War and warfare cause not only human suffering and displacement, as well as damage to homes and infrastructure, but also extensive destruction and degradation of the environment. These impacts persist long after the conflict has ended. In this article, the authors explore the structural deficiencies and uncertainties of the existing international legal framework – specifically international humanitarian law (IHL) and international environmental law (IEL) – to address the environmental impacts of armed conflicts.
The direct and indirect protections that IHL offers for the environment during armed conflict are of problematic value. Few IHL provisions explicitly address environmental protection during armed conflict, and those that do are inadequate. Conversely, IEL is an extensive body of law protecting the environment and provides a growing body of standards and mechanisms for addressing environmental harm – and increasingly including issues of liability – during times of peace. Whether and to what extent IEL continues to apply and provide protection during armed conflict, however, is a matter of debate.
The authors highlight a few key gaps and deficiencies in IHL, and then consider opportunities that exist at the intersection of these two bodies of law. The article begins with an analysis of three key weaknesses in the existing body of IHL. First, Additional Protocol I to the 1949 Geneva Conventions defines the threshold of impermissible environmental damage (it must be ‘widespread, long-term and severe’)Footnote 1 in a way that is both excessively restrictive and unclear. Second, the provisions of IHL on the protection of civilian objects do not include a sufficient protection of elements of the environment from harm during armed conflict. Third, the proportionality of harm to the environment deemed ‘collateral damage’ is difficult to determine. For each of these gaps, specific opportunities of redressing the problem are presented.
The article then examines ways in which IEL may address some of the inadequacies of IHL in protecting the environment during armed conflict. Yet questions remain about the general applicability of IEL during hostilities and also the application of particular types of IEL provision. Some treaties expressly state whether they continue during armed conflict (e.g. certain provisions of the World Heritage ConventionFootnote 2), while others only indirectly address the issue (e.g. the Ramsar ConventionFootnote 3) or remain silent on it (e.g. the Convention on Biological Diversity). Moreover, there are several theories regarding the appropriate method for determining whether IEL continues to apply during armed conflict, and whether the answer to these questions varies for multilateral environmental agreements, principles of IEL, and customary international environment law. While diverse, these approaches highlight opportunities to complement the existing IHL provisions for protecting the environment during armed conflicts and to help answer the question whether and to what extent these IEL norms, standards, approaches, and mechanisms may be applied to prevent, address, or assess liability for environmental damage incurred during armed conflict.
International humanitarian law: gaps and opportunities
Armed conflict causes both direct and indirect environmental damage, which can endanger people's health, livelihoods, and security. To address these risks, IHL has incorporated fundamental environmental protections into the legal framework governing armed conflict.
A number of scholars have examined in great detail environmental protection under IHL.Footnote 4 Without attempting to catalogue all the relevant IHL provisions, the following analysis highlights a number of gaps and opportunities in the existing IHL framework. This section includes a brief history of IHL lawmaking; threshold questions of Additional Protocol I; elements of the environment as civilian objects; and proportionality of environmental damage as collateral damage.
Armed conflict and environment: a brief history of lawmaking
In the early 1970s, two developments occurred: the international community began addressing environmental protection generally, and it also made a serious attempt to remedy the deficiencies of legal protection for victims of armed conflict. Both developments were prompted by a scandalization of public opinion triggered by a number of key events. In the international environmental realm, these were environmental disasters such as major oil spills, as well as a broad citizens' movement. As to the law of armed conflict, the developments were the Vietnam War, the protection of human rights in occupied territories (and specifically in Palestine), and the armed conflicts that occurred during decolonization. These two focal points of public opinion came together in a single event: the environmental and public health effects of defoliation associated with the use of herbicides (particularly Agent Orange) during the Vietnam War.
While lawmaking in the international environmental field left this problem at the margins, the issue was directly addressed by international conferences dealing with armed conflict – to the great dismay of those defending military interests. These conferences included the United Nations Committee of the Conference on Disarmament (CCD) and the Diplomatic Conference on the Reaffirmation and Development of International Humanitarian Law (CDDH, Geneva, 1974–1977).
In 1976, the CCD adopted the Convention on the Prohibition of Military or Any Hostile Use of Environmental Modification Techniques (ENMOD). This treaty regulates the use of environmental modification techniques as a means to cause harm to the enemy. ENMOD specifically prohibits ‘environmental modification techniques having widespread, long-lasting or severe effects as the means of destruction’.Footnote 5 The travaux of the CCD indicate how to interpret these key terms: ‘long-lasting’, for instance, signifies ‘lasting for a period of months, or approximately a season’.Footnote 6
The more difficult questions were addressed by the CDDH when it discussed the adoption of two important provisions in what became Additional Protocol I. First, Article 35, paragraph 3, states that ‘[i]t is prohibited to employ methods and means of warfare which are intended, or may be expected, to cause widespread, long-term and severe damage to the natural environment’. Second, Article 55 of Additional Protocol I states that ‘[c]are shall be taken in warfare to protect the natural environment against widespread, long-term and severe damage’. It further specifies that this protection ‘includes a prohibition of the use of methods and means of warfare which are intended or may be expected to cause such damage to the environment and thereby to prejudice the health or survival of the population’.
