Reprisals Against Enemy Civilians, the Object and Purpose of Additional Protocol I, and the United Kingdom’s Reservation

Editors’ note: This post considers one aspect of an article recently published by the author in The Military Law and the Law of War Review. It does not address the question of the legal effect and consequences of a determination that the United Kingdom’s reservation is invalid.

Belligerent reprisals—ostensibly unlawful acts resorted to in response to a prior violation of international humanitarian law (IHL), aimed at securing compliance with IHL—are controversial, brutal, and ineffective. One scholar described them as “a historically well-established mechanism for deterring violations of the law of armed conflict even as they provide an ostensible rationale for otherwise unthinkable atrocity.” Reprisals encourage an escalation of violence which negates any potential efficacy, and the victims tend not to be those responsible for the initial violation. Indeed, the UK Joint Service Manual of the Law of Armed Conflict accepts that reprisals are an “extreme” measure of coercion because “in most cases they inflict suffering upon innocent individuals.” This is especially true of reprisals against civilians.

Now largely discredited, IHL has progressively limited the use of reprisals. Initially prohibited against prisoners of war (POWs) in 1929, further prohibitions against reprisals were introduced in the 1949 Geneva Conventions and again in 1954. Efforts in the 1970s to revise and enhance IHL and, in particular, to strengthen the protection of civilians, led to the inclusion in Additional Protocol I (AP I) of a series of prohibitions so sweeping that reprisals remained possible only against military objectives. The focus of this post is Article 51(6) of AP I, which prohibits “[a]ttacks against the civilian population or civilians by way of reprisals . . . .”

The United Kingdom’s Reservation to AP I

Upon ratification of AP I, the United Kingdom lodged a reservation which included a claimed ability to target enemy civilians in response to similar violations by an adversary. According to the International Committee of the Red Cross’s (ICRC) 1987 Commentary to AP I, reservations to AP I are regulated by general international law as expressed in the 1969 Vienna Convention on the Law of Treaties (VCLT). Modelled on the International Court of Justice (ICJ)’s 1951 Advisory Opinion, Reservations to the Convention on the Prevention and Punishment of the Crime of Genocide, VCLT Article 19 provides that formulating a reservation is generally permissible unless:

(a) the reservation is prohibited by the treaty;

(b) the treaty provides that only specified reservations, which do not include the reservation in question, may be made; or

(c) in cases not failing under subparagraphs (a) or (b), the reservation is incompatible with the object and purpose of the treaty.

AP I does not prohibit reservations outright, nor does it prohibit reservations to specific articles, or any particular category of reservation. Article 19(a) and (b) are accordingly inapplicable in this context. Article 19(c) may apply, however. This provision seeks to accommodate the broadest possible participation within a treaty regime whilst protecting its integrity by preventing reserving States from undermining the treaty’s raison d’être. Two questions then arise. First, who is responsible for determining compatibility with the treaty’s object and purpose? And second, how is the assessment to be made? The VCLT is silent on both questions.

Who Determines Compatibility?

Prima facie, compatibility is a matter for the other States parties. The VCLT regime permits States to respond to reservations lodged by other parties, either by accepting the reservation (explicitly via Article 20(4), or implicitly, after twelve months, via Article 20(5)) or objecting to it. This subjective approach inevitably invites disagreemnent and the logically incoherent position whereby a reservation is valid between the reserving State and some parties, but invalid between the reserving State and others.

The ICJ assumed States would mitigate this problem, believing that States would be “guided in their judgment by the compatibility or incompatibility of the reservation with [its] object and purpose,” in seeking to preserve what is “essential to the object of the Convention.” Subsequent practice paints a gloomier picture, however. Whilst the ICJ still refuses to discount entirely their absence, objections are relatively rare. As ICJ Judge Abdul Koroma observed, “States are often remiss in . . . objecting to reservations,” even where they consider them invalid.

