2.1. The environmental gap in international criminal law

By way of a beginning, it is important to note that this article shares some common ground with the Panel's position. Most fundamentally, it agrees that ecocide would make a welcome addition to international criminal law. As Alberro and Daniele explain, ecocide could allow action to be taken against corporate executives accused of ‘driving the mass deforestation of Indonesia to produce palm oil, threatening species like the orangutan, while leaders like Brazilian president, Jair Bolsonaro, could potentially be prosecuted for the assault on the Amazon forest’.Footnote ²³ Similarly, as Higgins hoped, by imposing individual criminal responsibility in this context, ‘the cycle of destruction and accrual of silent rights (the right to pollute, the right to destroy) will die’.Footnote ²⁴ For Liana Minkova, the criminalisation of ecocide would stigmatise environmental atrocities as ‘socially unacceptable … and instil norms of appropriate behaviour’.Footnote ²⁵ Likewise, for Donna Minha, part of the importance of ecocide lies ‘in its symbolic value [and the] transformative nature inherent in declaring certain behavior as criminal’.Footnote ²⁶ The author shares these sentiments and the view that the existence of ecocide would act as a key deterrent against, and punishment for, serious environmental damage.

Unfortunately, as it currently stands, international law is a patchwork when it comes to environmental protection, and it is very easy for malign behaviour to slip through the gaps in its stitching. This is because international law recognises destruction of the environment as a crime only when there is some added element – it does not recognise environmental destruction in isolation. The Panel was aware of this and that none of the existing crimes provide a conceptual basis for ‘purely environmental damage’.Footnote ²⁷ The existing crimes in this context are ‘war crimes’, ‘crimes against humanity’ and ‘genocide’ (the crime of ‘aggression’ is not relevant here). Each will now be considered briefly in turn to demonstrate why they do not criminalise purely environmental damage and thus why, collectively, they create a lacuna in international law that might be filled by ecocide.

2.2. The environmental war crime

First, under the Rome Statute, it is a war crime to ‘intentionally [launch] an attack in the knowledge that such attack will cause … widespread, long-term and severe damage to the natural environment’ (the environmental war crime).Footnote ²⁸ This language aligns with the text found in Additional Protocol I to the Geneva Conventions (Additional Protocol I) which provides 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’.Footnote ²⁹ Of course, Additional Protocol I imposed a substantive international humanitarian law obligation on states, whereas the Rome Statute sets the international criminal law sanction for individuals. Still, the underlying idea was the same.

The inclusion of the environmental war crime in the Rome Statute was a positive step. However, it has two key limitations. The main limitation is that, as the text of Article 8(2)(b)(iv) makes clear, its scope is limited to ‘international armed conflicts’ between states and it does not stretch to ‘non-international armed conflicts’ (NIACs) between states and non-state groups or between two or more non-state groups. Obviously, environmental destruction is not limited to armed conflicts. As Minkova observes, ‘much environmental damage occurs outside of international armed conflicts and during peacetime’.Footnote ³⁰ Similarly, as Mark Drumbl notes, environmental destruction is not limited to armed conflicts and can occur through activities such as reckless management of nuclear reactors.Footnote ³¹ Even when there is an armed conflict, for Falk, the types of environmental protection are of a ‘manifestly incidental character’ given that international humanitarian law is concerned primarily with mitigating harm to civilians.Footnote ³² Furthermore, the fact that the environmental war crime does not apply in NIACs is problematic because, as Nils Melzer notes, urbanised NIACs have become increasingly common in recent decades.Footnote ³³

A further limitation of the environmental war crime is that the destruction is criminalised only if it ‘would be clearly excessive in relation to the concrete and direct overall military advantage anticipated’.Footnote ³⁴ In other words, an individual who has carried out environmental destruction – regardless of its extent – is exculpated if the damage was nevertheless proportionate in the context of the military goal. In this sense the environmental war crime is contingent on the anthropocentric consideration of military gain and so it is not ‘genuinely ecocentric’.Footnote ³⁵ Thus, counterintuitively, there is not much protection for the environment to be found under the banner of the environmental war crime.