These provisions were negotiated and adopted with knowledge of the CCD text.Footnote 7 Thus, the differences between the texts are not an oversight but intentional. The three conditions, or the threshold, of the prohibition in Additional Protocol I are cumulative (joined by ‘and’), while the conditions in ENMOD are alternatives (joined by ‘or’). Furthermore, the meaning of the three adjectives limiting the scope of prohibited damage differs (at least according to the negotiating history of the two provisions) depending on whether they are being interpreted in the context of Additional Protocol I or ENMOD. The competent Conference committee intensively discussed all three in the context of Additional Protocol I. With regard to the time or duration element, the report of the committee states that the duration:
was considered by some to be measured in decades. Reference to twenty or thirty years were made by some representatives as being a minimum … [I]t is impossible to say with certainty what period of time might be involved. It appeared to be a widely shared assumption that battlefield damage incidental to conventional warfare would not normally be proscribed by this provision.Footnote 8
This definition radically differs from the one accepted in the travaux of ENMOD just quoted. As a consequence, negotiators assumed that, in practice, these provision would ‘not impose any significant limitation on combatants waging conventional warfare’.Footnote 9 It might leave room for application to biological and chemical warfare. As to nuclear warfare, however, most NATO countries claim that Additional Protocol I does not apply thereto,Footnote 10 and at least the United States and the United Kingdom contest that these two provisions constitute customary law.Footnote 11 The environmental community has never considered this situation satisfactory.
As a result, a debate arose about further lawmaking. A number of events, primarily the oil spills caused by the 1980–1988 and 1990–1991 Gulf Wars, sustained public debate on the issue. There were, for example, far-reaching demands for new international law – a ‘Fifth Geneva Convention’. Those efforts, however, met considerable resistance from important military powers. The only tangible result of the efforts to adopt new documents was a set of Guidelines on environmental protection during armed conflict, elaborated by the International Committee of the Red Cross (ICRC) and published in 1994.Footnote 12 These Guidelines, however, did not constitute any significant progress for better protection of the environment during armed conflict, and even this modest document received a somewhat hostile reception at the UN. The UN General Assembly politely buried it in 1994.Footnote 13
The next step in lawmaking was the 1998 Rome Statute,Footnote 14 establishing the International Criminal Court. In its definition of war crimes, the Rome Statute contains a provision protecting the environment in times of international armed conflict:
Intentionally launching an attack in the knowledge that such attack will cause incidental loss of life or injury to civilians or damage to civilian objects or widespread, long-term and severe damage to the natural environment which would be clearly excessive in relation to the concrete and direct overall military advantage anticipated.Footnote 15
This provision is related, but not identical to, three provisions of Additional Protocol I. These include Article 51(5)(b), prohibiting attacks that cause ‘excessive’ collateral damage to civilians or civilian objects, and the two provisions concerning the environment quoted above (Articles 35(3) and 55).
To assess the significance of the Rome Statute provision, one has to bear in mind that criminal law contains secondary norms. They constitute means to enforce a primary obligation. As such, they need not be identical to those primary obligations. While a provision of international criminal law presupposes a primary norm prohibiting the behaviour that constitutes a crime, many primary obligations do not trigger criminal sanctions. Thus, a more narrow criminal provision on environmental damage during armed conflict (holding individuals personally and criminally liable) does not alter the underlying primary obligation of states, under existing international law, to prevent a broader range of environmental damage.
The lack of clarity surrounding the bounds of treaty obligations to avoid damaging the environment during armed conflict raises the question of environmental protection under customary humanitarian law and the treatment of the question by the Customary International Humanitarian Law Study published by the ICRC in 2005.Footnote 16 The Study declares a simplified version of the provisions of Additional Protocol I and of ENMOD to constitute customary law, stating that the ‘use of methods and means of warfare that are intended, or may be expected, to cause widespread, long-term and severe damage to the natural environment is prohibited. Destruction of the natural environment may not be used as a weapon’.Footnote 17
A real step forward was accomplished via another rule articulated in the Study. Rule 44 states that
[m]ethods and means of warfare must be employed with due regard to the protection and preservation of the natural environment. In the conduct of military operations, all feasible precautions must be taken to avoid, and in any event to minimise, incidental damage to the environment.Footnote 18
In addition, a ‘[l]ack of scientific certainty as to the effects on the environment of certain military operations does not absolve a party to the conflict from taking such precautions’.Footnote 19
The rule uses a variation of the general IHL rule that ‘precautionary measures’ must be taken to avoid damage to civilians and civilian objectives.Footnote 20 The last sentence of Rule 44, however, constitutes an application of the precautionary principle, well established in IEL, to the duty to take precaution in armed conflict, which in the latter context amounts to a revolution. The Study quotes limited state practice to support this rule. The majority of the support cited is the reading of two decisions of the International Court of Justice (ICJ), namely the 1995 Nuclear Tests case order and the 1996 Nuclear Weapons advisory opinion.Footnote 21 The Study in effect argues that recognition of the precautionary principle as customary international environmental law must be reflected in the law of armed conflict.
