Home EJIL Analysis IHL Does Authorize Detention in NIAC: A Rejoinder to Rogier Bartels

IHL Does Authorize Detention in NIAC: A Rejoinder to Rogier Bartels

Published on February 24, 2015        Author: 

We are grateful to Rogier Bartels for his thoughtful comments on our recent post and article in which we argue that IHL authorizes State parties to a NIAC to detain suspected insurgents. In this rejoinder, we briefly respond to Rogier’s main criticisms of our argument.

Equal protection versus equal status

The crux of Rogier’s criticism flows from his understanding of what the principle of equal application requires. For Rogier—as well as for Leggatt J in Serdar Mohammed and Dapo Akande and Lawrence Hill-Cawthorne (see here and here) —‘a principle of IHL has to apply equally to all sides; otherwise it cannot be a principle’. All parties to a NIAC, both States and non-state actors, must have exactly the same rights (including authorities) and obligations under IHL. Thus, our position that IHL authorizes States (but not organized armed groups) to detain produces unacceptably ‘asymmetrical rules’.

As we explain in our article, this ‘symmetry’ objection stretches the principle of equal application beyond its breaking point:

If the principle [of equal application] demands that all belligerents must enjoy the same status and rights and CA3 does not confer the full panoply of belligerent status and rights on non-State actors, then the only logical conclusion is that the parties to the Geneva Conventions and AP II gave up their status and rights as States and assumed the same status and rights as non-State actors. This not only contradicts commonsense, but also the plain language of CA3, which declares that it does not affect the legal status of the parties, State and non-State alike, to the conflict. In fact, CA3 thereby conserves any pre-existing inequality between the belligerent status and rights of State and non-State parties to a NIAC.

The principle of equal application requires that the protections and obligations under IHL apply to all parties to an IAC or NIAC whatever the lawfulness of resort to force under the jus ad bellum. Entitlement to protection is not dependent on how the conflict began or the relative justice of the causes involved. Similarly, the scope of IHL obligations should not be linked to organizational capacities or military rationales. However, none of this alters the fact that there is an undeniable asymmetry in the status of parties to a NIAC. One is a State and the other is not. The fact that an internal situation rises to the level of a NIAC does not transform the State party into a non-State actor or vice versa. As René Provost notes in his comments on the debate between Marco Sassòli and Yuval Shany referred to by Rogier:

Equality as it emerges from the ways in which it is invoked in discussions of the laws of war evokes an idea of equality as necessarily grounded in sameness. …[H]owever much we might push to make non-state armed groups like states, having courts with due process and so on, in the end they are not states at all. (here, at 439–440)

Our position preserves equality of protection while recognizing inequality of legal status and authorities under IHL. This balance neither delegitimizes IHL nor reduces the incentive (on either side) to comply with obligations. Rogier’s charge that this approach is unfit to bear the name ‘IHL’ is informed by his own view that IHL rules applicable in NIAC are purely protective.

Regulation v. authorization

Rogier also objects that we ‘fail to acknowledge’ the distinction between regulation and authorization and fail ‘to appreciate the realism that underlies IHL’. However, our position is not based on the assumption that IHL must authorize status-based operations (targeting and detention) simply because it regulates warfare. It is, of course, entirely possible for international law to regulate a matter without authorizing it. This is why, as Rogier notes, we ground our position in the ‘rationale and principles of IHL’.

Similarly, we do not deny that IHL is pragmatic or realistic in its outlook. Neither, however, is IHL entirely agnostic when it comes to the legality and legitimacy of using lethal force against insurgents. Rogier’s eloquent historical narrative of the development of the law of NIAC is, of course, broadly correct. However, his view that the law contains ‘only protective rules’ rests on an incomplete account. First, contrary to Rogier’s suggestion, Article 4 AP II was not simply transplanted from Article 75 AP I, but was based directly on Common Article 3 of the Geneva Conventions of 1949 and the relevant provisions of the International Covenant on Civil and Political Rights  (see Official Records of the Diplomatic Conference on the Reaffirmation and Development of International Humanitarian Law Applicable in Armed Conflicts, Vol VIII, 324). More importantly, the reference in Article 4 AP II to ‘[a]ll persons who do not take a direct part or who have ceased to take part in hostilities’ was understood to extend to an adversary hors de combat (Official Records, Vol VIII, 335). This is an important point. It illustrates not only how insurgent fighters were considered as ‘combatants’ throughout the negotiations, but also that their treatment follows the same logic as the treatment of combatants in an IAC (status-based targeting subject to the exception applicable to those hors de combat).

