Codifying the Unconventional

The Swiss government’s January 2003 roundtable to discuss the Geneva Conventions and 1977 Protocols in light of the so-called war on terrorism is now the talk among experts in international humanitarian law. To some academics and NGOs, the discussions provide a long-overdue opportunity to consider updating the Conventions and Protocols – to tailor them for a war they were never designed to address. In that sense, a new round of codification would be as important a response to terrorism as was the ICRC’s diplomatic conference of the 1970s a response to anti-colonial wars and internal conflicts.

Although governments and other experts should ensure that the employment of coercion against terrorists is regulated according to some normative framework, I believe that most claims for a major reform of international humanitarian law are premised on a variety of misconceptions of that law. As a result, any intergovernmental process is not likely to lead to any significant new norms nor, I believe, should it. In this essay, I will suggest four misconceptions implicit in calls for major reform, and their implications for efforts to augment the Conventions and Protocols.

Misconception 1: International Humanitarian Law has a Major Gap Regarding the War on Terror.

A common assumption in favor of updating the law of war is that the existing corpus of international humanitarian law is ill-equipped to address uses of force by and against terrorist groups. While the laws of war provide detailed rules for interstate conflicts, conflicts between states and liberation movements, and conflicts between states and well organized insurgencies, they do not provide any guidance to those fighting in wars between states and terrorist movements. It is, of course, undeniable that the Conventions and Protocols only cover what they cover. If one side in the conflict does not meet the definitions provided in Common Article 2 of the Geneva Conventions, Article 1(4) of Protocol I, and Article 1 of Protocol II, the Conventions and Protocols qua treaties simply do not govern the conflict.

But international humanitarian law has a sizeable "place-holder" for all other conflicts – one that Protocol I explicitly recognizes: "the principles of international law derived from established custom, from the principles of humanity and from the dictates of public conscience." These terms are no mere rhetoric. At a minimum, their content includes: (a) the principle of distinction, i.e., that combatants must distinguish between other combatants and civilians, and that combatants must neither deliberately target nor indiscriminately or disproportionately harm civilians; (b) the status of hors de combat, under which combatants not taking part in hostilities shall be treated humanely; (c) limitations on methods of combat to those that do not cause unnecessary suffering. These basic notions are admittedly vague around the margins. But states have made clear that they recognize such obligations in all conflicts. Thus, while at first denying, and later limiting, the application of the Geneva Conventions to its operations against Afghanistan and Al Qaeda, the United States and its allies have repeatedly stated their acceptance of basic principles of humanitarian law.

Indeed, the full scope of customary international humanitarian law is far more detailed than these basic principles. If asked, most states would point to numerous provisions of Hague and Geneva Law as still binding on this conflict, even though it does not meet the definition of armed conflict in those instruments. For instance, the ban on killing someone in the course of a bona fide surrender, as well as the possibility to shoot those engaging in perfidy by faking surrender (with some gray areas as to how the bona fides of the surrender is determined), would seem well accepted.

This is not to suggest that every scenario in the "war on terrorism" is addressed by customary international law. That, after all, is why states draft treaties. Ambiguities over the definition of combatant, the meaning of hostilities (and the end of them), and other problems will remain beyond the scope of existing customary international law simply because customary law has not had time to develop regarding certain aspects of these new conflicts. The question, however, is whether those ambiguities can and should be resolved through a new round of codification, or whether other methods are open to decisionmakers.

The burden needs to be on those calling for revision to demonstrate exactly what needs to be improved. Are they asserting that, under current law, combatants are underprotected, as when members of so-called terrorist groups are taken prisoner and not granted POW status; or that they are overprotected, as when the U.S. army targets and kills an Al Qaeda leader in Yemen and claims this action is a legitimate engagement in an armed conflict? Are civilians currently underprotected, as when they are killed in collapsing buildings; or are they said to be overprotected, because many members of terrorist cells, e.g., those organizing or financing activities, never wear uniforms or ever show arms? Even now, while the United States denies the applicability of the POW provisions of Geneva Convention III to Taliban detainees (almost certainly unjustified under that Convention) or Al Qaeda detainees (far more convincing under extant law), the ICRC is able to monitor their conditions at a number of detention facilities.

More likely, the harms arise from the prospect that states, such as the United States, may be claiming the right to kill terrorists (as the U.S. defines them) anywhere or detain them indefinitely. This is a significant expansion of the notion of armed conflict, and eats away at much of the traditional ways of addressing transnational crime, e.g., through extradition and trial. But if that is the case, those inside and outside of governments should focus on those narrow issues, rather than assume that international humanitarian law has little to say about such conflicts.