The ‘due regard’ principle formulated in Rule 44 seems to be well accepted. This is elucidated by two private, yet semi-official, restatements of relevant rules. The San Remo Manual on International Law Applicable to Armed Conflict at Sea (1994) states that ‘[m]ethods and means of warfare should be employed with due regard for the natural environment …’.Footnote 22 The HPCR Manual on International Law Applicable to Air and Missile Warfare (2009) formulates rules along the same lines, stating that the ‘destruction of the natural environment carried out wantonly is prohibited’.Footnote 23 It also urges that ‘[w]hen planning and conducting air and missile operations, due regard ought to be given to the natural environment’.Footnote 24
It is submitted that the prohibition of ‘wanton’ destruction of natural environment elements and the due regard principle are both more favourable for the environment and more flexible than the provisions of Additional Protocol I.
Additional Protocol I and the threshold question
As noted above, the crucial problem raised by Additional Protocol I is the meaning of the three conditions attached to the prohibition on ‘long-term, widespread and severe’ damage to the environment.Footnote 25 The narrow scope of the prohibition is due both to the cumulative character of the three conditions and to their interpretation, which effectively sets the threshold very high, but also with some ambiguity. At least if interpreted in the light of the negotiating history,Footnote 26 it seems next to impossible that the threshold could be reached by conventional warfare. Regarding chemical warfare, the only practical case thus far (the use of herbicides in Vietnam) raises some doubts. If the yardstick for ‘long-lasting’ is several decades, the threshold might not have been reached. Nature has recovered in many but not all places in Vietnam; however, damage to human health remains, and may last for generations.
The above interpretation is not necessarily generally accepted nowadays, but it is safe to state that the parties to the treaty intended to establish a very high threshold. Natural resources and the environment are essential to post-conflict peacebuilding,Footnote 27 and significant environmental damage can undermine efforts to provide for livelihoods, promote economic recovery, and allow society to return to a ‘normal’ peacetime way of life. A framework that is too permissive of environmental damage during armed conflict can thus undermine long-term peace.
One can ask whether this high threshold is still valid, or whether it has fallen into desuetudeFootnote 28 in light of the continually increasing recognition of environmental concerns in international relations. This is arguable, but not certain.
This brings us to gap number one: for two reasons, the legal situation is highly unsatisfactory from an environmental point of view. First, the conditions attached to the prohibition of Articles 35 and 55 of Additional Protocol I are excessively restrictive, making the prohibition much too narrow from an environmental point of view. Second, the exact scope of this prohibition remains uncertain, and thus difficult to implement or enforce.
This gap presents an opportunity, however: an alternative form of legal protection for the environment in times of armed conflict appears in a number of documents purporting to reflect customary law – namely, the ‘due regard’ principle and the prohibition of ‘wanton’ destruction. Whether and how this can be clarified remains an open question.
Elements of the environment as civilian objects
Elements of the environment are most often civilian objects. As such, they are protected against attacks. To that extent, the restrictive conditions of Articles 35 and 55 of Additional Protocol I do not apply. This protection is, however, shaky, as environmental elements can easily become military objectives. Once armed forces are located in a protected area, the area may contribute effectively to military action and its neutralization may offer a definite military advantage. Thus, it becomes a military objective. In the case of herbicide use in Vietnam, the trees provided cover for the enemy. Their defoliation constituted a definite military advantage, and the trees – more precisely their leaves – became a military objective.
This brings us to the second gap in IHL: elements of the environment are all too likely to become military objectives, invalidating their protections as civilian objects. In theory, Articles 35 and 55 of Additional Protocol I could restrain environmental destruction, but this brings us back to gap number one.
Gap number two provides opportunity number two: the transformation of environmental elements into military objectives must be prevented. How can this be achieved? Articles 59 and 60 of Additional Protocol I (non-defended localities and demilitarized zones) may serve as a model. Because there is no military presence in these areas, they are not or are no longer military objectives, and therefore are immune from attack. In the same way, environmentally sensitive areas may be rendered immune against attacks by excluding any military presence. In the absence of a treaty to that effect, this cannot be achieved by a unilateral declaration. Whether a new treaty is possible is highly doubtful, at least in the near future. Nevertheless, the parties to a conflict can achieve the same result by mutual agreement, such as through mediation led by the ICRC, the UN, or a relevant environmental organization.Footnote 29 An international organization could call upon the parties to conclude such agreements. Finally, the UN Security Council, in the exercise of its powers under Chapter VII of the UN Charter, could designate such protected areas and oblige the parties to conclude such agreements.
Environmental damage as collateral damage: the question of proportionality
The environment may be damaged indirectly by attacks against military objectives. This is the case with oil spills (if the direct target is a military objective) or pollution caused by attacks against industrial installations. In this case, the elements of the environment that are affected constitute civilian objects. Damage to these environmental objects would then be ‘collateral damage’, which is permissible only to the extent that it is not excessive in relation to the concrete and direct military advantage anticipated as a result of the attack. This raises a number of difficult questions.
As a rule, there will be uncertainties regarding the environmental impacts of the collateral damage. Where damage occurs outside military objectives (for example, an oil spill caused by the attack against a power station situated near the coast), it is first necessary to assess the extent of the environmental damage, which can be quite difficult. To assess the damage caused by the attack, it is necessary to determine not only the extent of pollution and the harm caused after the attack but also the amount of pre-existing pollution and its effects. There is also the uncertainty involved in predicting long-term damage.Footnote 30
Assuming it is possible to clarify the relevant facts regarding the damage, what is the legal yardstick of proportionality? This is already difficult in the case of simple physical damage. It becomes all the more difficult in the case of environmental damage, in particular long-term damage. Is the precautionary principle relevant in determining the relative weight of environmental values in the so-called proportionality equation? In other words, can an environmental damage be ‘excessive’ within the meaning of Article 51 of Additional Protocol I even if the full extent and nature of the damage are not certain?
