International Humanitarian Law as Jus Cogens

International peremptory norms have received a variety of definitions and constructions. The definition relevant to the law of treaties states: a peremptory norm is such if it cannot be derogated from by a special legal act, i.e. if it cannot be replaced inter partes by a lex specialis rule, normally under the effect of voidness of the attempted derogatory act as set forth in Article 53 of the Vienna Convention on the Law of Treaties (VCLT). Under this lens, peremptory norms do not embody a legal hierarchy or sit at the apex of international law, giving them general precedence over all other norms. Although this latter proposition is often defended in legal writings on peremptory norms, it is not borne out by international practice. Moreover, the legal uncertainty such a construction would inevitably trigger is incompatible with the integrity of the international legal order.

It is often claimed that international humanitarian law (IHL) is in its greatest part peremptory, or at least a type of regulation akin to jus cogens and protected in the same way as peremptory law. The aim of this short post is to critically review the arguments advanced to buttress this position. That IHL contains peremptory norms cannot be doubted. The extent to which that is the case remains, however, debatable and debated. In this regard, the main arguments advanced to establish the peremptory character of IHL norms are of interest in the process of determining the precise scope of peremptoriness in our subject area.

Arguments for Peremptoriness

Contrary to other areas of international law, the arguments presented in favor of the peremptory character in IHL are not limited to the usual fundamental values embodied in the norm and the practices of States and international organs. The array of arguments advanced goes significantly further. The main arguments are set forth below. In their aggregate they present an impressive thrust in favor of the peremptory character of IHL norms. When disaggregated, however, the picture is less monolithic. And there remains in any case the need to give more precision to the scope of peremptoriness.

The Settled Views of States on International Law

One argument emphasizes that States have taken positions according to which the rules of IHL relating to the protection of the human person represent imperative law. This argument is of an inductive nature. It stands to reason that if the conditions for the emergence of a general rule of international law are met (in particular for universal customary international law), and States indicate that they believe not only that the rule is binding or ought to be binding (opinio juris), but that it is also peremptory, or ought to be peremptory (opinio juris cogentis), that rule becomes peremptory. Such an enquiry must be made rule-by-rule.

The Non-Derogation Clauses of the Geneva Conventions

Another argument focuses on the conventional clauses of non-derogation (articles 6/6/6/7 of the Geneva Conventions of 1949 (GC)) and non-waiver of rights by protected persons (articles 7/7/7/8 GC), as well as article 47 of GC IV (for occupied territories) and articles 51/52/131/148 GC (not allowing a carving out from grave breaches of the obligations of these conventions). This argument asserts that the fact that these conventions are practically universally ratified or acceded to allows a move from the conventional level to the customary level.

The non-derogation clauses of the GCs are closely akin to the jus cogens mechanism applicable to treaties: the general rule shall not be derogated from by special rules because of some collective interest in the integrity of the general rule. The collective interest in question refers to the humanitarian function of the rules contained in the GCs.

With respect to peremptory norms as codified in the VCLT, one difference springs to the eye. The mainstream jus cogens results in nullity of derogatory norms. But that is true only for pre-existing peremptory norms under Article 53 VCLT. For peremptory norms emerging after the adoption of the “derogatory” legal act, the legal consequence is only a termination of the obligations, and not their voidness, notwithstanding the confusing vocabulary of Article 64 VCLT.

Therefore, the question is where the emphasis is to be laid: is the peremptory mechanism triggered mainly by the non-derogation side (in that case the present clauses are plainly akin to jus cogens), or is it based mainly on the effect of the contravention, the nullity of the contrary legal act (in that case the present clauses will not be seen as a device akin to jus cogens). I tend to the first view and consequently perceive the close links of these non-derogation clauses with peremptory norms.

This peremptoriness is, however, softened or limited. Derogation is prohibited, but the legal effect of it is not the voidness of the derogatory legal act. It is rather the fact that if the derogatory act is implemented, the GCs will be violated, triggering international responsibility. There is a similar mechanism under Article 41 VCLT for derogatory agreements to a multilateral treaty. Here too, one could speak of softened or partial jus cogens. There is the injunction not to derogate, but without going to the sanction of voidness in case of contravention.

Finally, it must be noted that if many of the provisions of the GCs form peremptory norms under general international law, the non-derogation clauses discussed will be sided by a voidness flowing from the limb of general international law. Indeed, the effect of nullity will then stem from the “mainstream” peremptory character, under general international law. The non-derogation clauses will then become largely superfluous. But this is true only if the peremptory norms under general international law span as far as the non-derogation clauses. That is not proven.