Second, whatever the early origins of the law of NIAC, things have moved on. There has been greater convergence between the laws of IAC and NIAC, including in particular the rules on targeting. Moreover, it is now necessary to accommodate the parallel application of IHL and IHRL as separate regimes without rendering either redundant. Rogier’s position would require that primacy always be accorded to the more restrictive rules of IHRL, at least under the ECHR (and in the absence of derogations or Court-made ‘accommodation’). As a result, when it comes to status-based operations, IHL would be deprived of any meaningful effect. Our point in all of this is that the existence of authorization to detain is implied and demanded by the logic of IHL in NIACs.

Transplanting the law of IAC?

Rogier objects that reliance on the permissive aspect of the principle of military necessity would involve transplanting ‘wholesale’ the law of IAC into NIAC. Our argument does not depend on making a new conceptual leap. States have already recognized the legitimacy of using lethal force against ‘combatants’, based on their status as fighters and by extending the rules on targeting to NIACs. (We suggested that the ECtHR recognized this practice in Korbely. Yes, the reasoning of the Hungarian courts in that case was less than convincing, but the ECtHR made its own findings. While the ECtHR did not distinguish between a civilian participating directly in hostilities and a person in continuous combat function, the key point is that it based its analysis on the distinction between combatant and non-combatant status under IHL). Second, however, it does not follow that military necessity in NIAC is as broad as in IAC. The ‘complete submission of the enemy’ may not necessarily mean the same thing in each context.  However, this does not undermine our argument that the principle of military necessity applies in NIAC.

Capture, release and kill?

In our article, we criticize Leggatt J’s reasoning and put forward a positive case for a legal basis for States to detain insurgents in NIAC. Some of Rogier’s objections to our position are based on a conflation between these separate arguments. For example, we argued that Leggatt J’s finding that fighters may be targeted or detained only if they represent an ‘imminent threat’ would lead to an absurd circle of capture, release and immediate re-capture or targeting. Our point is that such an absurd circle does not reflect the logic of IHL in NIAC (nor is it compatible with the letter of the law, as Rogier helpfully points out).

Print Friendly

7 Responses

  1. Jordan

    Thanks, makes sense. And yes, insurgent fighters in a NIAC do not have pow status or combatant immunity (and are not actually “combatants” but fighters), so there is no perfect equality. But please do not adopt the phrase “imminent threat” (and imminent threat, logically, is not yet a present “threat”). Moreover, civilians are not targetable merely when they actually pose a “threat” (of what sort, to whom, and so forth).

  2. J.T.

    It wasn’t only the Hungarian judges, who misunderstood the law of armed conflict. The judges in Strasbourg too made a mess of the law in Korbely. Using that case as an authority is the same as saying that combatants are prohibited to target civilians only ‘when not justified by military necessity’, because the ICTY said this in Blaskic.

  3. J.T.

    That is not too say that Sedar Mohammed must be correct and THE authority to rely on (because it too is just a single case). However, when you aim to show that LOAC is not actually the way it is written, the burden is on you to come with authorities.

  4. Thanks, Jordan. The phrase “imminent threat” is Leggatt J’s, not ours. We reject that test not only due to its operational absurdity but also because, unlike international human rights law, targeting and detention in the law of armed conflict (IAC and NIAC) is status-based.

    J.T., “the way the law [of armed conflict] is written” neither expressly permits nor prohibits detention in NIAC. In the absence of settled authority, our interpretation of AP II and CA3 proceeds from context, object and purpose and travaux. As we explain above, Korbely is consistent with our interpretation.

  5. Rogier Bartels Rogier Bartels

    Dear Sean and Aurel,

    Thank you for this interesting reply. I had a couple of deadlines and am only now finding the time for some (belated) comments. Not a full sur-reply, but some things that are worth raising or clarifying.

    I have trouble understanding your comment that “things have moved on”. I had understood your argument to be that authorisation was implied. That would seem to mean from the beginning onwards. Are you now saying that it actually only later became implied (inter alia, due to the external factor of the emergence of an IHRL regime)? How does that work?

