Remarks of Professor John Cerone at the 31st San Remo Round Table on Current Problems of International Humanitarian Law

Topic: Peace operations and the complementarity of human rights law and international humanitarian law


I. Introduction

On February 3, 2000, chaos erupted in Mitrovica, Kosovo. Following the bombing of a local café, mass civil unrest erupted in the city. The UN Police and KFOR, the NATO peacekeepers deployed in Kosovo, were completely unprepared as mobs rampaged through the streets.

A number of ethnically-motivated attacks were carried out in the course of the rioting. Unidentified perpetrators threw grenades into homes and gunned down individuals attempting to flee. Many other homes and vehicles were torched. At least eight people were killed that night and dozens more were severely injured. While there was a great deal of chaos, with several unruly mobs roaming the streets, the door-to-door killings were carried out systematically, evincing a planned attack.

Where was KFOR? When the fighting broke out, those soldiers who were present at the scene withdrew to their base and provided no assistance to the UN Police who were trying to extract the wounded and vulnerable. No KFOR reinforcements arrived.

As a human rights officer working under the auspices of the UN Interim Administration Mission in Kosovo, I was confronted with a number of legal issues relating to the violence. One particularly difficult question was whether KFOR had a duty to protect individuals from violent acts committed by third parties. Underlying this question were a number of complex legal issues, including the interaction of human rights law and humanitarian law in a peacekeeping context, the question of responsibility in the context of multilateral operations, and the extent to which human rights obligations applied beyond a state’s sovereign territory.

This paper focuses on the analysis of these issues in the context of obligations of troop contributing or sending states.


II. Simultaneous Application of Human Rights Law & Humanitarian Law

For much of the Twentieth Century it remained unclear whether human rights law would apply to a state’s conduct during armed conflict or occupation. Despite continuing objections on the part of a handful of states, a consensus is evolving in favor of the view that human rights law applies alongside humanitarian law in times of armed conflict or occupation.

As stated by the International Court of Justice (ICJ) in its 1996 Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons: “The Court observes that the protection of the International Covenant of Civil and Political Rights does not cease in times of war, except by operation of Article 4 of the Covenant whereby certain provisions may be derogated from in a time of national emergency.” In a subsequent opinion, the ICJ noted further: “[T]here are thus three possible situations: some rights may be exclusively matters of international humanitarian law; others may be exclusively matters of human rights law; yet others may be matters of both these branches of international law.”

While the jurisprudence of the ICJ and other international bodies has shed some light on this relationship, much remains to be clarified in terms of how this complementarity is to be operationalized. However, for present purposes, it suffices to note that it is now increasingly clear that human rights law does not cease to apply by reason of the inception of state of armed conflict.


III. Attribution of Conduct in the Context of Multilateral Operations

Notwithstanding the significant expansion of international law in the past century, the principal subject of international law remains the state. Human rights treaties, such as the ICCPR, place responsibility for ‘respecting and ensuring’ human rights squarely upon states parties. Thus, only conduct attributable to the state can constitute an internationally wrongful act under these treaties, and only the state can be held responsible on the international plane for such violations. While certain norms of humanitarian law are now regarded as being addressed to individuals, this remains exceptional. While the breach of any rule of humanitarian law may give rise to state responsibility, individual responsibility arises only in response to certain serious breaches.

At the same time, states are abstract entities, incapable of acting as such. The conduct of states is the conduct of individuals whose acts or omissions are attributable to the state. As such, the issue of attribution must be addressed.

As an initial matter, it is important to bear in mind that the question of whether an actor’s conduct is attributable to a state is analytically distinct from the question of whether that conduct is internationally wrongful. The rules of attribution form part of the law of state responsibility. These rules are of a framework nature and are thus unconcerned with the separate question of whether the conduct at issue conforms to what is required by substantive norms of international law.

The first rule of attribution is that the conduct of an organ of a state, including that of any individual who is an official part of the machinery of the state, or of an entity legally empowered by a state to exercise elements of governmental authority is considered to be an act of that state. This would also include situations in which an organ is placed at the disposal of a state by another state and the organ is acting in the exercise of elements of the governmental authority of the former state. The conduct of such actors is attributable to the state even where an actor’s conduct is ultra vires, or beyond the scope of his or her authority, so long as he or she was acting in an official capacity.

Attribution in the Context of Collective Action

While the lines of responsibility are relatively clear when states act in an individual capacity, the issue of attribution becomes more complex in the context of collective action, particularly in light of the range of circumstances in which states may conduct collective operations.

States may simply deploy military forces jointly or through “coalitions of the willing”, which may or may not have separate legal personality. They may also contribute troops to UN or NATO forces in which operations are under the command and control of these organizations. Or they may deploy forces together with other states acting pursuant to a UN mandate, while retaining command and control. In these situations, chains of command may or may not be unified, states may or may not retain control over their contributed troops, and the lines of attribution may be muddled as a result.