2.3. Crimes against humanity

Second, individuals can be brought to account for damage to the environment when that damage amounts to a crime against humanity. Crimes against humanity are acts such as murder, extermination or torture, ‘committed as part of a widespread or systematic attack directed against any civilian population, with knowledge of the attack’.Footnote ³⁶ From a prosecutorial point of view, an advantage of using crimes against humanity over the environmental war crime as the basis for a charge is that it does not require an armed conflict to be raging to be available. Indeed, as Minha put it, the category of crimes against humanity is probably the most appropriate for the prosecution of peacetime environmental crimes, ‘since it does not require any nexus to an armed conflict’.Footnote ³⁷

However, context is still important for crimes against humanity and acts will fall under this heading only if they occur against the backdrop of an attack ‘directed against any civilian population’.Footnote ³⁸ As Kevin Heller notes, this contextual element of crimes against humanity is ‘not particularly well-suited to a crime that focuses on environmental destruction’.Footnote ³⁹ Take, for example, the crime of extermination.Footnote ⁴⁰ At first blush, one might imagine that an act of environmental destruction that kills many people could fall under this heading. However, the Rome Statute explains that extermination means ‘the intentional infliction of conditions of life, … calculated to bring about the destruction of part of a population’.Footnote ⁴¹ Clearly, extermination is inherently tied up with notions of deliberately killing part of a defined group of people; however, these considerations are unlikely to be at play in the context of an alleged ecocidal event. Consequently, crimes against humanity, such as extermination, do not lend themselves to the protection of the environment for its own sake. This is perhaps unsurprising given the inherently anthropocentric notion of a crime against humanity and the fact that environmental damage would again be merely incidental to the harm caused to humans.

A further problem is that the concept of crimes against humanity is simply not envisaged as covering environmental destruction; therefore, any prosecution on this basis would be, as Minha puts it in the context of the ICC, ‘heavily dependent on the willingness of the Prosecutor and the Court to go the extra mile and expand the scope of existing international crimes to cover new forms of harm’.Footnote ⁴² Given the despairingly low number of ICC convictions secured based on the existing understanding of the Rome Statute – ten according to the ICC itselfFootnote ⁴³ – it seems unlikely that the Prosecutor is looking to interpret his remit more broadly to include anything other than the most egregious crimes perpetrated directly against humans.

2.4. Genocide

Third, individuals can be brought to account if damage to the environment somehow amounts to genocide. Genocide involves ‘acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group’ and includes killings, preventing births and forcible transfers of children.Footnote ⁴⁴ There is precedent for the notion that environmental destruction might amount to genocide. In Al-Bashir, the ICC Pre-Trial Chamber held that there were reasonable grounds to believe that the contamination of water infrastructure was part of a genocidal policy of the former Sudanese president to inflict conditions of life calculated to bring about the physical destruction of parts of ethnic groups.Footnote ⁴⁵

However, again, Al-Bashir was concerned with environmental destruction only to the extent that it was used as a medium through which to harm humans – it was not concerned with environmental destruction by itself. When it comes to using genocide to criminalise environmental destruction that results in no attendant human harm, the prospects seem limited. As Heller put it, ‘there is nothing “group-like” about the definition of ecocide’.Footnote ⁴⁶ Moreover, ‘it would be neither desirable nor practically possible to limit the crime to the destruction of specific groups of animals or plants’.Footnote ⁴⁷ Finally, when it comes to environmental destruction, there is unlikely to be any intent to harm the environment purely for its own sake. Rather, damage to the environment is likely to be a by-product of pursuing some other objective. Thus, the lack of requisite intent seems likely to be a barrier to an environmental prosecution based on genocide in most, if not all, cases. As Kai Ambos summarised, ‘the attack on the environment is not a direct attack on a human … group and the attacker does not … act with a specific intent to destroy’.Footnote ⁴⁸

2.5. Summary

At present, the criminalisation of environmental damage in international law is contingent on some form of harm being caused to human beings. As Alberro and Daniele concluded, the protection of the environment is a ‘deeply anthropocentric’ affair.Footnote ⁴⁹ Or, as Minkova observed, the existing crimes ‘address damage that is inflicted upon human beings and that only “incidentally” harms the environment’.Footnote ⁵⁰ This is unacceptable. The environment itself deserves to be protected by international criminal law. A new international crime of ecocide would indeed fill a lacuna in the current law, and its creation is therefore necessary.