    Relatedly, it is difficult to understand the reference to Korbely in the above post. Korbely looked at the law in 1956; a time that the rules on targeting had not yet been extended to NIACs and things had not yet “moved on” in the few years since CA 3 was agreed upon and (a little later) had entered into force.
    In addition, I find it difficult to understand your reading of Korbely. The ECtHR did not conclude anything beyond that the alleged victim was not (at the time he was killed) a protected person under CA 3. That was reasonable and – based on the facts as summarised by the Court – the applicant had actually quite unfairly been convicted for the killing. Not only because it is doubtful (based on the summarised facts) that a NIAC existed at the time, but also because the victim could be considered as DPHing, or at the very least a situation of personal self-defence existed (“… [the victim] drew his gun with unknown intentions”). The Court’s reasoning may well have been outcome-based, but even if it was not, the Court did – as just mentioned – nothing more than concluding that no protective rule was violated: “… [the victim] did not fall within any of the categories of non-combatants protected by common Article 3”.
    Besides showing the Court’s focus on the protective scope of CA3 (indicating that deriving support for status-based targeting from the Court’s conclusion is a bit of a stretch), this short phrase shows (potential) misunderstandings of CA3, as well as IHL as a whole. Obviously there are no “combatants” or “non-combatants”, but the Court could be forgiven for using that terminology. It may have been a faulty translation from a draft judgment in French, or the Court may have just wanted to juxapose persons protected under CA 3 with those who are not protected. However, if the Court really believed combatant and non-combatant status existed, this would have a further misunderstanding as a consequence. If someone would have been a “combatant” before falling within the categories protected by CA3 of “members of armed forces who have laid down their arms and those placed ‘hors de combat'”, that person remains a “combatant” – just one who is protected. The Court’s conclusion then seems to indicate the opposite of a continuous combat function or status-based targeting. Moreover, if you want to suggest that it supports your argument, it also means that you are willing to forgive the EtCHR something that you criticise Leggatt J for: not appreciating that a “fighter” does not become a civilian when hors de combat. If Korbely does at all support your argument, it would only be the use of the word “non-combatant”. If every time the word (non-)combatant is used in the context of a NIAC is supposed to imply recognition of status-based targeting, there are other sources to refer to (the ICRC’s commentary to AP II, for example). But in most instances it will be meant to explain who is protected in a NIAC, rather than to say that detention of the non-protected group is authorised.

    As I already said in my post, your arguments requires us to first accept that status-based targeting is implied in the law of NIAC. Then only can your second step be taken. The alleged problems with not taking or agreeing with the second step, in part, arise because of your insistence that the first step is the only correct one. Similarly, your criticism of Leggatt J, and in your article also Debuf, in part, follows from their perceived unwillingness to consider continuous combat function as an authorisation to kill and consequently to detain. However, this again requires accepting that your point of departure is correct – as opposed to theirs. I merely tried to point out that relying on the continuous combat function does not provide all the answers, as it also leads to problems.

  6. Rogier Bartels Rogier Bartels

    The above quotation of René Provost’s piece in the IRRC is indeed fitting to support your argument, but René’s piece read as a whole (as well as his excellent longer piece in Benjamin Perrin’s Modern Warfare) does not appear to support your argument – quite the opposite. René’s discussion of about equality of belligerents relates to the obligations and duties, i.e. protective rules, and the practical difficulties non-State actors may have to ve up to these duties. He recognises that you cannot really call a State and a non-State actor to be formally equal, He therefore advocates for “substantive equality” and proposal a model by which the non-State actor interest is taken into account. As such, his piece is actually very much in disagreement with the inequality in ‘rights’, with the law of NIAC allegedly conferring rights or authorisations on States, advocated by you.

    Contrary to what it says above, I did not say that “Article 4 AP II was […] simply transplanted from Article 75 AP I”. I merely pointed out that the use of the word “persons” in AP II makes sense, as it is the NIAC-version of Article 75 AP I also using “person(s)”. I admit my choice of words was somewhat poor, and you are certainly correct that the origins for Art. 4 can be found in CA 3 and (mainly) the ICCPR, but I do not see how that helps your argument. The ICCPR’s use of the word persons referred to every human and, given its purpose, naturally, did not make any distinction between combatants/fighters and civilians.
    There appears to be no dispute that the protections of APII “extend to an adversary hors de combat”. Of course, persons hors de combat are protected by the fundamental guarantees. Nevertheless, I do not consider the fact that the Iraqi delegate remarked so to be relevant or particularly convincing. The Iraqi delegate made many more comments, several clashing with your interpretation of the law of NIAC. If we are to view all single remarks made by the delegates and not further addressed by the other drafters, as proof of a certain implied basis, there are many more remarks in the Official Records that do not support your argument (see, e.g., Jonathan Horowitz’s post on OJ). In sum, the fact that Art. 4 was meant to and does protect the various types of persons protected under IHL does not imply an authorisation. If the law prohibits A, but is silent on B, one conclusion could be that as B is not prohibited by the said law, it is permitted. The conclusion that B is authorised does not follow from the silence.

    You put forward an interesting case that the law of NIAC is different from the understanding that I and, e.g, Dapo, Lawrence, Kevin, Derek, Jonathan, have of it. Yet, if supported, your argument appears to only be applicable to so-called multinational NIACs, but I have not yet seen convincing evidence for the general law of NIAC as such. It is an interesting discussion to engage in, however. Thanks for the debate.

  7. Jordan

    Sean: just a note re: “status-based” — targeting and detention can also be conduct-based (e.g., where a civilian is a direct participant in hostilities or, because of conduct, a person who can be detained under GC, art. 5).