The Nature of IHL: Reciprocity

Another argument holds that IHL is not of a bilateral nature but of an integral, objective, or absolute type. This is emphasized also by Article 60(5) of the VCLT. This argument is often stated but is not entirely accurate. First, the assertion that IHL (often also human rights law) is not based on reciprocity is only partially true. What is true is that human rights treaties have a triangular structure: States assume obligations towards a “third party,” the individual, and this triangularity modifies bilateral reciprocity, but does not eliminate it entirely, neither in terms of reservations nor in terms of implementation (as shown, for example, by the institution of the inter-State complaint).

The International Court of Justice (ICJ), in a recent case (Case concerning the Application of the International Convention on the Elimination of All Forms of Racial Discrimination, Azerbaijan v. Armenia, 2024) held that the 1965 Convention against Racial Discrimination could only give rise to claims between the two States from the latest date of entry into force inter partes of this text, even if it had to be respected already by the State which ratified it or acceded to it first, because it had entered into force vis-à-vis other States (§§ 41ff). No trace of non-reciprocity here.

In IHL, reciprocity is even less eliminated, and the triangular structure is even less pronounced. Thus, the Eritrea/Ethiopia Arbitration Commission was able to go so far as to affirm, on a legally fragile basis, but taking into account realistic considerations arising from international practice, that a State is not obliged to release without delay all the prisoners of war it is holding at the end of the war.

This position is in literal disrespect of the wording of Article 118 of GC III. The Commission considered that the detaining State could retain prisoners to stimulate a reciprocal release by the other party, when the latter delays or procrastinates (Prisoners of War case, Eritrea Claim No. 17, 2003). The Commission’s motivation was to allow the State ready to fulfil its obligations to keep means of pressure in hand to force the other State to comply with its obligations. There is no trace of a ban on reciprocity here, on the contrary. Has the Commission violated IHL? It is difficult to say. The persistence of certain forms of reprisals in IHL (outside the rights of protected persons) also shows that reciprocity has not generally disappeared. In any case, the result is that the lack of reciprocity is not a determining argument or solid basis in favor of sweeping jus cogensassertions in our area.

The Prohibitions on Reprisals

Other arguments stem from the prohibitions of reprisals (in the GCs and Additional Protocols (AP)), the criminalization of violations (war crimes), the jurisprudence (the “intransgressible” rules mentioned by the ICJ in the Nuclear Weapons opinion, the discussion of typical consequences of massive violations of jus cogens rules in the Wall opinion, the assertions of the International Criminal Tribunal for former Yugoslavia in the Kupreskic case). Moreover, the rules of IHL also have an erga omnes (partes) scope, as the ICJ recognized in the Wall advisory opinion.

Reprisals and the limitation of their scope in international law are not entirely congruent with the phenomenon of peremptory norms. A peremptory norm may be protected against reprisals; but other norms, non-peremptory, may also be so protected. This can be graphically seen in Article 50 of the Articles on Responsibility of States for Internationally Wrongful Acts (2001), where peremptory norms are just one category of obligations not to be affected by countermeasures, i.e. peacetime reprisals, refraining from the use of force. If that is true, a conclusion from duties of non-reprisal to peremptoriness is not logically warranted; there must be some other element(s) added.

The same is true for the criminalization of IHL through war crimes. Peremptory norms can give rise to war crimes, but so do non-peremptory norms. After the Second World War, a tribunal treated under the heading of war crimes the scuttling of a ship. It can hardly be asserted that this offence is based on a jus cogens norm. Moreover, war crimes concern only a small fraction of IHL positions. A short comparison demonstrates this. The Hague Regulations (HR), GCs and APs contain roughly 550 provisions. Article 8 of the Rome Statute of the International Criminal Court includes 60 war crimes, spanning over international and non-international armed conflict. The ratio is roughly 1:9. This ratio is insufficient to conclude that “IHL” in general forms peremptory law. Finally, the erga omnes-quality of a norm is not tantamount to qualifying it as peremptory. There are more erga omnes (partes) norms that there are jus cogens norms. To pass from one to the other without additional elements is therefore again not fully consistent with legal logic.

The Character of IHL: Specificity

Another argument states that IHL is a sort of minimum law for cases of armed conflict, applicable when most other rules cease to apply, and concentrated on maintaining a modicum of protection in a sort of bulwark against barbarism. The argument is sound but too generic to draw precise conclusions to the peremptory perimeter in the context of IHL.