If an element of the environment is lawfully attacked because it constitutes a military objective, there might be long-term environmental damage beyond the actual destruction. Are Articles 35 and 55 of Additional Protocol I lex specialis for determining the lawfulness of that long-term damage, and thus not subject to the principle of proportionality under IHL? Alternatively, is this long-term damage a form of ‘collateral’ damage that would then be subject to the yardstick of proportionality?
This brings us to the third gap: there is a lack of clarity about the practical issues of proportionality where environmental damage is collateral damage, caused by attacks against military objectives.
These issues of proportionality also provide the third opportunity, which is to clarify proportionality. A new treaty, even assuming it were politically possible, would be unlikely to resolve the issue, at least for the substantive problems involved. It seems problematic – if not impossible – to provide a general solution that applies to all possible situations of environmental collateral damage. However, general rules for assessing environmental damage could be helpful. They could be established by resolutions of relevant international organizations. As to the substantive issues, the best solution would probably be for expert groups to analyse typical scenarios and develop a set of criteria for determining proportionality.
Conclusions on international humanitarian law
IHL provisions that are relevant for environmental protection during armed conflict constitute a body of treaty and customary law with significant gaps and deficiencies.
First, the primary provisions that directly protect the environment during armed conflict – Articles 35 and 55 of Additional Protocol I – do not adequately achieve this aim because the ‘widespread, long-term and severe’ damage threshold is imprecise and difficult to meet, leaving much of the serious environmental harm arguably outside the scope of current protections. Thus, these terms require clear and more appropriate definitions.
Second, because few IHL norms explicitly address environmental protection, indirect means may provide more effective protection by regulating the means and methods of warfare or by protecting civilian persons and objects. However, the gap here is that elements of the environment are likely to become military objectives, which brings us back to gap number one. This problem may be solved by using Articles 59 and 60 of Additional Protocol I (non-defended localities and demilitarized zones) as a model for rendering environmentally sensitive areas immune.
Third, there is a lack of clarity surrounding collateral damage to civilian objects as a result of attacks against military objectives. The gap here is a lack of clarity about the practical proportionality issues where environmental damage is collateral damage, caused by attacks against military objectives.
There are other deficiencies in the existing IHL framework.Footnote 31 These include, for example, uncertainties in IHL governing protection of the environment during non-international armed conflict.Footnote 32 This is problematic, as the vast majority of current armed conflicts are non-international.
Application of international environmental law during armed conflictFootnote 33
In the light of the significant gaps and deficiencies in the IHL framework regarding the protection of the environment during armed conflict, it must be asked whether some of these gaps could be addressed through the application of international environmental law (IEL). IEL provides a well-established body of norms, standards, approaches, and mechanisms preventing and redressing – including through responsibility and, increasingly, liability – damage to the environment during times of peace. With the development, maturation, and application of a substantial body of law governing environmental use and protection, the question is whether and to what extent these provisions of IEL continue to apply during armed conflict and whether they provide a meaningful protection against the specific risks of warfare. IEL rules protecting certain natural resources are one example to be considered. Their ability to restrain waste and destruction by warfare has to be assessed. Rules that prohibit the causation of specific types of environmental damage should also be taken into account. To use an example referred to above, if a power station is destroyed during a war or other military operation (as happened with the power plant at Jiyeh, Lebanon, in 2006), should the subsequent oil spill trigger an institutional mechanism for response and clean-up or a liability regime?Footnote 34 Would the World Heritage Convention, which protects sites of cultural and natural heritage,Footnote 35 prohibit targeting or causing extensive harm to a World Heritage site during military activities?
Questions regarding the application of IEL during armed conflict are complicated for two reasons. First, environmental law is still dynamic, and it keeps developing. Second, this is part of the general phenomenon of fragmentation of international law, which entails the question of how, in the case of overlapping fields of application, different regimes or bodies of international law relate to each other. Scholarship and commentary provide several perspectives on whether IEL applies during armed conflict. Starting in the 1990s, there has been a noticeable shift in the historic belief that laws applicable during war and peace were mutually exclusive. Contemporary perspectives increasingly bridge the two bodies of law, applying peacetime international law during armed conflict to varying degrees. Where both bodies of law apply concurrently, however, the question of their relationship (lex specialis) also has to be answered.
This development is clearly documented in the work of the International Law Commission (ILC). In 2004, the General Assembly approved the ILC's proposal to include work on the ‘effects of armed conflict on treaties’ in its long-term programme. In 2008, that work resulted in a set of draft articles that attempt to regulate the applicability of treaties during armed conflicts.Footnote 36 The draft articles state that the onset of armed conflict ‘does not necessarily terminate or suspend the operation of treaties’ between belligerents or belligerents and neutral parties.Footnote 37 Rather, this is determined by a complex body of different considerations: express provisions and subject matter of the treaty, treaty interpretation according to Articles 31 and 32 of the Vienna Convention on the Law of Treaties (VCLT) of 23 May 1969, the nature and extent of the armed conflict, and the effect of the armed conflict on the treaty.Footnote 38 What this means in practice has to be ascertained on a case-by-case basis.
