Guest Post by Erik Zouave: The Weaponization of the Falluja Dam and the River Euphrates: A Non-State Test for International Environmental Humanitarian Law
Erik Tristan Zouave (LLM, University of York) is a trainee at the Swedish International Development Cooperation Agency (Sida) Iraq Unit and a former research-assistant at the Swedish National Defense College’s International Law Center. He recently contributed to the book Regional Organizations and Peacekeeping: Challengers to the UN? (available here). Erik thanks Hayder Al-Shakeri (National Program Officer to Sida in Bagdad) for his help with media surveillance.
On the 9th of April 2014 and throughout the days leading up to the national elections of the 30th of April, the Iraq Army deployed troops, in a series of battles, to secure a dam from insurgent control near Falluja. The armed group known as the Islamic State of Iraq and the Levant (ISIL) had taken positions and fortified the area of Nuaimiyya, surrounding the dam, since February 2014. The ISIL’s control of the dam, controlling the water flow of the River Euphrates, resulted in the flooding of the lands upstream, possibly flooded the Abu Ghraib area in Bagdad, and lead to the exacerbation of the water shortages downstream.
Given that the current state of law regarding the protection of the environment and natural resources are often forgotten or sidelined in legal debates, it is useful to consider what an event like this can tell us about logic of the constraints in the environmental protection provided by the international humanitarian law, or the international environmental humanitarian law (IEHL), applicable to an armed group in a non-international conflict? This post will very briefly touch upon the basic understanding of environmental damage that the law provides, the logic of international legal protection for the environment in armed conflict, and the difficulties in applying this protection to armed groups and internal armed conflicts.
While the nature and extent of damage caused by the attack may at first glance seem unusually far-reaching, it is shown that the environmental protection offered by IEHL is severely curtailed with regard to non-international armed conflicts (NIACs) and the high thresholds of damage envisioned under the instruments in force. It is argued that it is the anthropocentric aspects of IEHL that offer the clearest and most developed forms of protection and are likely to do so given the current trend of law.
Background to ISIL’s military campaign
The insurgent campaign against the dam was in response to ongoing hostilities in the Anbar Province. The humanitarian crisis and internal conflict in Anbar Province has reached its most intense peak of violence since 2008, and has probably escalated concurrently with national elections. The Al-Mada Newspaper reported that capture of the dam and flooding of the adjacent areas were an attempt to force the Iraqi Forces to lift their siege on the ISIL stronghold of Falluja, 5km away from Nuaimiyya. The Government’s siege of Falluja has dragged out and come to a standstill since the beginning of the year, with Iraqi Forces biding their time as an assault the city would be too costly. Throughout the fight for the dam, ISIL managed to close the valves at least twice with consequences which are set out below (see here).
Extent of damage
Reportedly, the consequences of ISIL’s campaign have been far-reaching. The dam at Nuaimiyya is one of 25 major dams in Iraq, directing the water flow of the Tigris and Euphrates (see here). Between them, the two rivers provide 98% of Iraq’s fresh water supply.
Reports (available here) indicate that ISIL undertook the campaign against the dam to purposefully damage the environment as a tactic in its fight against Government Forces. Reports of intended environmental damage can also be corroborated by other ISIL campaigns, yet seem to suggest that ISIL is willing to go beyond targeting military objectives. The ISIL attack against the Bayji oil pipeline on the 17th of April, 2014, which contaminated the freshwater supply of Karkh and Baghdad (see here) is one telling example. The burning of approximately 400 acres of wheat crops, allegedly belonging to Turkmen farmers, while also planting mines to cut off access for fire control at Baquba in the Diyala province on the 7th of May, 2014, is another example.
Allegedly, the subsequent water shortage affected hundreds of thousands of civilians in cities downstream of the dam (see here). The downstream major riparian cities include Karbala, Najaf and Diwaniyya cities with approximate populations of 1 million, 700,000 and 400,000 respectively (see here). News sources also reported that the water shortage additionally affected the downstream energy supply, nearly halving the generation of electricity.