3.1. Overview

In an effort to fill the lacuna in international law in the context of environmental protection, the Panel was convened. Its proposed new international crime of ecocide is set out in its Commentary and Core Text document.Footnote ⁵¹ However, disappointingly, the Panel's vision of ecocide perpetuates the underlying problem that it was trying to address – anthropocentrism in international law – and, indeed, it extends this phenomenon into a realm that many, including the author, hoped would be freed from the selfish human mantra that all issues should be considered through the lens of the impact they have on humankind. Consequently, although there are attempts in the Panel's definition to capture ‘damage to the environment per se, independent from harm to human life’,Footnote ⁵² environmental damage is generally considered to be ecocide only if it somehow results in harm to humans. The environment receives little protection for its own sake and, if damaging the environment should somehow benefit humans, even those limited forms of protection often evaporate. To demonstrate how this lingering anthropocentrism manifests, this article will now deconstruct the Panel's blueprint for ecocide into its four principal criteria: (i) unlawful acts, (ii) wanton acts, (iii) knowledge of a substantial likelihood of environmental harm, and (iv) severe and either widespread or long-term environmental harm.

3.2. The ‘unlawful acts’ criterion

The first criterion for ecocide is that the act carried out is ‘unlawful’ (the alternative criminalising option, that the act is ‘wanton’, is discussed below). According to the Panel, ‘the qualifier “unlawful” captures environmentally harmful acts that are already prohibited in law’.Footnote ⁵³ However, for Karnavas, the Panel is ‘woefully superficial in failing to define what is lawful’ because it declines to pinpoint any particular applicable law.Footnote ⁵⁴ If lawfulness were to be tied solely to international law, this would be problematic given that – as highlighted above and as the Panel itself acknowledges – international law contains ‘relatively few absolute prohibitions, and it leaves the bulk of the protection to be formulated at the national level’.Footnote ⁵⁵ For example, as Robinson notes, ‘[o]ne of the few concrete prohibitions in [international environmental law] treaties is on “illegal traffic in hazardous waste”, but [even then] illegality hinges on the lack of national permits’.Footnote ⁵⁶ Consequently, Alberro and Daniele are probably correct to assume that the Panel's approach would, in practice, require ‘unlawfulness’ to be assessed based on what is ‘already illegal under domestic law’.Footnote ⁵⁷ Thus, for example, if ecocide were to be accepted into the Rome Statute, any alleged ‘ecocidal’ acts in France would first need to be illegal under French law before the ICC Prosecutor would even begin to consider proceedings (unless the ‘wanton’ nature of the act could be used as an alternative criminalising element).

3.3. The extension of domestic law loopholes

The Panel's reliance on domestic law has the effect of anthropocentrising ecocide in the sense that, rather than focusing on the harm to the environment, the first consideration is whether humans (or, more specifically, the states into which they have coalesced) have taken steps to criminalise that conduct. Only if the answer to that is affirmative will the prosecution typically proceed to the other criteria. This position is most unsatisfactory as it prioritises states over the environment and leaves protection of the latter largely at the mercy of the internal rules of the former. Thus, as Minkova notes, relying on unlawful acts would be of limited utility because, at the domestic level, ‘unfortunately, most environmentally harmful acts are lawful’.Footnote ⁵⁸ Humans, specifically governments, would be able to serve as gatekeepers to ecocide and could dilute their domestic law – assuming it was not diluted already – to ensure that even the most environmentally damaging acts would not satisfy the Panel's unlawfulness criterion. This is perverse. After all, a key motivation for an international crime of ecocide is to bridge the accountability gaps that exist (accidentally or by design) in domestic criminal laws that govern environmental protection. Ultimately, if the Panel's position is adopted, the sad consequence would be to import domestic law loopholes into international law.