Misconception 2: Jus in Bello Can and Should be Divorced from Jus Ad Bellum.

One of the hallmarks of IHL since Henri Dunant has been the need to protect victims of armed conflict regardless of its cause or the blameworthiness of the sides under the law governing recourse to force (jus ad bellum). Today, the Conventions and Protocols apply equally to entities acting unlawfully – in violation of the UN Charter’s ban on the use of force -- and those acting lawfully -- in self-defense, under UN authority, or in opposition to colonial domination (or even against an unjust regime, although international law tends to be rather agnostic on this issue). Without such a clear distinction, combatants would argue that the justness of their cause allows for all sorts of indignities against combatants and civilians, clearly defeating the whole enterprise of IHL. With regard to terrorism, the argument would be that the question of who is right and who is wrong in the decisions by certain states and non-state groups to use violence against each other should not detract from the need to regulate that conflict in a humanitarian manner.

The jus in bello/jus ad bellum distinction is thus premised on the idea -- based on fundamental principles of humanity -- that all combatants and civilians deserve protections regardless of the merits of their side. In addition, international actors may well have other reasons for granting both sides protection. They may understand the difficulty of knowing (or at least reaching a consensus among states) in certain conflicts which side has violated jus ad bellum by initiating the war; in these cases, it is better to grant both sides protection.

But the reality is not a clear-cut as this distinction suggests. First, the inclusion of wars of liberation in Protocol I’s coverage stemmed from the sense among many governments that certain struggles against colonial or alien domination were legitimate; and even Protocol II’s protections recognize that insurgent groups might be engaging a legitimate activity. Second, even for wars that are illegal, i.e., clear aggression under the UN Charter, states seem to have found these not so obnoxious to merit denying even the offenders humanitarian protections. States quietly accept that wars still happen; that aggression still occurs; and that they might, after all, even be aggressors themselves one day. They are not willing to say that the aggressor is so evil – that his goal is so beyond the pale of civilized conduct -- that he forfeits all protections for his troops and his civilians.

But what happens when we have non-state actors whose goal is simply to kill innocent people and terrorize a population? While not all groups labeled as terrorist by states have this goal – many use attacks on civilians as a means to gain control of specific territory (e.g., the IRA or Palestinian groups) -- there seems to be no such goal for Al Qaeda. It has no leaders in waiting to take over the United States or the United Kingdom. It claims to want those states to change their foreign policies on various issues, but there is also evidence that it is simply the way of life practiced in these states that it finds a threat. In these situations, can and should governments make the leap of faith that such conflict against organized states deserves regulation by detailed protections of international humanitarian law? When the goal of a group is so beyond acceptable conduct that it finds no defenders among governments, extending the protections of international humanitarian law to such conflicts serves a legitimating function. Equally significant, key governments likely have the same fear, and are unlikely to engage in a process that they will regard as tarnishing IHL.

Misconception 3: The Non-Reciprocal Nature of IHL Demands Protection for Even Those Entities That Insist on Violating it.

Another mantra of international humanitarian law is that, unlike much international law, the targets of its norms deserve protection even if they themselves violate them. Thus, traditional norms of treaty interpretation that permit a state to suspend its obligations in the event of a material breach by the other side do not apply in many key situations. At a minimum, reprisals – i.e., otherwise unlawful acts taken in response to prior unlawful acts -- against POWs, civilians, and wounded, sick, and shipwrecked combatants are explicitly prohibited by Geneva Convention III (article 13), Geneva Convention IV (article 33), and Protocol I (articles 20 and 51(6)). Some might call for the abolition of all reprisals based on the underlying purposes of IHL. Another form of non-reciprocity appears in Protocol I’s requirement (article 44(2)) that combatants do not lose their status even if they violate most norms of IHL in the course of their operations. Under this non-reciprocity model, the violations of IHL by various terrorist groups should not be an excuse for denying them certain protections.

But non-reciprocity is not and should not be all-encompassing. First, current IHL does not preclude reprisals during combat against combatants that might violate IHL. Second, even the bans on reprisals in Protocol I have their detractors, such as the United Kingdom, which issued a reservation to that treaty allowing for the possibility of measured reprisals against civilians if the opposing party itself engaged in serious, deliberate attacks on civilians. Lastly, the granting of protections in Protocol I was, in fact, part of a reciprocal bargain, not simply extending protections to guerrilla groups as a gesture of goodwill, but creating obligations for those movements as well. Thus Protocol I denies combatant status to guerrilla groups that do not follow certain requirements regarding open carrying of arms.

