don’t have the combatant’s privilege. Because that is the case, NSAs are in effect absolutely prohibited from using force against government soldiers: any such use of force (except in the case of personal self-defence) will violate domestic criminal law. Despite that absolute prohibition, however, IHL nevertheless regulates how NSAs use force against government soldiers. NSAs are categorically prohibited, for example, from causing a government soldier unnecessary suffering. But Murray’s argument says that situation is not possible: IHL can’t regulate what is absolutely prohibited. So either IHL implicitly authorises NSAs to use force against government soldiers — which would come as something of a surprise to states — or all of IHL’s “regulations” concerning how NSAs use force are invalid. Neither position, however, is tenable.

Murray’s article is interesting and provocative. But it is no more successful than previous attempts at arguing that IHL authorises detention in NIAC. It doesn’t.


3 Responses

  1. Interesting post, Kevin. I can certainly see both sides of the argument. I think you have the better of it, though, contrary to my own earlier inclinations.

    I would say this, though. The issue of the actual legal authority to do anything in NIAC is typically one of domestic law. I therefore do not think what international law permits or or implicitly authorizes is relevant to who may prosecuted under domestic law (although this may be semantics to some extent). At some point I plan to write more about why that is the case.

  2. I agree that under customary and treaty law there is no right of non-state armed groups to detain in NIAC -unless expressly authorized to have POW by States-. Moreover, arguing on behalf of such a “right” risks generating a slippery slope and making the aforementioned groups think that they are entitled to engage in the respective conduct. I am a Colombian international lawyer and argued in that sense in this blog some years ago, here: http://opiniojuris.org/2014/11/19/guest-post-suspension-colombian-peace-talks-illegality-deprivation-liberty-members-state-armed-forces-non-international-armed-conflicts/

Trackbacks and Pingbacks

  1. […] IHL does not authorise detention in a NIAC: a response to Murray.  […]