Guest Post: Detention in NIACs: A Pledge in Favour of the Application of IHL

Guest Post: Detention in NIACs: A Pledge in Favour of the Application of IHL

[Ezequiel Heffes holds an LL.M., Geneva Academy of International Humanitarian Law and Human Rights and is a lawyer, University of Buenos Aires, School of Law.]

Recently, the High Court of England and Wales delivered a judgement in Serdar Mohammed v. Ministry of Defence [2014] EWHC 1369 (QB) holding, among other things (see here for an explanation of the whole case), that the United Kingdom lacks detention authority under international humanitarian law (IHL) with regard to individuals it captures in the course of the non–international armed conflict (NIAC) in Afghanistan. In the present case, Justice Leggatt held that Common Article 3 (CA3) and/or Additional Protocol II (AP II) do not provide legal power to detain in the context of NIACs.

Much has been written about this in the blogosphere (see here, here, here and here by Gabor Rona a few years ago). From a theoretical perspective, these writings have raised several interesting arguments. This post, however, will focus on certain practical issues.  I will offer four arguments to suggest that it would be simply counter–intuitive not to recognize that IHL already regulates detentions in NIACs, even though it seems to be explicitly silent on this question.

The protection gap argument

The fact that CA3 and AP II neither mention internment nor elaborate grounds of detention has led to different positions on the legal basis for internment in NIACs. International bodies have prohibited such actions in cases other than when it is necessary for reasons related to the conflict. Here, the Inter–American Commission affirmed with regard to detentions carried out by the Colombian AOGs that “international humanitarian law also prohibits the detentions or internment of civilians except where necessary for imperative reasons of security”. The same view was held by the UN Commission of Human Rights (Resolution 1995/77) when it appealed to AOGs to refrain from “arbitrary” detention of civilians. As Zegveld points out, these bodies seem to have derived this prohibition from the IHL applicable to international armed conflicts, in particular Geneva Convention IV (Zegveld, Accountability of Armed Opposition Groups, at 65).

On the other hand, other resolutions by the UN Commission on Human Rights, such as Resolution 1995/74, deny that IHL permits certain civilian detentions on the grounds that human rights law is, in principle, a body of law only addressed to States, and only States have authority to arrest and detain persons. This would mean that, in the present case, IHL is silent while the international human rights law (IHRL) provision on arbitrary detention (Article 5 ECHR) would only be applicable towards the UK. This, however, represents a protection gap for detainees held by AOGs during the NIAC.  If IHL and IHRL do not apply upon them, then AOGs are able to operate within a legal ‘black hole’ and can in principle detain with impunity from an international law perspective (Somer, at 667–668). This necessarily implies that those detained by non–state actors have less protection than those detained by States, a situation that in the context of an armed conflict could not exist since IHL recognizes the principle of the equality of the parties (see the fourth argument by Kubo Mačák).

The judicial guarantees argument

CA3 affirms that “the passing of sentences and the carrying out of executions without previous judgement pronounced by a regularly constituted court, affording all the judicial guarantees which are recognized as indispensable by civilized people” is prohibited with respect to protected persons. Article 5 of AP II complements this by including several standards based on the more rigorous provisions of GC III and IV.

These provisions and possible detentions in NIACs should be seen through the same prism. Indeed, by granting AOGs the possibility to “regularly” constitute courts and to legislate in order to meet the judicial guarantees component (CA3), States have recognized AOGs’ legal capacity to run a parallel non–state legislative and judicial system outside of State authority (Somer, at 657). If States have accepted this guarantee (and therefore AOGs can declare someone innocent or guilty, or even permitting the person detained to challenge his or her detention), it would be simple logic to accept that they can also detain individuals under the same legal framework. In fact, they are both related since the application of judicial guarantees may serve to prevent indefinite detention in either situation.

The hostage taking argument

Alternatively, if AOGs are not able to detain members of State forces, then there is no practical difference between that situation and hostage taking, which is forbidden by CA3. The 1979 Convention against the Taking of Hostages provides a useful definition: “any person who seizes or detains […] in order to compel a third party, namely, a State […] a natural or juridical person, a group of person, to do or abstain from doing any act as an explicit or implicit condition for the release of the hostage commits the offence of taking of hostages […]”. Even though it was not drafted with a NIAC in mind, the definition contained therein could be considered appropriate in times of armed conflicts (Prosecutor v. Sesay, Kallon and Gbao, para 579) and might cover AOG detentions in the absence of authority under IHL.

As Sivakumaran explains, the means by which an individual enters the custody of the hostage–taker may be through lawful and unlawful means (Sivakumaran, The Law of Non–International Armed Conflicts, at 269). Certainly, the “hostage” label would come after the person has been taken away, regardless of how he or she is taken, but having in mind that an AOGs’ detentions will always be illegal under domestic legislation and not regulated as such by international law, the only characterisation that could frame such conduct under the latter regime would be the “hostage–taking” one. If AOGs cannot detain under IHL, then every person under their control against their will would be a hostage and therefore each detention would constitute an automatic violation of international humanitarian law.