Given this complex array of possibilities, the issue of attribution must be assessed in light of the particular features of each operation. In general, the conduct of a state’s military forces will be attributable to that state while those forces are acting in their national capacity. However, if troops are fully seconded to an intergovernmental organization, or another entity with separate international legal personality, such that they are acting on behalf of that organization or entity and are no longer acting on behalf of their state of nationality, then their conduct may no longer be attributable to their state of nationality.

In reality, the sending states of troops contributed to UN or regional peace-keeping operations retain a significant degree of control over their troops. In such situations, the precise scope of the troops’ national capacity versus their intergovernmental peacekeeping capacity may be difficult to delineate. Indeed, it may be possible that the troops are operating in both capacities simultaneously, in which case their conduct may be attributable to their sending state as well as to the intergovernmental organization through which they have been deployed.

This latter approach appears to be embraced by the rules of attribution set forth in the ILC’s Draft Articles on the Responsibility of International Organizations. Article 5 contemplates situations in which seconded personnel are under the effective, though not necessarily exclusive, control of an international organization. In such cases their conduct is attributable to the organization. Article 5 does not comtemplate full secondment, in which situation the seconded personnel would be under the exclusive control of the organization. Such situations are governed by Article 4 of the Draft Articles, dealing with agents of the organization. The ILC commentary indicates that Article 5 applies in situations where the conduct of the contributed personnel would likely remain attributable to the sending state, notwithstanding the simultaneous attribution of the same conduct to the organization. Assuming the minimum level of control is met to satisfy Article 5, the relative degrees of control, as between state and international organization, do not go to the question of attribution, but to the apportionment of responsibility (e.g. proportion of compensation to be paid by each).

Attribution of the Conduct of Non-state Actors

The conduct of non-state actors may also be attributed to a state under certain circumstances. The conduct of a non-state actor may be imputed to a state when the actor is in fact acting on the instructions of, or under the direction or control of, a state in carrying out the conduct; when the actor is exercising elements of governmental authority in the absence or default of official authorities; when the conduct is subsequently adopted by a state; or when the conduct is that of an insurrectional movement that becomes the new government of a state.

These standards establish a fairly high threshold of state involvement or, alternatively, de facto state action by non-state actors accompanied by state authorization or disengagement. Instances of simple complicity of state organs in the conduct of non-state actors are not sufficient to render such conduct attributable to the state under the traditional rules of attribution.

However, the law of state responsibility admits the possibility of lex specialis where ‘special rules of international law’ may govern. Special rules may be evolving through the practice of universal and regional human rights mechanisms. These institutions have increasingly found degrees of state involvement not rising to the level established for attribution under the Articles to be sufficient to render the state responsible for the acts of non-state actors. Indeed, a growing corpus of international human rights jurisprudence and practice supports the proposition that the conduct of non-state actors may be attributable to the state where state actors are complicit in such conduct.

Caveat: Positive Obligations and the Attribution of Omission

As noted above, the question of attribution is in principle separate from the content of international obligations. However, this distinction may become difficult to discern in the context of a failure of a state to fulfill positive obligations in relation to the acts of non-state actors. In such situations, it is essential to distinguish between whether the conduct of non-state actors is attributable to a state and the separate question of whether a state has failed to fulfill an affirmative obligation, should one be imposed by a primary rule of international law, in relation to the conduct of non-state actors.

The attribution of conduct consisting of omissions presents conceptual difficulties in part because conduct consisting of omissions is, in a sense, always attributable. As omission is a lack of action, an actor is not required. Hence, the state is essentially in a constant state of omission. However, in order for an omission to constitute a basis of responsibility, there must be a duty to act. The question of establishing a duty to act will turn on the content of the relevant primary rule. Thus, in these circumstances, the issue of attribution collapses into the content of the primary rule.

The Behrami Judgment

The Behrami judgment of the European Court of Human Rights misapplies these rules in several ways. Most significantly, the Court concluded that because it had found the conduct at issue attributable to the UN, it could not be attributable to the sending state. It thus failed to recognize that the same conduct may be attributable both to an international organization and to a sending state. Another defect in the Court’s approach is its formulating the primary issue as one of attribution instead of addressing the antecedent question of the existence of a positive obligation on the part of the Respondent State operating in an extraterritorial context.


IV. Extraterritorial Application

International Humanitarian Law

Unlike human rights law, the law of armed conflict was designed to apply primarily in an inter-state context. Thus, the vast majority of its provisions would clearly apply to a state’s extraterritorial conduct, specifically in the territory of the opposing state.