There are several possible reasons for this. For example, “many administrations are simply not equipped to keep under constant review reservations to multilateral conventions formulated by other States” (p. 63). Objecting may also destabilise relations between the objecting and reserving States. Ultimately, as Professor Goodman explains, “A state may find another state’s reservations objectionable, or even violative of the basic understanding of the treaty, but unless the former’s own domestic interests are directly implicated, monitoring, scrutinizing, and reacting to other states’ reservations are simply not priorities.” Whatever the reason, States are less than rigorous in objecting to problematic reservations.

It is true that no State has objected to the United Kingdom’s reservation, but silence does not necessarily imply approval. As the Human Rights Committee indicated in 1994, “The absence of protest by States cannot imply that a reservation is either compatible or incompatible with the object and purpose of [a treaty, and] . . . it is not safe to assume that a non-objecting State thinks that a particular reservation is acceptable.” Judge Koroma and the International Law Commission (ILC) agree. Parties to AP I may have opted not to draw attention to the United Kingdom’s reservation, but this does not mean that it is compatible with the object and purpose of the treaty.

This flows logically from the VCLT, which does not determine permissibility by the presence or absence of objections. The VCLT makes no link between Articles 19 and 20 (or 21) and does not suggest that impermissible reservations are open to acceptance. Instead, the VCLT specifies that States may not formulate reservations inconsistent with any element of Article 19, which operates in a different context and timeframe from Article 20. Article 19 regulates the formulation of a reservation while Article 20 regulates reactions to an established reservation. Given reservations falling foul of Article 19(a) and (b) are void, it follows that reservations that violate Article 19(c) are also void. To suggest otherwise, and that a prohibited reservation could nonetheless be accepted, would make Article 19(c) meaningless and grant the impermissible reservation precisely the same effect as a permissible reservation. As Bruno Simma and Gleider Hernández explain, this is “unacceptable, both in policy and logic.”

The absence of objections does not, then, mean that a reservation is compatible with the object and purpose of the treaty. Article 20(5) cannot circumvent this (VCLT Commentary; ILC Guide to Practice), and Article 19 must instead “be interpreted as stating the criteria of ‘intrinsic’ validity of reservations” on an objective basis (VCLT Commentary). For treaties with supervisory machinery attached, authoritative objective assessment may be possible. Human rights bodies have certainly demonstrated a willingness to engage in the activity (see, e.g., here and here). Several ICJ Judges in Armed Activities (here and here) also insisted that courts and tribunals are competent to objectively determine a reservation’s compatibility with a treaty’s object and purpose.

Like most treaties, dedicated oversight is not a feature AP I enjoys. The absence both of objections and of a supervisory mechanism cannot mean, however, that a reservation’s permissibility is beyond objective scrutiny. Authoritative determination of contested issues may be rare in international law, but external scrutiny and “unofficial” judgments therefore become an important function of the international legal academy.

How to Determine Compatibility?

Article 19(c) does not include the VCLT’s only reference to a treaty’s object and purpose. Seven other articles also refer to this, although none indicate how it should be discerned. Nor do the travaux préparatoires. According to the Commentary, “All that can be deduced . . . is that one should . . . place oneself at a sufficiently large level of generality” in order to ascertain a treaty’s “essence.” This is easier said than done. Even in relation to the Genocide Convention—perhaps the archetypal single-purpose treaty—the ICJ has found it difficult (e.g., here, here, here and here). Little wonder things are more complex in relation to wider-ranging treaties where, to borrow Sir Ian Sinclair’s description, “a search for the common intentions of the parties can be likened to a search for the pot of gold at the end of a rainbow.”

The ILC suggests deducing the object and purpose of a treaty via its interpretation: as a whole, in good faith, in its entirety, in accordance with the ordinary meaning to be given to the terms of the treaty in their context, including the preamble. Furthermore, recourse may be had to the preparatory work of the treaty, the “circumstances of its conclusion” and, where appropriate, the subsequent practice of the parties.