3.4. The Panel's ‘global north vs global south’ justification

According to Voigt, the panellists were aware of the problems associated with situations where actions that are lawful under national law can nonetheless result in severe, widespread or long-time environmental harm.Footnote ⁵⁹ However, she went on to explain that the Panel was reluctant to ‘leapfrog’ to the conclusion that environmental damage amounted to ecocide regardless of whether it was legal or not in domestic law.Footnote ⁶⁰ Oddly though, this discomfort with leapfrogging national law was not prompted by a desire to ensure that individuals would be held to the same consistent standard in domestic and international law. Rather, the Panel hints that it was prompted by a fear that such leapfrogging may prejudice the global south where states generally have looser environmental protection laws. The Panel's logic was that, without this limiting mechanism, ecocide would represent a proportionately greater tightening of the criminal law in the global south than it would in the global north.Footnote ⁶¹ According to Voigt, to make up the shortfall left by this choice, wantonness was added as an ‘additional “unjustifiability” threshold’.Footnote ⁶²

Accommodating the global south by making special provision for developing countries is not unprecedented in international law, despite the fragmentation and complexity it generates. Indeed, the notion of ‘common but differentiated responsibilities’ has been a key theme in international environmental law for some time.Footnote ⁶³ It helps to ensure that our shared dependencies on, and interests in, the condition of the planet are recognised, while also ensuring that developing states are not subjected to environmental requirements that would unfairly inhibit their economic growth. After all, the developed countries that industrialised during the nineteenth and twentieth centuries were able to pollute their way to growth in an entirely unfettered manner.

Nonetheless, despite the arguable merit and fairness of a ‘differentiated’ approach, the regard paid to the developed or developing status of states is further evidence of the anthropocentrism that pervades the Panel's work. The environment does not distinguish between the (arguably more justifiable) harm caused by the global south and the (arguably less justifiable) harm caused by the global north. Damage is damage, no matter which person or state caused it to be inflicted. Moreover, there is every reason to believe that the global south stands to be more severely impacted by environmental degradation than the global north, so it is in the interests of developing countries to be especially tough on environmental degradation. For example, according to the Environmental Investigation Agency (a UK- and US-based NGO), Pacific Island countries ‘contribute less than 1.3 per cent of the mismanaged plastics in the world's oceans yet are one of the main recipients of plastics pollution and its impacts’.Footnote ⁶⁴ If it seems that this line of thinking would place a disproportionate burden on developing states, recall that much of the environmental damage caused within their borders is done by powerful corporations. Many of these originate in the United States and China where, as Karnavas notes, protecting the environment ‘is calculatedly given short shrift’ in comparison with maintaining, or gaining, strategic superiority.Footnote ⁶⁵

Ultimately, tighter ecocide rules across the board (in the global north and global south) are the only way to tackle multinational polluters. If the Panel's ‘unlawful act’ criterion is retained, those corporations will be able to relocate destructive operations to states that adhere to the lowest common denominator in terms of environmental regulation. Of more concern still, the criterion could well prompt a race to the bottom in the global south (as we already see in the case of labour, taxation, corporate transparency, and so onFootnote ⁶⁶), the regimes of which – often in desperation – compete for investment from those malefactor multinationals. If that unintended consequence were to manifest, the effect of the Panel's proposal could be to degrade domestic environmental protection laws across the global south. This would be a deeply disturbing outcome. It is a possible legacy that the Panel has not envisaged and, as will now be discussed, one for which the alternative criminalising element of ‘wantonness’ cannot act as a cure.

4.1. The definition of ‘wanton’

According to the Panel's proposal, even if environmentally destructive acts are lawful at the domestic level, they may nevertheless qualify as ecocide if they were ‘wanton’.Footnote ⁶⁷ The Panel stated that wanton means ‘with reckless disregard for damage which would be clearly excessive in relation to the social and economic benefits anticipated’.Footnote ⁶⁸ It should be emphasised again that the wantonness criterion is an alternative criminalising element to ‘unlawfulness’: it is the means by which the Panel hopes that domestically legal acts could become internationally criminal acts. This alternative criterion was added by the Panel to fill the gap that would emerge if only domestically unlawful acts could qualify as ecocide.Footnote ⁶⁹ However, its addition undermines the edifice of ecocide by providing another conduit for anthropocentrism. After all, considerations of social and economic benefits are inherently anthropocentric. Moreover, the criteria for wantonness set a very high threshold for prosecutors to satisfy should they need to rely on it. By including reference to anthropocentric issues, and by setting such a high bar for wantonness, the Panel has added a convenient escape hatch into its ecocide blueprint for potential ecocidaires.