There has been little reporting of the effects of the flooding around Falluja itself. However, since 14 April the UN Assistance Mission for Iraq (UNAMI) has reported the flooding of Abu Ghraib, Bagdad. UNAMI reported (here and here) on the Abu Ghraib flooding that:
12,000 families have lost their crops, livestock and were constrained to be displaced inside and outside Abu Ghraib, while another 11,000 families were found at risk, some of them starting to flee, with urgent need for food, water tanks, hygiene kits and health care.
UNAMI further reported on a two-day power outage in Falluja as a result of the floods.
The severity of the insurgent campaign was further aggravated by the humanitarian situation in the region. Falluja and the Anbar Province have suffered the bulk of the internal strife between ISIL and the Iraqi Forces. The Iraqi fresh water infrastructure is thus suffering the strain of internally displaced persons from the conflicts, Syrian refugees, and displaced persons from various natural floods along the rivers (see here and here).
The notion of environmental damage in humanitarian law
Environmental Law instruments, especially those pertaining to liability, often envision damage through the relationship of specific agents acting to create measureable impact on human interest connected to the environment, i.e. property, health, and economic interest or costs of measures of prevention and reinstatement/restoration of damaged environment.
While generally not specifying the definition of damage as a matter of a specified agent inflicting damage on certain subjects, international humanitarian law approaches general environmental damage through the prohibition of certain means and methods of warfare, with key suggestive subjects incorporated.
Within the Additional Protocols to the Geneva Conventions, this general approach has only been adopted in Additional Protocol I (API), and was left out of Additional Protocol II. Article 35(3) API states:
It is prohibited to employ methods or means of warfare which are intended, or may be expected, to cause widespread, long-term and severe damage to the natural environment.
Article 55(1) presents a rationale for the prohibition of environmental damage relating to the health and survival of the population:
Care shall be taken in warfare to protect the natural environment against widespread, long-term and severe damage. This protection includes a prohibition of the use of methods or means of warfare which are intended or may be expected to cause such damage to the natural environment and thereby to prejudice the health or survival of the population.
In addition to these explicit provisions, international humanitarian law also implicitly protects the environment by certain further provisions which are examined below.
The 1976 Convention on the Prohibition of Military or Any Other Hostile Use of Environmental Modification Techniques (ENMOD) goes further in the prohibition against certain methods of warfare. The innovation it introduces to international humanitarian law is the concept of environmental modification techniques, as set out in its Article I:
Each State Party to this Convention undertakes not to engage in military or any other hostile use of environmental modification techniques having widespread, longlasting or severe effects as the means of destruction, damage or injury to any other State Party.
And Article II of ENMOD provides a suggestive list of the manipulation that ‘environmental modification would entail:
As used in article I, the term “environmental modification techniques” refers to any technique for changing – through the deliberate manipulation of natural processes – the dynamics, composition or structure of the Earth, including its biota, lithosphere, hydrosphere and atmosphere, or of outer space.
The logic of environmental protection in non-international armed conflict
It is often noted that a main reasoning behind environmental protection in humanitarian law falls within its greater anthropocentric objective, to protect human interests. The overall relevance of anthropocentric protections, given the specific impact that the insurgent campaigns around the dam has had on civilian populations seems easy to adduce in this case from provisions of international humanitarian law which although not explicitly referencing the environment, seem to have a concern for the environment within their scope.
This logic is firstly evident in provisions stemming from or related to the Additional Protocols to the Geneva Conventions involving distinction between civilian and military objectives (see ICRC’s discussion here and §179 ICJ Nuclear Weapons case). Although Iraq is not party to Additional Protocol II (APII), some key provisions from the Additional Protocols nevertheless appear to have gained customary status with regard to NIACs. Article 8(d)(i) of the Rome Statute further provides a basis for individual criminal liability for directing attacks against civilians.
More specifically, the go-to-prohibition in the case of attacks against dams is the prohibition of attacks against works and installations containing dangerous forces (Articles 15 APII and 56 API) causing “severe losses among the civilian population.” The article is clearly concerned to protect installations withholding natural forces or forces that could act on the environment to affect human life (such as through water, pollution or radioactive material).