AP I’s title refers to the “Protection of Victims of International Armed Conflicts,” while the Preamble recalls the need to “reaffirm and develop the provisions protecting the victims of armed conflicts and to supplement measures intended to reinforce their application.” Neither is necessarily dispositive as to compatibility; the United Kingdom may accept the importance of protecting civilians from attack while believing that the ability to respond in kind should its own civilian population be attacked is essential to achieving this goal. Its reservation could therefore represent an attempt to achieve the object and purpose of AP I.

This approach is difficult to sustain, however. The inevitable cycle of reprisal and counter-reprisal cannot possibly enhance the protection of civilians. If we accept civilians as victims of armed conflict—which we must—there can be no place for violence directed against them, even as a reprisal, in a treaty explicitly seeking to reaffirm, develop, and reinforce their protection. Reprisals against, for example, enemy POWs are not accepted as a legitimate mechanism to prevent the mistreatment of a State’s own captured troops, and there is no reason why the balance between humanity and the availability of reprisals as a tool for enhancing respect for IHL should be any different with respect to civilians.

To argue that civilians share responsibility for the actions of their State or offer some broader military utility would be to accept that collective responsibility is tolerated by international law, or that first strikes against enemy civilians are lawful; civilians would be a legitimate means to an end, destroying the fundamental principle of distinction. Rather, as the International Criminal Tribunal for the former Yugoslavia indicated, “a slow but profound transformation of humanitarian law under the pervasive influence of human rights has occurred [whereby . . . ] belligerent reprisals against civilians and fundamental rights of human beings are absolutely inconsistent legal concepts” (para. 529).

The travaux préparatoires suggest instead that reprisals against civilians are incompatible with AP I’s object and purpose. A range of concerns and aims on the part of delegations is evident in opening discussions. These include the link between human rights and IHL, recognition and regulation of wars of national liberation and their participants, and regulating means and methods of warfare causing unnecessary suffering or with indiscriminate effect. However, a widespread desire to enhance civilian protection is also evident. Indeed, the ICRC saw the reaffirmation and development of norms protecting civilians from hostilities as the “the primary reason for the Diplomatic Conference,” even going so far as to say “the Conference would have been considered a failure if the legislative work had not been successful on this point.” Crucially, numerous delegations indicated the central importance of this to the Protocols.

In the more specific context of discussions around Article 51 AP I, various States again placed special importance on its provisions. Poland, for example, believed that it “contained the most important provisions of the Protocol, such as . . . the prohibition of attacks by way of reprisals,” and several other delegations gave particular support to the prohibition as corresponding to the “objectives” of AP I. Indeed, the German Democratic Republic said that the prohibition of reprisals was so important that a reservation on the same would be “incompatible with the humanitarian object and purpose of the Protocol.” Mexico likewise said such a reservation would “undermine [the Protocol’s] basis.”

Protection of the civilian population undeniably represents the object and purpose of at least Part IV of AP I, and one of the primary aims and objectives of AP I as a whole. Article 51(6) was inserted to help achieve this goal. Even if the illegality of reprisals against the civilian population was doubtful in 1977 (the prohibition may not have represented a “reaffirmation” of IHL), it is difficult to maintain their compatibility with a treaty also seeking to “develop” the protection of those not involved in hostilities. Violations of IHL in relation to precisely the individuals that the rules are designed to protect cannot be justified as a means to achieve compliance with those rules.

Conclusion

Dino Kritsiotis sees a treaty’s object and purpose as something that “exists in the eye of the beholder, and [which . . . ], like any proposition of law . . . is there to be interpreted, argued, contested and adjudicated.” Widespread argument and adjudication may both be absent here, but in the eye of this beholder, the United Kingdom’s reservation is incompatible with the object and purpose of AP I. It is impermissible as per VCLT Article 19(c) and is therefore invalid. The silence of other States parties cannot remedy this.

***

Photo credit: Sgt Nick Johns RLC