4.2. Justifications for the wantonness test

According to the Panel, it included the wantonness test because not all environmentally damaging acts are ‘illegitimate, or even undesirable [and] international [law] must permit legitimate development’.Footnote ⁷⁰ It gave examples of ‘socially beneficial acts, such as housing developments and transport links’.Footnote ⁷¹ In other words, when assessing whether an act that causes environmental damage (that is lawful under domestic law) might qualify as ecocide for the Panel's purposes, one must consider the benefits to human society that the act brings. For Robinson, the Panel was correct to include this cost-benefit orientated wantonness calculation. He acknowledges that ‘we need greener energy, transport, products, and food, as well as less consumerism [and that] an incorrectly blunt criminal law will not achieve that societal transformation … nor would it attract a critical mass of support’.Footnote ⁷² He supports this with reference to food production, which has a ‘major ecological footprint’ but nevertheless remains essential for everyone.Footnote ⁷³ It is certainly true that environmental damage is usually a by-product of some other, societally desirable, process. However, this does not mean that subjecting ecocide to an anthropocentric cost-benefit analysis is the answer.

4.3. Problems with the wantonness test

There has been much academic criticism of the inclusion of the cost-benefit orientated wantonness criterion in the definition of ecocide. According to Heller,Footnote ⁷⁴

Ambos expresses the same sentiment:Footnote ⁷⁵

For Alberro and Daniele, the inclusion of wantonness implies that it is acceptable to cause environmental harm ‘as long as the damage is not “clearly excessive” in relation to the anticipated benefits for humans’.Footnote ⁷⁶ For Minkova, the incorporation of wantonness contradicts ‘a message to the broader international audience that harming the environment is wrong and punishable’.Footnote ⁷⁷ Taken collectively, these concerns boil down to the observation that the Panel's inclusion of the wantonness criterion means that its vision of ecocide, rather than being a watershed moment for the protection of the environment, simply perpetuates the status quo and subordinates the environment to human wants and needs. This is precisely the well-entrenched cultural attitude that ecocide was supposed to break, or at least bend, and therefore the inclusion of wantonness seems to compromise the Panel's design.

Moreover, the inclusion of wantonness is problematic in practical terms. As Minkova observes, the benefits drawn from activities that harm the environment are often ‘more tangible than the costs which may occur years later’.Footnote ⁷⁸ This is surely correct. For example, there is an immediate boost to amenity when a new housing block is opened, and there is an immediate boost to the profitability of industry when transport networks are expanded, allowing faster transit of goods. On the other hand, the impact on the environment caused by redirecting waterways to build that housing block, or by clearing woodlands to construct those networks, are less noticeable to humans – at least in the immediate sense. One might call this the ‘perceptibility problem’ and it has two consequences. First, the benefits of environmental destruction are given disproportionately high weighting as they can be immediately appreciated. Second, the costs of environmental destruction are given disproportionately low weighting because they may not emerge for years and, even when they do, they are more likely to have an impact on flora and fauna than on humans. In certain cases the impact of environmental degradation may even be felt in an entirely different country: consider, for example, the issue of plastic pollution mentioned above. Plastics are used in a plethora of industries, which include construction, one of the most polluting industries of all. For the state in which buildings are constructed, the benefit of having new housing, offices or schools is immediately apparent in boosting the economy and living standards. The costs, however, are diminished or entirely invisible within that state if the tonnes of discarded plastic used in construction are dumped in the sea or exported to be burnt.Footnote ⁷⁹ As touched on earlier, an egregious aggravator here is the fact that the polluting state is more likely to be in the global north, while the state that suffers the pollution is more likely to be in the global south – in Africa or the Pacific Islands.Footnote ⁸⁰ In cases such as these, even if an anthropocentric approach was seen to be desirable, it is hard to see how an accurate assessment of the costs and benefits to humanity can be made in practical terms. Against this backdrop, it may be difficult or even impossible to determine with much certainty whether a given act was wanton or not. Thus, the wantonness criterion adds a gaping loophole through which environmental malefactors may slip.

