A ‘Clear’ War Crime Against the Environment?

On the morning of June 6, 2023, the first news agencies reported that the Nova Kakhovka dam in the Russian-controlled area in southern Ukraine was destroyed. The spilling water masses threatened the city of Kherson and 14 more settlements. Both Ukraine and Russia blame the other side for the dam’s destruction. What exactly caused the dam busting is a matter that has to be investigated in the upcoming months; however, early analyses indicate that Russian forces are responsible. It also does play into the Russian government’s hands, as the Ukrainian armed forces now have to deal with the humanitarian catastrophe in the area and delay the slowly launched counteroffensive (here). Besides the numerous deaths and destroyed livelihoods, this incident has also been catastrophic for the ‘silent victim’ of any war – the environment. The Ukrainian president, for instance, described the destruction as an act of ‘ecocide’. The German and British foreign secretaries referred to it as a ‘clear’ war crime.

This blogpost seeks to present the extent to which the injustice of environmental destruction can be prosecuted under international criminal law (ICL). In doing so, it identifies the shortcomings of Article 8(2)(b)(iv) Rome Statute (RS) and explains why prosecution under this provision will not be successful but could still be important for the development of ICL. Subsequently, an alternative approach under Article 7 RS is presented, which seems more promising.

The Only (War) Crime That Protects the Environment Directly

The RS enshrines exactly one ecocentric crime, which is Article 8(2)(b)(iv). It prohibits “intentionally launching an attack in the knowledge that such attack will cause […] 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.” Historically, the wording of this norm is reminiscent of Articles 35 and 55 Additional Protocol I to the Geneva Conventions (AP I) and Article 1 of the Convention on the prohibition of military or any hostile use of environmental modification techniques (ENMOD). The main difference between the AP I and ENMOD Articles is the cumulative standards of widespread, long-term and severe damage in AP I that Article 1 ENMOD does not include. Similar to those IHL rules, Article 8(2)(b)(iv) RS requires in its actus reus a widespread, long-term and severe (w-l-s) environmental damage, and is heavily criticized for these high standards, which are arguably the reason that no criminal has ever been convicted of violating these provisions (Lawrence, Heller, p. 75). Since neither the Rome Statute nor the Elements of Crime (EoC) define these notions, let us approach their interpretation by considering their (historical) context.

As Article 8(2)(b)(iv) RS has historically been derived from AP I and ENMOD, definitions could possibly be found in these conventions. While the travaux préparatoires of AP I define ‘long-term’ as “damage that lasts for decades” (Lawrence, Heller, p. 73),the text of AP I itself does not provide any explicit definitions.. When discussing widespread and severe damages, scholars refer to the definitions given by the ENMOD understandings: widespread is understood as “a scale of several hundred square kilometers” and severe as “significant disruption or harm to human life, natural and economic resources or other assets”. Most ICL scholars endorsed those definitions (e.g., Low, p. 433 or Ambos p. 426). Although the dam’s destruction arguably meets these requirements (widespread, long-term and severe damage), it is not set in stone that the ICC will adopt these definitions in any potential proceeding; or whether it might define them in a different way.

Furthermore, this legal uncertainty presents a problem with mens rea, creating a ‘circular argument’. Under Article 30(3) RS, there must be an “awareness […] that a consequence will occur in the ordinary course of events”, which was later concretized to require virtual certainty (Bemba Confirmation Decision, para. 362). A perpetrator cannot have known something ‘virtually certain’ that was not even defined by the Statute or case-law beforehand.

Secondly, the available data on forecasting environmental damage by attacks over years (or decades) have in the past been poor, even for experts, as has been seen in the example of the Gulf War (Schmitt, p. 59). One of the main reasons why the burning of oil wells in Kuwait 1991 were not described as violations of Articles 35, 55 AP I was that scientists were hesitant to make predictions about whether these damages will (definitely) last for decades (Weinstein, p. 708). It is then hardly possible to expect perpetrators to ex-ante anticipate the (long-term and widespread) consequences of an attack.

Thirdly, even if these standards were met, Article 8(2)(b)(iv) RS requires a proportionality test. The environmental damage has to be ‘clearly excessive’ in relation to the military advantage anticipated. There are no standards to weigh military advantage against environmental destruction in ICL yet. This puts the final nail in the coffin of adequate environmental protection under Article 8 RS – in the words of Lawrence: the provision is a ‘virtual nullity’ (Lawrence, Heller, p. 73).