The nature of terrorist organizations – whose modus operandi typically includes targeting of civilians (although it is not limited to that) – pushes the need for non-reciprocity even further. One side is determined from the outset to carry out a struggle regardless of even the most fundamental principles of humanity. If these acts are not simply an aberration but its principal way of operation, why should its members be afforded anything more than treatment consistent with those basic notions of humanity? (Some will suggest that they do not even deserve that treatment, of course.) States will not and should not tolerate a legal regime whereby only one set of combatants benefits from the protections of IHL.

This solution is not a recipe for a free-for-all in the war against terrorism. As noted, basic principles of humanity still apply and are accepted by governments. They would mean, for example, that terrorists cannot be tortured upon capture and that their families cannot be targeted. But to suggest that the IHL’s non-reciprocal approach to protections requires granting them a vast array of other protections, such as the combatant’s privilege (against prosecution) or POW status, is unwarranted.

Misconception 4: The Paradigms of Combat and Combatants are Out of Date.

One argument made for the revision of the Conventions and Protocols is that they reflect an outdated notion of what constitute armed conflict and those participating in it. Thus, Al Qaeda’s actions against the United States – whether against civilians or military personnel, whether within the United States or abroad – are clearly a form of armed conflict; and the U.S. actions in fighting, capturing, and killing Al Qaeda forces – whether in Afghanistan, Pakistan, Yemen, or elsewhere – are equally armed conflict. The argument is not that customary law does not offer some protections, but that the Conventions and Protocols, which on their face govern only three sets of conflicts (state vs. state, state vs. national liberation movement, and state vs. organized insurgency), need to respond to the changing nature of armed conflict.

This apparent need for updating seems bolstered by the very use of the war paradigm by states engaging with Al Qaeda. Thus, the United States and its allies have used the rhetoric of armed conflict to respond to the attacks of September 11. U.S. officials refer to their operations as the "war" on terrorism; and the United States has invoked its rights under jus ad bellum – Article 51 of the United Nations Charter – in attacking Afghanistan and Al Qaeda targets and personnel around the world. The U.S. relies on this characterization to avoid treating Al Qaeda as simply a criminal organization that can be targeted only through traditional law enforcement activities, e.g., police investigations, extradition requests, and civilian trials with full due process. Advocates for change say that the U.S. is trying to have it both ways – asserting various rights under jus ad bellum while denying the applicability of key aspects of jus in bello.

It seems unquestionable that the United States and others are engaged in an armed conflict with Al Qaeda. But the subsidiary concepts of combat operations and combatants that permeate Hague and Geneva Law are not and should not be infinitely elastic. Part of the core of international humanitarian law is the creation of physical and temporal space in which it is perfectly legal for certain categories of people to kill each other: that space is combat and those people are combatants. The combatant’s privilege means that combatants (at least in international conflicts) may not be punished for lawful combat operations, though, of course, they can be punished for war crimes. The notions of armed conflict, combat, and combatant have changed since the days of the Hague Conventions, as Protocol I makes especially clear; today, the problems of determining whether the unconventional nature of some armed forces’ garb (like the Taliban) serves to deny them combatant status are well known.

Yet, even with the expansion of the notion of armed conflict to cover acts by national liberation movements in Protocol I, there remain the ideas of the military engagement and the military attack. For instance, under Protocol I, if a fighter does not carry his arms openly during the engagements as well as during the deployment before an attack while visible to the adversary, he is not a lawful combatant (entitled, for example, to POW status). This compromise (though opposed by the United States) preserves the idea of combat operations and the special nature of the combatant, who must distinguish himself from the civilian population in combat and the time leading up to it. The ban on perfidy also reinforces this idea.

To expand the laws of wars to apply to any situation where an organization (or conceivably an individual) initiates force is to blur the distinction between cases where the law allows individuals to kill each other and those where the law prohibits it. Under such a view, every attack on a military installation, even if undertaken without any separation from the civilian population (indeed, this is the modus operandi of many terrorist operatives) is per se a combat operation and those who carry it out combatants. The result is to turn every act of violence into an act of war, and all those who commit it into lawful combatants who enjoy the combatant’s privilege. It infinitely expands the protected zone in time and space. Where does one draw the line between Al Qaeda attacks and those of the mafia or simply an insane person? For the combatant’s privilege to remain, as it should, a hallmark of international humanitarian law, it must be confined to a highly limited set of circumstances. States will not and should not agree to extend it to any individual or group that chooses to attack a military target.