This part of the article provides an overview and analysis of law and commentary addressing IEL applicability during armed conflict. It is organized in three main sections: (a) multilateral environmental agreements that might be applied; (b) customary international environmental law and soft-law instruments that might be applied; and (c) commentary on the applicability of these two bodies of IEL during armed conflict.
Multilateral environmental agreements
With regard to the question of whether and to what extent multilateral environmental agreements (MEAs) continue to protect the environment in times of armed conflict, a basic distinction has to be made, which is also a basic problem. As a rule, the law applicable in times of peace applies between belligerents and neutral states (more generally, states not parties to an armed conflict). This means that, at least as a matter of principle, an MEA must continue to apply during an armed conflict at least in the relation between the parties to the conflict and the states that are not parties. From this perspective, the continued application of an MEA during an armed conflict only constitutes a problem in the relationship between belligerents. In addition, and in line with this logic, the existence of a non-international armed conflict arising on the territory of a party to an MEA does not affect the application of the treaty. The situation is not, however, as simple as that. The fundamental question of whether a state involved in an armed conflict, whether an international armed conflict or a non-international one, can be expected to fulfil its international obligations in the same way as if the conflict did not exist has to be asked and answered. The clausula rebus sic stantibus Footnote 39 and necessity as a circumstance precluding wrongfulnessFootnote 40 may modify the relationship between parties and non-parties to an armed conflict in comparison to their relationship in time of peace. While the basic distinction between the relationship between belligerents and the one between belligerents and non-belligerents has to be maintained, a closer look at the scope of that distinction is necessary, taking into account the basic content of relevant MEAs.
How instruments of IEL address their applicability during times of armed conflict varies substantially. Some MEAs directly or indirectly provide for their continued application during hostilities, while others specifically state that they are automatically suspended, terminated, or inapplicable once armed conflict has begun; others remain silent on the issue. Unfortunately, most MEAs fall into the third category, so there is substantial uncertainty.Footnote 41
An MEA may indirectly provide that it continues to apply during armed conflict. Under the World Heritage Convention, the World Heritage Committee establishes and updates a World Heritage List of cultural heritage and natural heritage properties with ‘outstanding universal value’.Footnote 42 Inclusion on the list requires the concerned state's consent. In addition, the Committee keeps a ‘list of World Heritage in Danger’ that includes sites that require ‘major operations’ to conserve, for which assistance has been requested, and that are ‘threatened by serious and specific dangers’. Serious and specific dangers may include ‘the outbreak or the threat of an armed conflict’.Footnote 43
Another example is found in the Ramsar Convention, which establishes a List of Wetlands of International Importance.Footnote 44 The Convention does not expressly state whether it applies to belligerents; however, intent may be inferred from the Convention's language that a party to the agreement has the right, ‘because of its urgent national interests, to delete or restrict the boundaries of wetlands already included by it on the List’.Footnote 45 It is possible, although unclear, that situations of ‘urgent national interests’ may include armed conflict.Footnote 46 It is notable that urgent national interests do not allow a party to restrict the protective measures of a listed wetland, only to amend its boundaries. The question then arises whether a military use of the area constitutes a violation of the duties of protection established by the Convention.
Similarly, the UN Convention on the Law of the Sea (UNCLOS)Footnote 47 requires states party ‘to protect and preserve the marine environment’, as well as to take measures to prevent, reduce, and control marine pollution.Footnote 48 Article 236 of UNCLOS, however, establishes a variance for ‘any warship, naval auxiliary, other vessels or aircraft owned or operated by a State and used, for the time being, only on government non-commercial service’. Parties are then instructed to ‘ensure, by the adoption of appropriate measures not impairing operations or operational capabilities … that such vessels or aircraft act in a manner consistent, so far as is reasonable and practicable, with this Convention’.Footnote 49 Although the standards applied to military as opposed to non-military vessels and aircraft vary, UNCLOS may continue to apply during armed conflict.Footnote 50 More specific requirements are formulated in the San Remo Manual on International Law Applicable to Armed Conflicts at Sea.Footnote 51 Article 44 of the Manual states that ‘[d]amage to or destruction of the natural environment not justified by military necessity and carried out wantonly is prohibited’. It also provides that, when hostile actions are undertaken within the exclusive economic zone of a neutral state, belligerents shall ‘have due regard for the rights and duties of the coastal State, inter alia, for … the protection and preservation of the marine environment’.Footnote 52
In contrast, some MEAs explicitly suspend, derogate, or terminate the agreement between belligerents during armed conflict. For example, the Convention on Third Party Liability in the Field of Nuclear Energy (1960) exempts operators for damage directly resulting from armed conflict or similar activities,Footnote 53 although Austria and Germany objected to this provision and explicitly declared their right to hold operators liable for such damage.Footnote 54 This is equivalent to the established principle of insurance law that insurance against losses does not cover war damages. Seen in this light, the rule would apply not only to the relationship between belligerents but also to that between belligerents and non-belligerents.