5.1. Overview

Assuming it is possible to find a criminalising element in the actions of a putative ecocidaire – either because the act was unlawful under domestic law or because it was wanton – the next requirement is that the act creates a substantial likelihood of ‘severe and either widespread or long-term damage to the environment’.Footnote ⁸¹ The Panel expressly borrowed the impact indicators of ‘severe’, ‘widespread’ and ‘long-term’ from treaties, including Additional Protocol IFootnote ⁸² and the Environmental Modification Convention (ENMOD Convention).Footnote ⁸³ In many ways this is the core of ecocide in the sense that this is the part of the definition that focuses most on the environment itself and the destruction that a particular accused is alleged to have caused. If any part of the Panel's text could be expected to be ‘ecocentric’ then this would be it and, indeed, this expectation is borne out to some extent. Nonetheless, even here there is a distinct undercurrent of anthropocentrism. As above, this serves to direct the flow of the Panel's ecocide narrative away from its intended destination. For present purposes, this impact-oriented aspect of the Panel's proposal will be deconstructed into four components: (i) its semi-conjunctive formulation, (ii) the ‘severe’ criterion, (iii) the ‘widespread’ criterion, and (iv) the ‘long-term’ criterion.

5.2. Semi-conjunctive formulation

One of the issues the Panel had to resolve was how to formulate the severe, widespread and long-term tests. The Panel observed that ‘ENMOD uses the disjunctive (“widespread, long-lasting or severe”) [whereas Additional Protocol] I and the Rome Statute use the conjunctive formulation “widespread, long-term and severe”’.Footnote ⁸⁴ For its own formulation, the Panel chose to require that the acts must cause ‘severe and either widespread or long-term damage to the environment’.Footnote ⁸⁵ Thus, the Panel's position represents a ‘mid-point’ between the existing formulations by combining the ‘disjunctive’ and ‘conjunctive’ approaches (hence it is referred to here as a ‘semi-conjunctive’ formulation).Footnote ⁸⁶ Voigt reasoned that while a fully conjunctive threshold is appropriate when assessing environmental harm caused during armed conflict, it would be ‘too high’ for (peacetime) ecocide; therefore, in addition to being severe, it was sufficient that the damage was either widespread or long term.Footnote ⁸⁷

The academic response to the Panel's approach on this point has been generally positive. As Minkova put it, the inclusion of the ‘severe’ requirement as the only compulsory element of Article 8 ter is ‘consistent with the ICC's limited mandate, which focuses on the “gravest” crimes’.Footnote ⁸⁸ Equally, according to Heller, the fully conjunctive approach in international humanitarian law exists only in recognition of the fact that it is ‘impossible to engage in armed conflict without causing some environmental damage’.Footnote ⁸⁹ This means that the fully conjunctive test should not be carried over to ecocide as ecocide is intended for application in peacetime too, when environmental damage is easier to avoid. However, there is cause for concern here also. The Panel settled on its semi-conjunctive test based on considerations such as the emphasis in the Rome Statute on the ‘gravest’ crimes (a human classification imposed based on human sensibilities) and on whether the environmental damage occurs in peacetime or wartime (conditions factually created and legally declared by humans). This reveals further anthropocentric bias. The level of protection accorded to the environment against ecocide is framed against a contextual backdrop of human interactions and political conditions.