The prohibition, as envisioned for non-international conflicts, in APII extends even to dams that constitute military objectives, as it likewise does in API. However, the ICRC Customary International Law study finds that the norm applicable to both NIACs and IACS is consistent with Article 56 of API, in which the prohibition ceases when a dam is used for a function “…other than its normal function and in regular, significant and direct support of military operations and if such attack is the only feasible way to terminate such support”.] This analysis is criticized by Sivakumaran for failing to recognize that the APII provision is absolute and favored by several military manuals in the case of NIACs.
Another relevant anthropocentric provision is the prohibition against “[s]tarvation of civilians as a method of combat” as enshrined in Articles 14 APII and 54(1) API. This provision is specifically linked to the prohibition of the attack, destruction, removal or rendering useless of objects indispensable to the survival of the civilian population. The fact that environmental protection was conceived to be relevant to this provision is evident in the included list of indispensable objects, i.e. “foodstuffs, agricultural areas for the production of foodstuffs, crops, livestock, drinking water installations and supplies and irrigation works.”
The phrasing “attack, destroy, remove or render useless objects indispensable to the survival of the civilian population” in Article 14 of APII and Article 54(2) of API suggests that most, if not all, methods of attack likely to result in the starvation are prohibited. However, it is uncertain whether the customary norm follows the reasoning of API which prohibits attacks with the effect of starvation regardless of motive, or the reasoning of AP II, which only prohibits attacks carried out for the purpose of starvation.
If custom follows the provision in API, it may be hard to argue that the campaign against the dam in Fallujah falls within this rule because the intent behind the ISIL’s military campaign against the Nuaimiyya dam with regard to the civilian population may be difficult infer. Human Rights Watch’s study on the starvation of civilians indicates that “[p]articularly relevant to assessment of intention is the effort the attacker makes to comply with the duties to distinguish between civilians and military targets and to avoid harming civilians and the civilian economy.” Intent could thus be inferred by the group’s failure to use less discriminate means. The simultaneous attacks on the oil pipeline and other attacks involving the burning of crops seem suggestive of ISIL’s intent, or at the very least willingness to employ methods of starvation as a tactic.
Finally, amongst these provisions, Article 13 APII on civilian protection might also be included. This includes the positive obligation of “general protection of the civilian population arising from military operations” (see here) and the prohibition of “[a]cts or threats of violence the primary purpose of which is to spread terror among the civilian population…” (see here).
It is clear that the campaign against the dam was carried out as an insurgent tactic by a group affiliated with Al-Qaeda. However, the difficulties of defining terrorist acts under International Law and Humanitarian Law generally, makes the application of the prohibition against acts of terrorism (Article 4(d) APII) to attacks with environmental impact or attacks utilizing or redirecting natural forces uncertain. While it may be presumed that ISIL were positioned to comprehend that weaponizing the Euphrates would have indiscriminate effects, and have at other occasions shown that they don’t care to discriminate, ISIL’s exact purpose behind this attack remains uncertain. Reports merely indicate, and logically so, that one aim was to displace Iraqi Forces around Falluja.
The ICRC study of previous examples seems to suggest, illustrating with cases from the Yugoslavian conflict, that the rule prohibiting terror has been interpreted as applying to deliberate indiscriminate tactics in general, and acts relating to war crimes such as rape and torture. In the case of Yugoslavian shelling of civilian areas lacking clear military objective, the deliberateness was easier to ascertain however. In the Nuaimiyya campaign with the presence of military objectives in Falluja, the criterion of deliberateness as an indicator of purpose appears more circular, especially with regard to primary purpose.
Environmental Protection per se
The second rationale for environmental protection in international humanitarian law is the protection of the natural environment in its own right. The exclusion of this logic from provisions within AP II, and the scope of the instruments in which provisions generally do appear cast some doubt on their applicability to NIACs. However some arguments may be made for their analytical validity.
Firstly, environmental provisions are reiterated preambularly within instruments relevant to NIACs. Secondly, many of the means and methods that would result in the damage envisioned by these provisions would also compromise human interests as described above. Thirdly, the ICRC makes the argument that given current trends in international humanitarian law, it is plausible to believe that this norm will eventually develop applicability for NIACs. So what uncertainties can we expect in such a development of the law?