Many MEAs contain no reference to their applicability during armed conflict. Such MEAs include the Convention on Biological Diversity (1992), the UN Convention to Combat Desertification (1994), and the Convention on the Conservation of Migratory Species of Wild Animals (1979).Footnote 55 The effect of their silence, and whether it varies by type of convention, is uncertain. For example, commentary has posited that the Convention on Biological Diversity applies to belligerent parties, as it is analogous to human rights treaties that do not automatically terminate upon hostilities.Footnote 56 This uncertainty raises questions about how parties should proceed, such as whether belligerents should agree on sites to be placed off limits,Footnote 57 or if military entities should be instructed on the principles of the MEAs and charged to abide by them to the extent possible.
Customary international environmental law and soft-law instruments
Certain soft-law instruments explicitly refer to armed conflict. Other IEL principles and soft-law instruments may apply, although they do not address armed conflict directly. So-called soft-law instruments are not legally binding unless they rise to the level of customary IEL. An example is furnished by arguments discussing whether the precautionary principle and the right to a healthy environment constitute – or are emerging – customary IEL.Footnote 58 Even if a soft-law instrument is deemed to not constitute customary IEL, it may still inform the interpretation and application of international law.
The Declaration of the United Nations Conference on the Human Environment (Stockholm Declaration)Footnote 59 of 1972 articulated an overarching principle that may bear on IEL applicability during armed conflict. Principle 21 provides that ‘States have … the responsibility to ensure that activities within their jurisdiction or control do not cause damage to the environment of other States or of areas beyond the limits of national jurisdiction’.Footnote 60
Two decades later, the Declaration on Environment and Development (Rio Declaration) of 1992 stated in Principle 24 that: ‘Warfare is inherently destructive of sustainable development. States shall therefore respect international law providing protection for the environment in times of armed conflict and cooperate in its further development, as necessary’.Footnote 61 While the intent to protect the environment is clear, the provision's precise meaning is less so; it may mean that IEL applies during conflict, or it may simply reiterate required state adherence to relevant IHL provisions.Footnote 62
The Rio Conference adopted similar language in the Programme of Action for Sustainable Development (Agenda 21) in Article 39(6), detailing the means of implementation. It states that ‘[m]easures in accordance with international law should be considered to address, in times of armed conflict, large-scale destruction of the environment that cannot be justified under international law’. The Article specifies that the UN General Assembly and Sixth Committee should handle such efforts, with ICRC consideration.Footnote 63
More broadly, Principle 5 of the World Charter for Nature mandates that ‘[n]ature shall be secured against degradation caused by warfare or other hostile activities’.Footnote 64 This principle appears intended to prohibit environmental harm during armed conflict. But it is probably rather a political postulate than the expression of a legal rule. In the light of the controversies surrounding the customary law character of rules protecting the environment in times of armed conflict, such a sweeping rule could not reasonably meet the test of general practice and opinio iuris.
UN General Assembly resolution 47/37, adopted in 1993, urges states to take measures for complying with international law protecting the environment during armed conflict.Footnote 65 Although it also encourages incorporating such international law into military manuals, the precise import of these provisions remains unclear. Its reference to provisions ‘applicable to the protection of the environment’ may refer to relevant provisions within IHL, or to IEL. The resolution led to the ICRC's development of Guidelines for incorporating environmental protection into military manuals, discussed earlier.Footnote 66
While some principles expressly address armed conflict, others – such as the Trail Smelter principle – are silent. Nonetheless, they may be applied to armed conflict. The Trail Smelter principle arose from an arbitral decision resolving a dispute between the United States and Canada regarding transboundary air pollution in which a Canadian smelter harmed US crops and forests downwind of it.Footnote 67 The Trail Smelter arbitral panel held that Canada had a responsibility to prevent harmful transboundary air emissions from the smelter, and was liable for the damages that such emissions incurred. The decision was based on a fundamental responsibility to use one's territory so as not to cause harm to that of another. It must nowadays be considered as a rule of customary international law.
The Trail Smelter principle may afford protection to non-belligerent, neutral territories by establishing state responsibility for environmental damage caused outside the state where the acts or events entailing such damage occur. Certain comments suggest that such allocation might not apply if belligerent interests outweigh the victim state's harm.Footnote 68 Where damage is caused in neutral territory, this thesis is in contradiction with the general principle of the law of neutrality that the neutral territory is inviolable and that the neutral state, as a matter of principle, may not be affected by the armed conflict. There is no basis in state practice for the suggested exception to this customary rule of international law. The frequent reiteration of the Trail Smelter principle indeed indicates the rapid emergence of a state's right to environmental protection as customary IEL that also applies during armed conflict.Footnote 69
This conclusion results, in particular, from the jurisprudence of the International Court of Justice. In the Corfu Channel case, the ICJ practically extended the Trail Smelter principle to the actions of parties during conflict, although the case did not specifically address transboundary pollution.Footnote 70 In this case, the ICJ held Albania responsible for damage from mines laid in Albanian waters to British ships travelling through these waters, observing that international law obliges the state ‘not to allow knowingly its territory to be used for acts contrary to the rights of other states’.Footnote 71
The ICJ also recognized the Trail Smelter principle in the 1996 Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons. ICJ advisory opinions, although not legally binding, provide persuasive evidence of customary international law and the application and implementation of international law. In this advisory opinion, the Court notes that a state's general obligation ‘to ensure that activities within their jurisdiction and control respect the environment of other States or of areas beyond national control is now part of the corpus of international law relating to the environment’.Footnote 72 The decision continues by instructing states to account for environmental considerations when determining what constituted necessary and proportionate measures in light of military objectives.Footnote 73 Following the decision, states appear, at a minimum, to be required to abide by the Trail Smelter principle; the outer boundaries of mandatory environmental protection, however, are less certain.