5.3. The ‘severe’ criterion

If we move beyond the issue of how the impact indicators are formulated together, we can consider the individual tests themselves. The first criterion (and the only mandatory one of the three) is that the damage to the environment must be ‘severe’. By way of a comparative baseline, in the context of the ENMOD ConventionFootnote ⁹⁰ interpretive guidance can be found in the work of the Geneva Conference of the Committee on Disarmament (CCD), which defined ‘severe’ (in the absence of a definition in the ENMOD Convention itself) as ‘involving serious or significant disruption or harm to human life, natural and economic resources or other assets’.Footnote ⁹¹ For its part, the Panel defines ‘severe’ as ‘damage which involves very serious adverse changes, disruption or harm to any element of the environment, including grave impacts on human life or natural, cultural or economic resources’.Footnote ⁹² In one sense, as Minkova notes, the term ‘any element of the environment’ allows ecocide to include environmental harm that ‘does not directly affect human beings’.Footnote ⁹³ This broad drafting is generally positive and could be seen to reduce the risk of an anthropocentric approach being taken to the environment.

However, the Panel's examples of severe damage – ‘grave impacts on human life or natural, cultural or economic resources’ – are most unhelpful.Footnote ⁹⁴ Admittedly, this is only an indicative list, but it is strangely anthropocentric. ‘Human life’ is inherently anthropocentric, of course. ‘Cultural or economic resources’ are anthropocentric, too, as items can be of cultural or economic benefit only to humans. Most concerning, though, is the Panel's treatment of nature or, as the Panel puts it, natural ‘resources’ and ‘assets’. The dictionary definition of a ‘resource’ is ‘a useful or valuable possession’ or ‘something that can be used to help you’.Footnote ⁹⁵ Thus, by its ordinary meaning a ‘natural resource’ includes only elements of nature that are of use to humans. The same can be said of ‘assets’ – an asset is defined as ‘something having value, such as a possession or property, that is owned by a person, business, or organisation’.Footnote ⁹⁶ Consequently, the Panel has again framed nature through a human lens and further anthropocentrised Article 8 ter. The Panel would have put ecocide on a stronger footing if its examples made no reference to the impact on humans or, indeed, if no examples had been given at all.

In setting out these examples, the Panel was probably trying to align with the definition of severity used in the ENMOD Convention. As noted above, the interpretive guidance for the ENMOD Convention refers to ‘human life, natural and economic resources or other assets’.Footnote ⁹⁷ Needless to say, though, the ENMOD Convention is nearly 50 years old and its anthropocentric approach to the environment is dated. It is a great pity that the Panel has chosen to rehearse the notion of the environment as a resource worthy of protection only if it is useful to humans. Indeed, as Minkova observes, ‘“severe” [is] problematic, as environmental damage which occurs in an isolated section of the global commons that has not yet been valued by global markets, could fall outside its scope’.Footnote ⁹⁸ Again, it must be noted that the full text of the Panel's definition refers to ‘any element of the environment, including grave impacts on human life or natural, cultural or economic resources’.Footnote ⁹⁹ The effect is that ‘non-useful’ parts of the environment could still technically fall within the scope of Article 8 ter. Nonetheless, the Panel could hardly have chosen worse examples to illustrate the aspects of the natural world that ecocide might protect. Sadly, its phrasing here could lead to gaps in protection in the future by restricting how ecocide is viewed and interpreted.

5.4. The ‘widespread’ criterion

In addition to being severe, to qualify as ecocide the damage must be either widespread or long-term. Regarding the widespread criterion, according to the CCD writing in the context of the ENMOD Convention, ‘widespread’ is interpreted as ‘encompassing an area on the scale of several hundred square kilometres’.Footnote ¹⁰⁰ In other words, the CCD set out its view of a relatively clear minimum threshold for damage under ENMOD – roughly 300 square kilometres. In the context of ecocide, the Panel chose a different tack and defined ‘widespread’ as meaning damage which extends ‘beyond a limited geographic area, crosses state boundaries, or is suffered by an entire ecosystem or species or a large number of human beings’.Footnote ¹⁰¹

