When powerful States adopt a mistaken view of international law, should we—scholars, practitioners, and activists—resist their view and insist on our own? Or should we regard their mistake as a fait accompli and try to contain its adverse consequences? With Donald Trump poised to become the next President of the United States, I fear that we will all face such questions in the years to come.
In late September, I argued that, just as the law of international armed conflict applies to the first use of military force between States, the law of non-international armed conflict applies to the first military operation between States and organized armed groups or between such groups. Such first strikes are constrained by the law of armed conflict, and serious violations of those constraints may constitute war crimes.
Two weeks ago, Jonathan Horowitz wrote a serious and thoughtful response to my post, concerned that the application of international humanitarian law (IHL) to first strikes would preclude the full application of international human rights law (IHRL) to first strikes. In my original post, I anticipated that
[s]ome 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.
I went on to explain that IHRL applies to first strikes and continues to apply throughout armed conflict. The parallel application of IHL does not relieve States of their obligations under IHRL or determine the content of those obligations. Instead, these two bodies of law each offer important constraints on the use of force and corresponding protections against violence and abuse.
Importantly, Jonathan suggests that we should resist the application of IHL to first strikes, not because IHL actually displaces or reduces the protections of IHRL, but because some States mistakenly believe that IHL displaces or reduces the protections of IHRL. If these States accept my view that IHL applies to first strikes, then they may erroneously conclude that IHRL does not.
Jonathan raises a legitimate concern. Nevertheless, I do not believe that we should address this concern by resisting the application of IHL to first strikes. Instead, we should address this concern by insisting on the parallel application of IHRL.
In general, if States hold a mistaken view about an important legal question then we should resist their mistake. We should articulate, defend, and rally around an interpretation of international law that best reflects its underlying values. Indeed, Jonathan seems to concede this general point when he insists that IHRL applies to extraterritorial uses of force, a view rejected by many powerful States.
In particular, the mistaken view that IHL displaces or reduces the protections of IHRL is far more dangerous than any view about when IHL starts to apply. After all, whenever IHL applies, it is the relationship between IHL and IHRL that determines the legal protections that individuals enjoy against death, injury, and detention. We should directly challenge this mistaken view, not resign ourselves to containing its consequences at the margins.
In my view, the “Trojan horse” to which Jonathan alludes is not the application of IHL, but rather the mistaken view that the application of IHL displaces or reduces the protections of IHRL. We should not let it through our gates.
Parallel Application of IHL and IHRL
In a representative passage, Jonathan writes that
Once IHL applies, States will have more latitude to argue that they aren’t, or aren’t fully, bound by IHRL.
On this view, once IHL applies, the protections of IHRL are either displaced by the protections of IHL or must be narrowly interpreted to coincide with the protections of IHL.
The view that the application of IHL relieves States of their obligations under IHRL seems legally unsustainable. As I noted in my original post, Additional Protocol II states that human rights law “offer[s] a basic protection to the human person” while IHL aims “to ensure a better protection for the victims of  armed conflicts.” There is no indication that the protections of Protocol II were intended to displace the protections of IHRL during armed conflict.
The displacement view is also incompatible with the European Convention on Human Rights, which permits a State Party to take measures derogating from its provisions “[i]n time of war” but (a) only “to the extent strictly required by the exigencies of the situation” and (b) only if that State informs the Council of Europe of the measures taken as well as the reasons for taking them. It is not the existence of armed conflict—or the application of IHL—that determines the permissibility of such measures, but rather the affirmative act of formal derogation together with the concrete factual circumstances.
For example, while IHL applies throughout the territory of a State that is a party to an armed conflict, measures derogating from the ECHR may be strictly required only in certain areas. In other areas, the ECHR may apply with full force. In the latter areas, acts—including intentional killing—may violate the ECHR even if they do not violate IHL. Similarly, under the ECHR, a “shoot to kill” policy may be strictly required by the exigencies of some circumstances (such as pitched battle) but not others (such as an ambush of unarmed adversaries).
Finally, the displacement view has been explicitly rejected by the International Court of Justice (see here), by the Inter-American Court of Human Rights (see here), and by the African Commission on Human and Peoples’ Rights (see here).
The related view that, during armed conflict, IHRL must be interpreted to coincide with IHL deviates from the general principle that the interpretation of one treaty provision or customary rule need only “take into account … [other] relevant rules of international law.” On this approach, the interpretation of IHRL should be informed but not determined by IHL. In many cases, IHL may prove a reliable guide to which deprivations of life or liberty should be considered “arbitrary” during armed conflict and which measures are strictly required by the exigent circumstances that typically arise in armed conflict. In other cases, IHL may prove an unreliable guide. Crucially, in all cases, whether human rights have been respected or violated remains a question of IHRL.
The view that IHL determines the content of IHRL during armed conflict is largely rooted in an unfortunate passage in the ICJ’s Nuclear Weapons Advisory Opinion, which states that
In principle, the right not arbitrarily to be deprived of one’s life applies also in hostilities. The test of what is an arbitrary deprivation of life, however, then falls to be determined by the applicable lex specialis, namely, the law applicable in armed conflict which is designed to regulate the conduct of hostilities.