However, peace operations usually do not entail armed conflicts with states. Typically, where peacekeepers are engaged in armed conflict, the opposing party or parties are non-state organized, armed groups. Such conflicts are thus not governed by the law of international (i.e., interstate) armed conflict. But do they fall within the scope of the rules of humanitarian law that were developed to regulate non-international (i.e., non-inter-state) armed conflict, the central case of which would be internal armed conflict?

Recently, controversy arose as to whether Common Article 3 applied only to internal conflicts. Until recently, the US Government had taken the position that common article 3 could not apply to its conflict with Al-Qaeda, as this conflict was transnational in nature. It focused on the phrase “occurring in the territory of one of the High Contracting Parties,” arguing that the plain meaning of this language would limit the application of Common Article 3 to internal conflicts.

This was, of course, contrary to the position of the International Court of Justice that the rules of common article 3 “constitute a minimum yardstick, in addition to the more elaborate rules which are also to apply to international conflicts; and they are rules which, in the Court’s opinion, reflect what the Court in 1949 called ‘elementary considerations of humanity.’”

Since the US Supreme Court ruling in Hamdan v. Rumsfeld, the US Government has accepted application of common article 3 in transnational conflicts. There now seems to be a general consensus supporting the proposition that the law of non-international armed conflict applies extraterritorially.

As noted above, demonstrating the applicability of humanitarian law outside of a state’s territory is facilitated by the fact that the bulk of the law of armed conflict was designed to apply in an interstate context, presupposing that states would be acting on each other’s territory. That some of these rules are now deemed to apply even in an internal setting does not lessen the presumption that they will still apply extraterritorially, at least insofar as they consist of prohibitions and do not purport to impose obligations on third states.

The situation is more complex under human rights law, which was not primarily designed to apply extraterritorially

International Human Rights Law

Most of the jurisprudence of human rights bodies, which have greatly elaborated on the content of states' obligations under various human rights treaties, has been developed in the context of alleged violations committed on the territory of the respective state party. Can these same standards be transposed onto the state's conduct abroad?

When I first began examining this question, I was working in the UNMIK Regional Center in Mitrovica, Kosovo. Though I was fortunate to have a copy of Professor Meron’s article on the subject, accessing relevant judicial decisions was extremely difficult. In any event, there was very little to be found. The few international decisions that had analyszed this issue had employed an approach that was unsatisfactory on a number of levels. Since that time, a number of other decisions have been handed down, though still lacking a coherent approach.

In an effort to bring order to an otherwise chaotic array of judicial (and quasi-judicial)
decisions, I developed a framework for delineating the scope of human rights obligations by examining three different parameters: the scope of beneficiaries, the range of rights applicable, and the level of obligation. Structuring an analysis of current jurisprudence around these three parameters reveals a trend toward recognizing varying levels of obligation.

The scope of beneficiaries refers to the range of individuals in relation to whom the state has any obligation under the relevant human rights treaty. States parties to the ICCPR are not bound to respect and ensure the rights of all individuals everywhere. For example, it is clear that, absent special circumstances, states parties are not required to protect the rights of individuals living in other countries from violations perpetrated by the governments of those countries or by non-state actors operating there.

A common feature of the major human rights treaties is that the scope of beneficiaries is typically limited to those individuals who are subject to the state party’s jurisdiction. While it was initially unclear whether this language could encompass individuals situated outside of a state’s territory, the extraterritorial application of human rights treaties has now been clearly established in the jurisprudence of several international judicial and quasi-judicial bodies.

The ICJ has held that when a state party is operating outside of its territory, “international human rights instruments are applicable ‘in respect of acts done by a State in the exercise of its jurisdiction.’” The regional human rights institutions have endorsed the position of the Human Rights Committee that a State may be responsible for violation of the rights and freedoms of persons who are in the territory of another State but who are found to be under the former State's authority and control through its agents operating in the latter State.

Under human rights treaties, the range of rights applicable within a state’s territory will normally be the full range of rights set forth in each treaty. However, this may not be the case when the state is operating abroad. In such situations, the range of applicable rights may be limited by the scope of the state’s authority or control in the circumstances. In general, it may be reasoned that as human rights law is generally predicated on a state’s authority and presumed capacity to control individuals and territories, a state’s human rights obligations while acting abroad would not be as extensive as when it acts on its own territory. Similarly, it may be the case that the application of certain rights requires a higher threshold of control. Indeed, the ICJ and the regional human rights institutions have implied as much.

For example, it may be argued that certain rights cannot be applied where a state is performing a very narrow function in the territory of another state. Where a state is not trying individuals, the right to a fair trial is simply not implicated.

However, another approach is to focus the inquiry not on the question of which rights the state is obliged to secure, but instead on the level of obligation upon states with respect to those rights.

Varying Levels of Obligation

As noted above, the obligation to “respect and ensure” rights set forth in the ICCPR, or, in the words of the European Convention, to “secure” rights, entails a substantial degree of positive obligation.