One of the most widely incorporated mechanisms for defining the scope of environmental damage is the tripartite provision relating to ‘widespread, longlasting and/or severe’ environmental damage (see Article I of the ENMOD, Article 35(3) of API and 55(1) of API, above).
In Articles 35(5) and 55(1) of API, these three requirements are cumulative but in the ENMOD the ‘and/or’ wording makes clear that they are subtractive. Additionally, Article 8(b)(iv) of the Rome Statute provides a basis for individual criminal responsibility relating to the cumulative standard. However, Article 8(b) applies only to international armed conflicts.
The scope of damage under the cumulative test, deriving from Articles 35(3) and 55(1) API and the subtractive test deriving from Article I of the ENMOD, arguably varies significantly. The UNEP’s legal overview on the law protecting the environment in armed conflict notes that while the terminology of the ENMOD is well defined, the scope of damage under the API is unclear. Some interpret the travaux prepartoire of API to draw a minimum extent of damage that far exceeds that of the ENMOD. While “widespread” damage under the ENMOD envisions hundreds of square kilometers, API seems to envision tens of thousands of square kilometers. The severity of damage is linked with human interests under both instruments, yet while the ENMOD extends its protection to natural and “economic resources” API seems to restrict itself to the protection of human “health and survival” (see above). Additionally while the temporal extent of the damage under the ENMOD test for “longlasting” damage is measured within months or a season, the respective API test might envision decades. That being said, other commentators suggest the two standards might be the same. Wyatt notes that:
the fact that no higher threshold was provided in materials associated with either Additional Protocol I or the ICC Statute may mean that the low ENMOD standard of widespread damage has been adopted by default. Indeed, this is the interpretation given to the provision by many states’ military handbooks.
A key difference, of the ENMOD and API, not relating to the threshold of damage, is the scope of activities covered by the provisions. Both Articles 35 and 55 limit their scope of application to attacks that are “intended or may be expected” to result on environmental damage. The applicability of the ENMOD on the other hand is valued by the effects of an attack, rather than their deliberateness.
At present, only the cumulative tripartite threshold of damage appears to apply in relation to NIACs, as Article I of ENMOD only applies to States. Evidence that the cumulative test applies in NIACs is found in the Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which May be Deemed to be Excessively Injurious or to Have Indiscriminate Effects (CCW) that the relation to NIACs is made. The CCW preamble reiterates that “it is prohibited to employ methods or means of warfare which are intended, or may be expected, to cause widespread, long-term and severe damage to the natural environment,” while Article 1(2), as amended 2001, defines the scope to apply to situations as envisioned under Common Article 3 of the Geneva Conventions i.e. NIACs.
Should a customary rule further develop applicable to NIACs, the question of whether events like the ISIL’s military campaign against the dam in Nuaimiyya will be covered by this rule will rest on whether it is a cumulative or a subtractive test. It will also depend on the threshold of damage envisioned in those test.
It is clear that the effects of the campaign in Nuaimiyya were “severe” in terms of numbers of people and families affected by water and electricity shortages, damage to agricultural grounds and activities. The relevance to economic resources and assets is clear from the reporting of these incidents. Effects on human health and survival are less certain. It may be presumed that Iraq and the humanitarian resources mobilized within Iraq are effective enough to prevent the campaign from prejudicing the survival and health of the population. The reporting of the incidents is also ambiguous regarding the size of the areas affected, leaving the relevance to the criteria of “widespread” damage doubtful. It is uncertain whether the effects of the damage could qualify as “longlasting”. The effects of lost crops and damage to land and property may potentially affect individuals and families for months or for a whole season yet there are few situations where a temporary flooding can produce effects that span a decade.
