When and where does the law of non-international armed conflict apply? Since most contemporary armed conflicts are fought between states and organized armed groups, or between such groups, these are important questions for both international lawyers and policy makers. The answers may affect the jurisdiction of U.S. military commissions, the detention of Taliban commanders and ISIL members, legal constraints on Saudi-led military operations in Yemen, and accountability for war crimes in Syria.
In this post, I’ll discuss the trigger and threshold of non-international armed conflict (NIAC). My point of departure is the much-discussed 2016 Commentary on the First Geneva Convention recently released by the International Committee of the Red Cross (ICRC). The University of Georgia School of Law recently hosted a fantastic event examining a number of issues raised by the Commentary, including the duty to “ensure respect” for the Convention by other Parties, incidental harm to sick and wounded combatants, and the classification of conflicts. This post grows out of that rich discussion.
The ICRC’s Commentary clearly states that an international armed conflict (IAC) “can arise when one State unilaterally uses armed force against another State even if the latter does not or cannot respond by military means.” Accordingly, the law of armed conflict constrains the first use of armed force by one state against another. Let’s call this a unilateral trigger.
In addition, “there is no requirement that the use of armed force between the Parties reach a certain level of intensity before it can be said that an [international] armed conflict exists.” Accordingly, minor skirmishes between state armed forces, or the capture of a single soldier, “would spark an international armed conflict and lead to the applicability of humanitarian law.” Let’s call this a nominal threshold.
Unfortunately, the Commentary is not so clear with respect to non-international armed conflict. The Commentary endorses the view that NIACs “are protracted armed confrontations occurring between governmental armed forces and … one or more armed groups, or between such groups.” This passage, as well as some cited authority, seem to suggest a bilateral trigger, requiring “armed clashes,” “combat zones,” or, simply, “fighting.”
The Commentary also states that, for the law of NIAC to apply, “[t]he armed confrontation must reach a minimum level of intensity.” Read alongside the Commentary’s discussion of IAC, it seems that this “minimum level of intensity” would not be met by minor skirmishes or by the capture of a single soldier or fighter.
The Commentary seems to accept a unilateral trigger and nominal threshold for IAC (quadrant 1) but a bilateral trigger and significant threshold for NIAC (quadrant 4).
In my view, we should accept a unilateral trigger and nominal threshold for both IAC and NIAC.
First, if an armed group is sufficiently organized, then a first use of armed force by or against that group should trigger a NIAC. Consider the following case:
Daesh: Daesh fighters pour over the Syria-Iraq border, killing Iraqi civilians, capturing Iraqi territory and taking over Iraqi government institutions. Iraqi forces flee, offering no resistance.
If we accept a bilateral trigger for NIAC, then the law of armed conflict does not apply until Iraqi forces “respond[s] by military means,” resisting Daesh’s advance. Until that time, Daesh fighters do not violate the law of armed conflict or commit war crimes. This result seems deeply unattractive. Although the Daesh fighters violate Iraqi criminal law, it seems hard to accept that they do not violate the law of armed conflict.
Now consider the following scenario:
Consent: State A attacks organized armed group G on the territory of State T, with the consent of State T. There is no pre-existing armed conflict between State A and group G. State A does not take feasible precautions in attack and recklessly kills many civilians.
If we accept a bilateral trigger for NIAC, then the law of NIAC does not apply until group G responds with military force, resulting in “armed clashes.” Since State T consents, the law of IAC does not apply either. It follows that State A does not violate the law of armed conflict or commit war crimes. This result seems intolerable.
Importantly, human rights law may not be sufficient to protect civilians or armed forces in cross-border cases like those described above. On most views, human rights law does not apply to the conduct of non-state armed groups that do not yet exercise territorial control and fulfill government-like functions. Moreover, according to some militarily active states, human rights law does not constrain extraterritorial lethal targeting by state armed forces. Yet, in my view, such conduct should be constrained by international law.
We should also accept only a nominal intensity threshold for NIAC. Consider the following case:
Capture: Members of organized armed group G mistakenly cross the unmarked border between State T, in which they normally operate, and State A. They encounter a unit of State A’s soldiers, and a minor skirmish ensues. No one is killed, but one group member is captured by the soldiers while one soldier is captured by the group and taken back across the border into State T.
In this case, it seems that both the group member and the soldier should be entitled to humane treatment under Common Article 3 of the Geneva Conventions. Moreover, if there are civilians present when the skirmish occurs, then it seems that the conduct of the skirmish should be constrained by customary rules including distinction, precautions, and proportionality If those rules are flagrantly violated, then those violations should amount to war crimes.
In my view, if an organized armed group has the capacity to sustain military operations then any military operation by or against that group should be constrained by the law of armed conflict. The organization and capacity of the group is sufficient to distinguish military operations by or against the group from “internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature.”
Some might worry that applying the law of armed conflict to first uses of low intensity force will displace or reduce the protections of human rights law. Fortunately, that is not the case. Even during armed conflict, states may choose not to derogate from their ordinary human rights obligations. Alternatively, derogation may be strictly required only with respect to certain measures, or only in certain parts of a state’s territory, or only in certain situations, even if the law of armed conflict provides additional constraints on all acts with a sufficient nexus to the conflict.
Most importantly, killings that are not prohibited by the law of NIAC may be prohibited by human rights law. In particular, “[w]here military necessity does not require parties to an armed conflict to use lethal force …, but allows the target for example to be captured rather than killed, the respect for the right to life can be best ensured by pursuing this option.” In situations of armed conflict, the law of NIAC may aid the interpretation of human rights law but does not exhaust the content of human rights law.
Finally, the law of armed conflict cannot authorize what human rights law forbids. As the ICRC observes elsewhere, “[t]he law relating to the conduct of hostilities is primarily a law of prohibition: it does not authorize, but prohibits certain things.” Human rights law and the law of NIAC do not conflict with each other but instead complement one another, as both impose constraints on violence rather than licenses to commit violence. As Additional Protocol II makes clear, human rights law “offer[s] a basic protection to the human person” while the law of NIAC aims “to ensure a better protection for the victims of  armed conflicts.”