As with the range of rights, the level of obligation also may be limited where the state operates abroad. The level of obligation may similarly be tied to the scope of a state’s extraterritorial activities or authority to act. In particular, it is arguable that human rights obligations requiring the adoption of affirmative measures may be more limited in an extraterritorial context.

This position finds support in the international jurisprudence referred to above. Seen collectively, the bulk of international authority indicates a trend toward recognizing varying levels of obligation.

In particular, the ever-lowering threshold for “jurisdiction” in the context of negative obligations seems to indicate that these obligations apply whenever a state acts extraterritorially (at least with respect to intentional human rights violations, as opposed to indirect consequences), but that the degree of positive obligations will be dependent upon the type and degree of control (or power or authority) exercised by the state.

Positive obligations are limited by a scope of reasonableness even when applied to a state’s conduct within its territory; there is no reason why application to a state’s extraterritorial conduct would not similarly be bounded by a scope of reasonableness, such that the adoption of affirmative measures is only required when and to the extent that the relevant party de jure or de facto enjoys a position of control that would make the adoption of such measures reasonable. This approach would preserve the integrity of the respective treaties and would vindicate the universal nature of human rights, which is proclaimed in the preambles of all of the human rights treaties considered in this analysis.

At the same time, it would not place unreasonable burdens on states parties. Even though this approach contemplates world-wide application of human rights law, it does so, absent a significant degree of control, only in respect of negative obligations. Thus, a state is bound only to the extent it chooses to project its power, and, even then, is bound simply to refrain from directly perpetrating human rights violations.

It is also possible that a negative obligation can flip into a positive obligation, but this again will depend on an assertion of authority by the state. For example, the state generally has an obligation not to arbitrarily detain. Once it begins detaining people, this converts to a positive obligation to create a regulatory procedure, to ensure humane treatment of the detainee, to provide compensation if the detention is wrongful, etc.
Another example is the negative obligation not to subject someone to an unfair trial. Again, this would not be implicated unless the state it trying people.

Textual Argument Limiting Application to a State Party’s “Territory”

The United States has consistently taken the position that the ICCPR applies only within the territory of the State Party. It bases this position on the text of Article 2 of the Covenant, which reads:

“Each State Party to the present Covenant undertakes to respect and to ensure to all individuals within its territory and subject to its jurisdiction the rights recognized in the present Covenant, without distinction of any kind…”

The US argues that the phrase “within its territory and subject to its jurisdiction,” if interpreted according to the generally accepted rules of treaty interpretation, must mean that States Parties to the ICCPR do not have any obligation to respect and ensure the rights of those who are outside of the their territory. This is contrary to the position taken by the Human Rights Committee and the ICJ that territory or jurisdiction is sufficient.

To further support its position the US points to the travaux preparatoires of the Covenant. During the negotiations, it was twice proposed that the phrase “within its territory” be deleted from the text, and both times the proposal was defeated. The US delegates consistently pointed out that it was essential to retain this phrase in order to avoid any obligation to ensure the rights of those outside of a state’s territory. This sentiment was echoed by several other delegations that supported the retention of the phrase.

While this argument appears persuasive on its surface, a closer read of the travaux reveals another possible interpretation. Every delegate that expressed concern about extraterritorial application of the Covenant limited their concern to the obligation to “ensure” rights. In other words, no delegate claimed that they were concerned about the extraterritorial application of the negative obligation to “respect” rights. This holds true even for the later negotiations, where the text already expressly included both the obligation to respect and the obligation to ensure.

Thus, the travaux equally support an interpretation that is consistent with the trend identified above. Indeed, a re-examination of the text demonstrates that the most reasonable interpretation of article 2, using the customary rules of treaty interpretation, is that the phrase “within its territory and subject to its jurisdiction” modifies only the obligation “to ensure”.

As such, article 2 essentially provides that Each State Party undertakes to respect the rights recognized in the Covenant and also to ensure these rights to all individuals within its territory and subject to its jurisdiction. This interpretation is consistent with the travaux and also with the ultimate decisions of the ICJ and regional human rights bodies, all of which have generally limited their findings of extraterritorial violations to breaches of negative obligations. The few instances where they have found violations of positive obligations were situations in which the State acting extraterritorially was occupying the territory, and thus had a large degree of authority and control over the territory, enabling the Court to assimilate that territory to the state’s own.

Negative & Positive Obligations in the Context of Collective Action

In light of the trend identified above, it may be unnecessary to examine the question of jurisdiction for negative obligations. The only relevant issue would be to which subject of obligation the human rights violative conduct is attributable.

For positive obligations, it would be necessary to establish “jurisdiction,” or, as this term is increasingly understood in a human rights context, control. In the context of collective action, it may be appropriate to consider whether individuals or territory are under the control of the relevant entity (e.g. peacekeeping mission) as a whole, and then to parse out responsibility among the participating states and member states.