In general, it difficult to imagine how the high threshold of damage could be achieved unintentionally. It is also unclear how much damage an organized armed group like ISIL would purposefully want to inflict on the population constituting their recruitment base. Clearly, ISIL has been willing to perpetrate indiscriminate attacks irrespective of which religious communities might be impacted. Even though it may be expected that the bulk of the water and electricity shortages were carried by the Shi’ite population downstream, any flooding and pollution damage around Bagdad is likely to affect Sunnite as well as Shi’ite families. The crop burnings were further perpetrated in Sunnite Turkmen areas. Regardless of this damage, ISIL’s existence is dependent on attracting fighters from the Sunnite communities. While an attack reaching the high threshold of damage in Iraq would generate media coverage for ISIL, the severe damage such an attack would also have within the Sunnite community would inevitably risk a hefty reduction in their potential recruitment base. If this does not deter a group like ISIL, its internal interests should. Many pathogens, pollutants or other forces with that magnitude of destruction are difficult to control. There is an overriding risk of backlash on areas or objects of personal interest to ISIL fighters. Even if a customary rule should develop, there will probably be few attacks perpetrated by organized armed groups that would reach the threshold of damage, leaving it unused.
Additionally, were a customary rule premised on prohibition of environmental damage per se develop which is applicable to NIACs according to the higher thresholds of damage, it is doubtful it could bring any added value to the humanitarian law analyzed above. Any attack reaching the high threshold of API would most likely also knowingly, if not purposefully breach the anthropocentric provisions evident within the Additional Protocols. With regard to installations withholding dangerous forces, an attack with effects qualifying under the environmental logic, would also be likely to prejudice agricultural areas under the prohibition of starvation as method of warfare, and more generally prejudice the protection civilian’s enjoy from dangers arising out of military operations.
What ISIL’s campaign at Nuaimiyya illustrates is the tall order that international humanitarian law makes with regard to environmental protection and protection from modification of environmental forces. ISIL is a relatively well-organized and well-resourced insurgent force that not only successfully perpetrated the campaign at the dam, but were successful in staving off Iraqi Forces and retaking the installation. Despite these capabilities and despite its ability to affect the dam for over a week, manipulating a natural resource of immense importance to the civilian population of Iraq, ISIL’s campaign is unlikely to measure up to the thresholds for current tests were they to become customary to NIACs.
There are few situations where organized armed groups could perpetrate attacks that affect human health and life across tens of thousands kilometers over decades. There are fewer situations still where such an attack could be carried out without a preconception of their indiscriminate effects. I imagine that an organized armed group relying on their appeal among civilians to recruit young men for their cause would rarely find attacks with that magnitude of damage a viable strategy in their own backyard. Considering the difficulty in perpetrating an attack of that magnitude in a “targeted” manner, not prejudicing the interests of key elements within the group’s own forces, the application of such a custom appears even thinner.
As a result, it is my view that it is the anthropocentric provisions of humanitarian law that conceivably relate to natural forces or natural resources preconditioning human life that provide the clearest and most extensive guidance. Given the ambiguities regarding the standards of damage in IEHL, it is conceivable that the anthropocentric provisions will continue to be the main source of that guidance, even while a norm for environmental protection in NIACs develops.
 M. Barbooti, M. Bolzoni, G. Mirza, A. Peosi, M. Barilli, L. Kadhum, G. Peterlongo, ‘valuation of Quality of Drinking Water from Baghdad’  Science World Journal 35.
UNAMI, ‘Militants burn large areas of wheat fields northeast of Baquba’  Public Information Office Media Monitoring Report on Iraq Print & Online Media Report (May 7, 2014)1, 12.
 UNAMI, ‘Fallujah suffering water and power shortage’  Public Information Office Media Monitoring Report on Iraq Print & Online Media Report (April 29, 2014)1, 6
 S. Sivakumaran, The Law of Non-International Armed Conflict (1st, Oxford University Press, Oxford 2012) 69.
 J. Pejic, ‘The right to food in situations of armed conflict: The legal framework’  RICR 1097, 1097-1103
 H. P. Gasser, ‘Acts of terror, “terrorism” and international humanitarian law’  RICR 547, 552
J. Wyatt, ‘Law-making at the intersection of international environmental, humanitarian and criminal law: the issue of damage to the environment in international armed conflict’  IRRC 593.
 Wyatt (n. 8) 624; M. Bothe, C. Bruch, J. Diamond, & D. Jensen, ‘International law protecting the environment during armed conflict: gaps and opportunities’  IRRC 569, 576.
 Wyatt (n.8) 624.