The realistic argument

If none of the abovementioned arguments are enough, then we should just move towards a more realistic approach. This alternative proposes that it is simply unreasonable to consider that AOGs cannot detain individuals from an IHL perspective. As Sassòli (at p. 19) correctly suggests, “[p]arties to armed conflicts intern persons, hindering them from continuing to bear arms, as to gain a military advantage. If the non–state actor cannot legally intern members of government forces it is left with no option but either to release the captured enemy fighters or to kill them”. This implies that AOGs’ members might attack government soldiers instead of trying to legally arrest them. Even if under domestic law the killing of State forces is inherently illegal, no one says that it is prohibited per se under IHL. Yet, according to Sassòli’s argument, we could simply analyse possible detentions by AOGs from a “military advantage” perspective.

To conclude

This post has attempted to demonstrate some possible arguments as to why detentions in NIACs should be logically framed under IHL having in mind the recent decision in Mohammed of the High Court of England and Wales, which held the IHL to authorize such detentions. Certainly, these arguments do not solve all the issues raised in the context of NIAC detentions, but there can (and should be) room for new paradigms, particularly in light of how NIACs operate in the real world and the practical protection concerns that arise if the law were to remain truly silent.

International Human Rights Law
Notify of

This is an important addition to the discussion.  From my perspective, it seems quite appropriate to imply authority to detain from some of the materials that you address as well as overall general patterns of practice and general opinio juris or expectations about the legal propriety of those practices over many years. Yet, I would consider that your realistic argument is based in reality but would be considered to be a policy-serving argument — in any event, quite relevant.  Further, perhaps I missed something, but common art. 3 expressly prohibits that taking of hostages. In any event, the U.K. court was wrong to consider the ongoing armed conflict in Afghanistan with many outside state actors participating and continued fighting by the Taliban (which is at the very least a “belligerent” and a true belligerency, as opposed to a mere insurgency, is an armed conflict of an international character under customary international law that requires application of all of the customary laws of war — and also one with a “Power” per common art. 2), as being merely a NIAC.  For many reasons, it is an IAC.  See generally Paust, Bassiouni, et al., International Criminal Law 677-78, 682, 685, 697-98 (4 ed. 2013, Carolina… Read more »

John C. Dehn

Nice post, Ezequiel. The essential misunderstanding of the court and some commentators seems to be that law must affirmatively sanction measures taken in war. This is a simplistic and uninformed view of how international humanitarian law developed, which was as a constraint on permissible measures of war.

Just as international humanitarian law nowhere affirmatively sanctions the use of tanks, planes or machine guns, or even of drones or guided missiles, it does not exhaustively prescribe other permissible measures of war. Rather, it permits any war measures potentially helpful to defeating an enemy armed force that it doesn’t prohibit or delimit. As you correctly note, Marco Sassoli calls this military advantage. The U.S. calls it military necessity. It is the Lotus principle (what international law does not prohibit is permitted) applied to war. Whatever the continuing validity of the Lotus principle generally, it is still essential to understanding how international humanitarian law functions to constrain hostilities in armed conflict.

Ian Henderson

I can almost feel the response from Jordan coming as I type this, but why would international law, which principally is about the relations between states (and international organisations) authorise actions by non-state actors?
If international law authorised the detention of government forces by non-state actors, wouldn’t that mean in monist states that no domestic criminal law offence was being committed by non-state actors capturing and detaining members of the government forces? Would it also mean that, again in monist states, that no domestic criminal law offence was being committed when non-state actors killed members of the government forces?


Ian: the first SSRN (free download as annonymous) article identifies various non-state actors that have had formal participatory roles over the last 300 years or so.  Among them, at least by the 19th Century, have been “belligerents.”  “Insurgents” did not seem to have any formal status until the 1949 GCs, common art. 3.  As you know, common art. 3 does not provide combatant status or combatant immunity for  mere insurgents.  They remain liable to prosecution for murder or manslaughter with respect to what would otherwise be lawful targetings of persons in an IAC.  I asume that there would be no change in the law of war regarding killings merely because insurgents can have an implied authority to detain certain persons during a NIAC.  However, just because government troops have an implied authorization to detain insurgents and others during a NIAC (which would mean that they are privileged to do so) does not necessarily compel one to conclude that the insurgents have such an implied authorization, especially because they remain unprivileged fighters with respect to killing.  What would be important would be identification of general patterns of practice and general patterns of opinio juris regarding this precise issue.  I suspect that insurgents would remain unprivileged… Read more »


[…] Opinio Juris Guest post: detention in NIACs: a pledge in favour of the application of IHL.  […]


Ezequiel:  At least when the U.S. used military force in Afghanistan on Oct. 7, 2001, there had been a belligerency between the Taliban and the Northern Alliance to which all of the customary laws of war applied (including combatant status and combatant immunity for members of the regular armed forces of each belligerent), and at least three states continued to recognize the Taliban as the de jure authority in Afghanistan (while most did not).  The four traditional criteria for an insurgency were met: (1) semblance of a govt.; (2) responsible milt. command structure; (2) control of significant territory (Taliban over 90%, NA some terr.); and (4) fielding milit. units in sustained, continued, protracted, open hostilities.  Plus, outside recog. as belligerents or above (e.g., de jurs govt. vs. NA).  And once the U.S. and other armed forces engaged in sustained hostilities against the Taliban (not merely al Qaeda), there was certainly an armed conflict of an int’l character.  I suspect that the Taliban is still at least a belligerent vis a vis the govt. of Afg. and the armed forces of the U.S., etc. p.s.  I see the logic and policy-type claim, if not “reality,” re: privileged status for a mere… Read more »