The same concept of neighbourly protection from harm is seen in the 2010 decision by the ICJ in the Pulp Mills case.Footnote 74 The Court decided that the construction and operation of Pulp Mills in Uruguay required the country to undertake a transboundary environmental impact assessment. The Court recognized that a state must take specific measures to prevent harm to its neighbours, thus extending the general principle of the Trail Smelter decision.
Commentary
MEAs, rules of customary environmental law, and soft-law instruments of IEL afford numerous examples of rules and decisions that establish norms (including on responsibility and liability) relating to environmental damage, and most commentary holds that many of these provisions continue to apply during armed conflict. The next section examines a few key lines of doctrine on the issue.
Commentary on the continued applicability of MEAs in times of conflict
A number of lines of argument have been developed concerning the question as to whether and to what extent MEAs apply during armed conflict. These include classification theory, intention theory, the context and nature of the MEA, and a sliding-scale system.
‘Classification theory’ assigns environmental laws to categories that determine their application during armed conflict. For example, Vöneky found that the current categorization methodology assumes IEL applies during times of both peace and armed conflict for three categories of treaties: if the legal instrument expressly says it continues to apply, if it is compatible with armed conflict, or if it is a jus cogens or erga omnes obligation.Footnote 75
By combining classification theory and intention theory – that is, the approach of holding as closely as possible to the original intent of the parties at the time that the treaty was formed – Boelaert-Suominen addressed instances in which an international agreement does not clearly fit into pre-defined categories.Footnote 76 The classification system has been criticized as overly simplistic for complex international relations; conversely, intention theory holds as closely as possible to the parties' original intent at the time that they formed the treaty.Footnote 77 Boelaert-Suominen concluded that, when an MEA does not clearly fit into a particular category, a case-specific analysis of the parties' intention should decide the matter.Footnote 78
A third set of theories holds that MEA applicability during armed conflict can be based on the context and nature of the agreement. Schmitt identifies a theory of differentiation, which holds that continuance of a treaty during armed conflict depends on the consistency of continuance with the context of the agreement. He reasons that, in the absence of express termination language or clear inconsistency, MEAs continue during armed conflict.Footnote 79 Sharp published a related theory according to which a treaty containing no termination clause may be denounced or withdrawn depending on ‘the nature of the treaty’.Footnote 80 He suggested that, if the treaty is directed at ‘sovereign’ relations, it should be suspended or terminated during armed conflict. Conversely, a treaty regulating non-military conduct or indirect state interactions should not automatically conflict with a state of hostilities.Footnote 81
Another approach is the sliding-scale theory, which reflects a balancing of environmental protection against military mission success.Footnote 82 It expresses an inverse relationship between IEL's effect and the degree of military operations: when military operations are low-intensity (e.g., during training), environmental laws are in almost full effect, but as the military operations grow the effect of environmental laws lessens. Unfortunately, this approach does not provide concrete explanations or criteria regarding which rules bind a military entity during different types and phases of military engagement.
These lines of argument are not mutually exclusive. A large part of these theories is reflected in the ILC draft articles on the effect of armed conflict on treaties.Footnote 83
Commentary on the continued applicability of customary IEL in times of conflict
Although customary international environmental law and MEAs are both binding, there may be differences in their application during armed conflict. This distinction may stem in part from the difficulty of determining whether a specific measure is a customary international law and what its precise legal scope is. Most commentators posit that customary IEL applies during armed conflict in a similar manner to MEAs. Citing the ICJ Statute, Parsons emphasized that environmental norms are equally relevant and applicable during armed conflict.Footnote 84
Some support for this argument may be derived from the so-called Martens Clause, which addresses the role of norms, custom, and practice as the law of war develops.Footnote 85 The clause appears in a number of treaty provisions, in particular in the Hague Convention on the laws and customs of war (versions of 1899 and 1907) and in Additional Protocol I of 1977, which are similar in their essential content. They refer to sources of law other than the treaties in which the clause is included. In the 1977 version, the clause reads:
In cases not covered by the present Protocol and other international agreements, civilians and combatants remain under the protection and authority of the principles of international law derived from established custom, from the principles of humanity and from the dictates of public conscience.Footnote 86
In the current stage of development of the international community, it is appropriate to subsume the protection of the environment under all three of the sources of protection enumerated in the clause.
Some observers note that particular international environmental principles presently, or may soon, constitute customary IEL. Soft-law instruments are quoted in this respect. One article notes that ‘since [the World Charter for Nature] was adopted by a significant number of States, at the very least the Charter is incorporated in customary international law’.Footnote 87 The question of whether and to what extent resolutions of international organizations can be used as proof of customary international law can in principle be answered affirmatively, but each case has to be analysed on its own merits. The fact that a resolution was adopted by consent and without declarations to the contrary is an important circumstance. In Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), the ICJ based part of its decision on the parties' consent to a UN resolution. The Court explained that, in the instant case, opinio juris – one of the two elements of customary international law – ‘may … be deduced from, inter alia, the attitude of the Parties and the attitude of States towards certain General Assembly resolutions’.Footnote 88 This holding of the Court related to the question of whether and to what extent the so-called Friendly Relations DeclarationFootnote 89 and the definition of aggressionFootnote 90 adopted by the General Assembly constituted customary international law. If widely accepted UN resolutions may constitute evidence of opinio juris, at least some of the soft-law documents previously discussed would be closer to constituting customary international law and becoming binding on all states.