The main principles of IHL may be peremptory and many obligations contained in the GCs can also be considered such in the light of the non-derogation clauses. But there are also a series of norms of an administrative character, which cannot just be equated fully with peremptory positions. Thus, the detaining power of prisoners of war must fill in capture cards of a certain type under Article 70 of the Third Geneva Convention (GC III). It would seem preposterous to claim that this is in all its contents a peremptory norm and that a delayed filling in of such a card (as happened to the United States in the Iraq war in 2003) would be a breach of a jus cogens norm. However, whether derogatory agreements to the duties of Article 70 would stand is manifestly another question, especially when looked at under the light of GC III, Article 6.

IHL as Applicable Exclusively to Breaches of Peremptory Norms

Another argument emphasizes that IHL applies normally when a peremptory norm has been breached, namely that prohibiting the use of force (Article 2(4) of the UN Charter). It would be illogical to consider that the rules applicable in case of a breach of a peremptory norm are not themselves of a peremptory nature.

The statement that the violation of a jus cogens norm triggers only legal regimes which are themselves jus cogens rests on a legal mistake. In our context, the use of force will not be in all cases based on a violation of the non-use of force rule contained in Article 2(4) of the UN Charter or customary international law. When the UN Security Council authorizes States to use force for a defined end under Chapter VII of the UN Charter, and if the acting States remain within the four corners of the authorization, the initial use of force is lawful; and yet the mandated States are obliged to respect IHL.

In case of an aggression, the State reacting against it is acting in self-defense. It therefore also uses force without the violation of a jus ad bellum norm, and yet it is also bound by IHL. Moreover, there are several rules triggered by an unlawful use of force which are not peremptory, e.g. the faculty to take countermeasures or sanctions; or to suspend treaties. It would be strange to suggest that because these acts respond to an unlawful use of force (seen as breach of a peremptory norm), they all become peremptory themselves.

Common Article 1 of the Geneva Conventions

Lastly, another argument posits that Article 1 of the GCs requires States to respect and ensure respect for their norms in all circumstances. The argument goes that to require respect of obligations under a treaty is apparently a useless utterance, because the principle pacta sunt servanda suffices to that effect (Article 26 VCLT, 1969). The only way to give sense to this word in Article 1 GCs is to interpret it as emphasizing the peremptory nature of the provisions contained in these conventions.

The duty to respect IHL “in all circumstances,” as stated in Common Article 1, can be interpreted in several ways. It is somewhat strange to see in it the hallmark of jus cogens, even if that interpretation is not absurd when looked at in the context of today. However, when the GCs were adopted, in 1949, peremptory law was not yet part and parcel of positive international law. It would therefore have been strange, at that time, that these terms should be read as referring to a notion which established itself much later, in 1969.

This shows that the drafters must have had something different in mind. What they did have in mind is to restate concretely the principle pacta sunt servanda, by recalling States that all their organs, also outside the military branch, and all mandated entities, whichever, should be instructed and controlled in order to ensure respect for IHL. This certainly shows the importance attributed to the subject matter but does not give a concrete hint to the extension of the jus cogens phenomenon in the polymorphic cradle of IHL.

Conclusion

All these sweeping arguments have a certain strength when looked at in conjunction, but they fail to give a precise account of the concrete extension of peremptory norms in the context of IHL. To be sure, more rules than in other branches of international law can be regarded as non-derogable in this area. However, to determine their exact extent and complexion, a painstaking analysis must be made.

What can be said is that the main principles of IHL are certainly peremptory: e.g. the principle of limitation (Article 22 HR, Article 35 AP I), the principle of distinction (Article 48 AP I), the principle of proportionality for excessive collateral damage (Article 51(5)(b) AP I), the principle that there must be an effort at precautions (Article 57 AP I), etc. The same can be said of many humanitarian norms contained in the GCs. Derogatory agreements could here not stand. To go further than these general statements, further precise legal analysis is needed. It will not be sufficient to utter sweeping arguments.

***

Robert Kolb is a Professor of Public International Law at the University of Geneva and former legal staff member of the ICRC. Prof Kolb is also a member of the legal section of the Swiss military high command (IHL section).

The views expressed are those of the author, and do not necessarily reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense. 

Articles of War is a forum for professionals to share opinions and cultivate ideas. Articles of War does not screen articles to fit a particular editorial agenda, nor endorse or advocate material that is published. Authorship does not indicate affiliation with Articles of War, the Lieber Institute, or the United States Military Academy West Point.

Photo credit: U.S. Army, Sgt. Terron Williams