Priorities for Normative Development

Of course, as noted, the United States government conceives of the struggle against Al Qaeda as a war, and it wants to expand the geographical zone of legitimate combat significantly – to cover, for instance, the killing by remotely piloted vehicle of a suspected Al Qaeda leader in Yemen. The United States also wants to expand the notion of combat temporally. It refuses to talk about an end to the hostilities, partly to justify the indefinite detention without trial of Al Qaeda and Taliban personnel.

These two expansions of the notions of armed conflict, combat, and combatant – geographical and temporal – are the most vexing questions for IHL today. On the geographical plane, it is hard to accept the U.S. position that treats Al Qaeda operatives around the world as legitimate targets for wartime killing while rejecting the idea that Al Qaeda operatives are acting as lawful combatants when they kill U.S. soldiers anywhere in the world. The U.S. squares the circle by relying on the notion of the "illegal combatant," something that IHL already accepts in terms of spies, mercenaries, and guerrillas who do not meet the special requirements of carrying arms openly. This allows a state to kill them but afford them no protections. But is there no limit to how and where someone becomes such an illegal combatant?

The simplest solution is to say that the armed conflict paradigm does not apply at all – that Al Qaeda’s members are criminals and that the United States should use law enforcement techniques to try and punish them, just like European states are rounding up suspected members of Al Qaeda cells. Human rights law, which governs peacetime law enforcement, still recognizes significant discretion for police acting in genuine self-defense, while protecting criminals from arbitrary killing by the state. The question is whether this attitude will suffice in a world where Al Qaeda can gain access to weapons of mass destruction. If states treat the campaign against Al Qaeda as an armed conflict, then they need to do so in a way that does not entail a wholesale change in the notion of combat. The conflict with Al Qaeda needs to have boundaries beyond which the special privileges that the law of war gives to combatants do not apply.

As for the temporal question, namely the U.S. willingness to tolerate indefinite detention of Al Qaeda fighters, it is clearly unsustainable in the long term. (As noted, their position on the Taliban seems to be a misreading of the Conventions.) Humanitarian and human rights law make clear that individuals – whether lawful combatants, civilians, or others -- cannot be held indefinitely without trial. But rather than recognizing some form of combatant status for Al Qaeda and then trying to determine at what point hostilities cease (so that they would have to be released), another possibility is available. They should enjoy protections consistent at a minimum with basic principles of humanity. Beyond that, states might apply the list of minimal protections found in Article 75 of Protocol I to any captured terrorist. This list currently applies to anyone in the control of a party to an interstate conflict who does not enjoy better treatment under other parts of the Convention and Protocol. These include the basic human right to a trial for suspected crimes.

In summary, as a predictive matter, it seems exceedingly unlikely that states will agree to a significant revision or augmentation of the laws of war to address the sorts of unconventional threats posed by transnational groups such as Al Qaeda. To date, only a very small part of those activities that some states refer to as terrorism – guerrilla warfare as part of a legitimate war of national liberation – has come under the protection of international humanitarian treaties. The remainder of unconventional activity is still subject to baseline principles of humanity, and states see little reason to grant more. Protocol I was expanded to cover liberation movements because a group of states – those in the developing world – insisted upon such protections. Al Qaeda and other groups have no such vocal constituency, however, as much as some states may support them sub silencio. If, for some reason, enough states were to find the absence of treaty law in the war on terrorism an unacceptable gap, a Protocol III is likely to be far more limited than Protocol II turned out to be. Indeed, I suspect it will do nothing beyond recognizing the most basic principles of humanity.

As a normative matter, any codification that goes beyond the obvious principles of humanity is at best premature. The legitimization function accomplished through codification sends a signal to these unconventional fighters that their tactics are acceptable – that they are lawful combatants, even if they consider their great victories to be crashing a plane into a commercial building or incapacitating a major city via germ or chemical weapons. The immediate threat to public order around the world from groups whose mission is to create civilian casualties – who themselves reject the most basic principles of the law of war – argues against enveloping their activities in a set of detailed law of war norms. Any such process will do little to protect the victims of such acts, as those people are already illegitimate targets under international criminal law, humanitarian law, and human rights law. So it seems to benefit only one side in the conflict. Whatever protections those fighting colonial domination may have deserved, Al Qaeda and its allies fall into a completely different category.

Clearly, states do need to figure out some limits to armed conflict as a geographic and temporal matter, and academic and governmental discussion of such issues can help elaborate whether the existing norms work or need some further elaboration. But beyond that, as long as those fighting terrorism respect the basic principles of humanity in that struggle, international humanitarian law ought to live with something close to the status quo.

Back to Top

Rethinking the Geneva Conventions
January 30, 2003