Remaining questions
The preceding argument has shown that there are a number of approaches for determining when and how IEL applies during armed conflict. Nevertheless, the question is open and there is probably no general and simple answer to it. There is no ICJ decision clarifying the issue, and the ILC draft articles leave questions because they offer a variety of considerations to determine the continued application of treaties during armed conflict. In the meantime, MEAs continue to be negotiated, implemented, and enforced, but many of them similarly fail to provide clarification regarding their application during armed conflict. A remarkable exception is shown by the recent articles elaborated by the ILC on the protection of transboundary aquifers, which contain a specific provision on protection in times of armed conflict.Footnote 91 Military manuals of many countries incorporate environmental provisions, yet these tend to focus on the requirements found in IHL. Specific interpretations and practice remain inconsistent. Further analysis and clarification is necessary to make the approaches that have been discussed operational, for example deciding how to implement a sliding scale.
Several general questions of both IHL and IEL have a bearing on the issue discussed in the present article. The first one is the distinction between international and non-international armed conflict. It seems at least plausible that environmental treaties signed by a state on the territory of which a non-international conflict takes place would not be affected by the existence of this conflict. But what if the treaty applies to parts of the territory that are no longer under the control of the governmental party to the conflict? How far can non-compliance with the treaty be justified on the basis of a state of necessity?Footnote 92 Second, the relationship between the rules concerning the continued application of treaties during armed conflict and the law of neutrality remains to be clarified. If the relations between a belligerent and a neutral state in principle continue to be subject to the law of peace alone, do the duties of abstention and impartiality, being essential elements of the law of neutrality, not modify certain obligations flowing from environmental treaties, be they multilateral or bilateral? Third, what is the relationship between MEAs and customary environmental law during armed conflicts? Fourth, what is the relationship between the private law liability for environmental damages and state responsibility for the causation of the same damages? Modern environmental treaties rely to a great extent on private law liability of the private actors involved. Can this solution be maintained in a situation of armed conflict? Finally, there is the general question regarding the relationship between different MEAs that overlap in certain instances. Whatever answer is given to the question in general, is it the same for the situation of peace and for that of an armed conflict?
This article has attempted to address the question of whether IEL continues to apply during armed conflict. The subsequent questions, which remain unanswered, are when and how it might apply in specific circumstances. For example, if area-based agreements such as the World Heritage Convention and the Ramsar Convention continue to apply, they might protect specified areas in the absence of military necessity. Although the protection offered might be limited, it is a relatively clear application. Conversely, consider the Convention on International Trade of Endangered Species of Wild Flora and Fauna (CITES) in the context of conflict resources.Footnote 93 If the agreement continued to apply, would it spur states receiving such goods to take action against the country of origin, if it were party to CITES? To what effect? How might the analysis change if rebels are engaging in the trade in CITES-listed species? Alternatively, the effect of MEAs may be to create a basis for providing assistance in certain circumstances; or to provide a framework of standards and methodologies, for example for conducting environmental assessment or remediation in conflict contexts (as the UN Compensation Commission borrowed peacetime environmental methodologies for assessing and valuing natural resource damages). These are but some of the questions and scenarios that arise when considering the practical effect and import of IEL during armed conflict.
None of these questions have clearly defined answers, and all involve significant considerations that must be carefully balanced.
Conclusions
Significant efforts to protect the environment during armed conflict began in the 1970s, mainly in response to events that had taken place during the Vietnam War; there was a resurgence of international attention to the issues following the deliberate oil fires and spills during the 1990–1991 Gulf War. However, the framework that has developed over the past four decades contains numerous gaps and ambiguities.
Although IHL reacted very soon after the development of a broad environmental awareness in the late 1960s, three key gaps or deficiencies remain:
— the regrettably narrow, but also unclear definition of impermissible environmental damage in the provisions explicitly addressing environmental protection during armed conflict;
— unsatisfactory protection of the elements of the environment as civilian objects owing to the easy transformation of such elements into military objectives; and
— doubts as to the practical application of the principle of proportionality to environmental damage, which is so-called collateral damage resulting from attacks against military objectives.
This article has shown a few ways in which these deficiencies can be remedied. It has also explored the chances that could be provided by a continued application of international environmental law, created essentially for the situation of peace, during armed conflict. In this respect, too, major uncertainties remain.
More clarity regarding the extent of environmental protection during armed conflict is necessary. Some questions (such as how various provisions of IEL apply during armed conflict) require additional research. Others (such as the threshold for ‘widespread, long-term and severe’ under Articles 35 and 55 of Additional Protocol I) require international action. Belligerents are also called upon to address some of the existing deficiencies on a case-by-case basis through mutual agreement, and the UN Security Council can play a role in this regard. Despite the doubts and challenges discussed above, it appears that international law continues to maintain the concern for future generations – which is the essence of environmental law – in times of war.