Notwithstanding the shortcomings that Article 8 RS reveals concerning environmental protection in this instance, it remains to be seen how the Office of the Prosecutor (OTP) will act in the case of a potential prosecution regarding Nova Kakhovka. In its 2016 policy paper (p. 14), the OTP stated – well aware of these shortcomings-  that it intends to focus among other things on environmental crimes. It could use the proceedings to allow judges to interpret the w-l-s standards, thereby creating legal certainty in a conclusive manner. This unique opportunity should not be missed. Still, the resulting issues for the circular argument in mens rea and unknown standards to properly undertake the proportionality test will make prosecution under Article 8 RS extremely difficult.

Environmental Destruction as Methods of Crimes Against Humanity

In response to the focus on environmental crimes proclaimed by the Office of the Prosecutor (OTP), some authors advocate a ‘green criminology’ approach e.g. Freeland or Killean). This approach tries to integrate environmental components into the remaining core crimes of the RS, even though they do not explicitly mention the environment, e.g. Article 7 RS :the Crimes against Humanity (CaH).

The idea behind the ‘greening approach’ of Article 7 RS is to understand environmental destruction as a means and method to achieve the objectives of CaH. For instance, the water masses pouring out of the destroyed dam could be seen as a means to fulfill Article 7 RS. A similar linkage was already drawn in the Al-Bashir Case (p. 56), where the perpetrator was charged with genocide under Article 6(c) RS, for inter alia ordering the deliberate destruction and poisoning of water sources. The Pre-Trial Chamber mentioned these conducts in its summary of evidence regarding genocide, which has also been described as the “genocide-ecocide-nexus”.

The CaH consist of two components. First, a “widespread or systematic attack” on the civilian population must be proven. If this is established as a contextual element, the crimes listed in Article 7(2) RS must be fulfilled in combination with a link to that context. To be a “widespread or systematic attack”, the attack must be directed against a quantitatively large number of victims (“widespread” – Ambos, p. 159) and/or must be subject to a “systematic” plan or policy of the executing group (usually the government) against the civilian population (Ambos, p. 160). The specific conduct (dam-busting) needs to stand in line with other ‘systemic’ attacks against the civilian population. (Lambert, p. 723). Statements by the Russian leadership and the numerous documented attacks (here, here or here) on civilian objects by the Russian army suggest such an ‘attack’.

Among the crimes mentioned in Article 7(2) RS, the crime ‘forcible transfer or population’ (Article 7(2)(d) RS) is particularly relevant for the Nova Kakhovka destruction. This crime is defined as “forced displacement […] by coercive acts […] without grounds permitted under international law” (Article 7 (2) (d) RS). The act must carry a certain degree of coercion and must lead to displacement. According to Ambos (p. 190) and Lambert (p. 727), environmental destruction is such a form of a ‘coercive act’. Furthermore, over 2700 people have already fled the Kherson area in the aftermath of the dam-busting. Regarding mens rea, the OTP only need to have proven that the perpetrator had ‘knowledge’ of the systematic attack (contextual element) and fulfilled the regular standards of Article 30 RS regarding the crime of ‘forced displacement’. The perpetrators must have wanted to destroy the dam (this is to be assumed according to the current state of facts) and either must have meant to cause the consequence (people fleeing) or at least must have been aware that it would occur in the ordinary course of events, which is defined as ‘virtually certain’ (see above). The subjective evidence that the perpetrators were aware that a large number of people would flee from the surrounding area if the dam was destroyed, seems relatively easy to provide.

Conclusion

Although the destruction may have barely estimable consequences for Ukraine and its environment, the RS does not appear to have a forthright response to ecological destruction during armed conflict. Even though the actus reus requirements of Article 8(2)(b)(iv) RS may have been met, its definitional deficits present an almost insurmountable obstacle to individual prosecution. However, a conviction will most likely fail because of the mens rea and proportionality test requirements. The injustice could indirectly be met by Article 7 RS, particularly the crime of ‘forcible transfer of population’, even though this crime was not meant to address environmental destruction in the first place. The requirements of this crime seem to be fulfilled and comparatively easy to prove. Thus, the destruction of the environment by the blasting of Nova Kakhovka could still be accounted for in ICL. Notably, this protection only exists via the indirect link to humankind – a genuine and practical ecocentric prohibition does not exist in the Rome Statute (Minkova, p.68) yet. Should one wish to change this, it would be quite fruitful to consider the introduction of Ecocide to the RS.

The “Bofaxe” series appears as part of a collaboration between the IFHV and Völkerrechtsblog.