This passage suggests that, during armed conflict, a deprivation of life is arbitrary under IHRL if and only if it violates IHL.
On this view, the content of IHRL depends on the content of IHL whatever the content of IHL happens to be. Let me try to illustrate the illogic of this view. The ICCPR entered into force in 1976. Additional Protocols I and II entered into force in 1978. Suppose, for the sake of the illustration, that Protocol I and Protocol II exhausted the customary law of targeting and attack in international armed conflict and non-international armed conflict, respectively, at the time of their entry into force.
On these assumptions, it would follow that killings in non-international armed conflict that would not violate Protocol II would not be arbitrary under the ICCPR. For example, if a fighter does nothing to verify that a person is a lawful target before killing her, or neglects feasible precautions that would avoid incidental harm, or inflicts excessive incidental harm, or uses a civilian as a human shield, then that fighter does not violate Protocol II. On this view, it necessarily follows that that fighter does not arbitrarily deprive his victims of their lives. This logic seems impossible to accept. The content of IHRL cannot depend on the content of IHL in this way.
Next, recall that Protocol I prohibits many killings that Protocol II does not. On this view, it follows that, when Protocols I and II entered into force, a wide variety of killings would be arbitrary deprivations of life if committed in international armed conflict but not if committed in non-international armed conflict. This logic seems incompatible with the very idea of a single, unitary human right to life. The content of the right to life cannot depend on conflict classification in this way.
Of course, IHL has evolved since 1978, particularly in non-international armed conflict. For a contemporary example of killing in armed conflict that may fail to respect human rights without violating IHL, consider this recent statement by the African Commission on Human and Peoples’ Rights:
Where military necessity does not require parties to an armed conflict to use lethal force in achieving a legitimate military objective against otherwise lawful targets, 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.
As the Commission observes, IHL does not exhaust the content of the right to life in armed conflict, which may be disrespected by killings that do not violate IHL. Nevertheless, IHL has an important role to play, because “[a]ny violation of international humanitarian law resulting in death, including war crimes, will be an arbitrary deprivation of life.” On this view, every killing that violates IHL is arbitrary, though not every arbitrary killing violates IHL.
During armed conflict, the specific rules of IHL provide more detailed guidance to fighters than the general prohibition of arbitrary killing in IHRL. Nevertheless, the general prohibition retains independent content. By way of analogy, the specific rules of driving (speed limits, turn signals, etc) provide more detailed guidance to drivers than the general prohibition of reckless driving. Nevertheless, driving that violates no specific rule may still be reckless.
In my view, the relevance of IHL to the interpretation of IHRL depends on the content of IHL, specifically on how well IHL protects individuals from arbitrary killing in armed conflict. Fortunately, in its latest statement on the relationship between IHL and IHRL, the ICJ did not mention lex specialis at all, but instead simply observed that, during armed conflict, “both branches of international law, namely international human rights law and international humanitarian law, would have to be taken into consideration.” Let us hope that the ICJ is returning to ordinary principles of legal interpretation, according to which IHL would be relevant to—but hardly dispositive of—the interpretation of IHRL.
Application of IHRL in Armed Conflict
Of course, one cannot apply IHRL during armed conflict if one adopts a view of IHRL that does not respond to the factual circumstances of its application. At one point, Jonathan writes that
Under IHRL, force should be applied in a graduated manner and lethal force can be used only as a means of last resort and only when necessary in the face of death or serious risk of injury.
Similarly, Jonathan writes that IHRL prohibits State agents from using force “if it knows with certainty that it will kill bystanders in the process.”
In most circumstances, IHRL would indeed place such constraints on the use of force by State agents. However, as we have seen, in times of war or other public emergency States may take measures derogating from such constraints to the extent strictly required by the exigencies of the situation. Similarly, whether or not a deprivation of life should be considered arbitrary depends on the factual circumstances of that deprivation.
For example, if a group of shooters is firing automatic weapons into very large crowds of people, then it may not be arbitrary for police to fire at the shooters even if the police will certainly miss the shooters at least a few times and thereby kill an innocent bystander.
Fortunately, such exigent circumstances rarely arise in ordinary life. Unfortunately, such exigent circumstances regularly arise during armed conflict. Accordingly, intentionally killing enemy fighters and incidentally harming civilians will prove arbitrary in some circumstances but not in others.
When, exactly, is intentional killing and incidental harm consistent with human rights? That is too large a topic for a blog post. (Indeed, it is one of the main topics of my forthcoming book.) However, we can all imagine circumstances during armed conflict in which such conduct would be justified rather than arbitrary, or arbitrary rather than justified.
It is the sensitivity of IHRL to the factual circumstances of its application that allows it to apply alongside IHL during armed conflict, sometimes offering greater protection than IHL but never offering less protection than IHL. Importantly, it is not the application of IHL but the concrete factual circumstances that determine what IHRL permits, forbids, or requires.
In my next post, I will address the mistaken view that IHL provides affirmative legal authority for States to use deadly force, including deadly force prohibited by other applicable legal rules. I will then return to the topic with which this exchange began, namely the application of IHL—and the parallel application of IHRL—to first strikes by States against armed groups.