Because of the way it is framed, the Panel's definition has the potential to be construed more broadly, or more narrowly, than the CCD definition. For example, the phrase ‘beyond a limited geographical area’ implicitly suggests that the threshold for extent is intended to be much lower than the several hundred square kilometres envisaged under the ENMOD Convention. This is because it is possible that even 10 square kilometres might qualify as more than a ‘limited area’. On the other hand, the example that environmental damage will be covered if it ‘crosses state boundaries’ could be read to imply that the requisite extent for ecocide might be higher than that under the ENMOD Convention. Similarly, the Panel's example that the damage would be sufficiently widespread if a ‘whole species’ was affected would seem to imply that a greater area needs to be affected than is required under the ENMOD Convention, as many species are distributed across vast areas. Having said that, different species diffuse themselves across the globe to varying extents. Some species are limited to very narrow geographical areas: for example, the proboscis monkey is found only on Borneo and the Tasmanian Devil is found only on Tasmania.Footnote ¹⁰² In contrast, some species cover much larger areas and some, such as the peregrine falcon, manage to make themselves at home on every continent except Antarctica.Footnote ¹⁰³ Given these considerations, it is hard to say in abstract how broadly, or narrowly, the Panel's definition would be construed.

What can be said, however, is that the Panel's attempt at setting the relevant threshold for ‘widespread’ has been influenced by anthropocentric considerations. Minkova is correct in saying that the Panel recognises ‘that the term [widespread] may relate not only to a large number of human beings but also to [ecosystems or species]’.Footnote ¹⁰⁴ However, the very fact that the Panel included impact on ‘a large number of human beings’ as a relevant indicator demonstrates anthropocentric bias. Moreover, the reference to whether the damage ‘crosses state boundaries’ seems to be an oddly human-orientated conception of distance as borders are often arbitrary, manmade, lines. Consider, for example, the ‘49^th parallel’ border between the United States and Canada or the ‘38^th parallel’ border between the Republic of Korea and the Democratic People's Republic of Korea. Equally, by quirks of history, some states are huge and geographically isolated (such as Australia), with the effect that environmental destruction within them is unlikely to spill over into other states; whereas some states are tiny and closely surrounded by neighbours (such as Luxembourg) so that pollution could much more easily cross their borders. More generally, as Drumbl put it, ‘environmental harm … does not know boundaries’.Footnote ¹⁰⁵ It would have been preferable for the Panel to omit reference to human-drawn boundary lines altogether and focus solely on geographical extent in providing indicators of what might constitute ‘widespread’.

5.5. The ‘long-term’ criterion

Finally, in addition to being severe, and as an alternative to being widespread, the environmental damage caused by the accused must be ‘long-term’. According to the Panel's proposal, ‘“long-term” means damage which is irreversible or which cannot be redressed through natural recovery within a reasonable period of time’.Footnote ¹⁰⁶ There is a significant prospect that ‘long-term’ could end up with an anthropocentric interpretation, too. According to Drumbl, ‘environmental harm … does not know [temporal] boundaries’.Footnote ¹⁰⁷ This is not quite correct. Rather, the difficulty is that in ecological terms, time is a relative and contextual affair as a ‘variety [of] … temporal scales … govern ecosystem processes’.Footnote ¹⁰⁸ Consequently, whether a particular phenomenon is ‘long-term’ fluctuates depending on the context. For example, animals are likely to operate on different temporal scales from plants, and one species of animal or plant will operate according to a different scale from others. Therefore, it is an oversimplification to contend that a single timeframe can be used to determine the point by which all ecological damage has had a reasonable opportunity to recover naturally. It would be wrong to argue that (i) any damage taking longer than a given period to heal is automatically long term, or that (ii) any damage taking less than that period to heal is not long term.

Even if we set aside the point above and assume that it is theoretically possible to select a single timeframe over which damage to an ecosystem could be said to recover naturally, there is the matter of determining precisely what that period is. Recall that, according to the Panel, the accused's actions will qualify as ecocide only if the environmental harm cannot recover naturally within this period, so it is a critical issue. Unfortunately, as Voigt acknowledges, the Panel gave no detail on how to interpret terms such as ‘recovery’ and ‘reasonable period’; therefore, its interpretation would be ‘in the hands of the judges … and developed on a case-by-case basis’.Footnote ¹⁰⁹ If that transpires, future courts are likely to look to the ENMOD Convention interpretive guidance for some initial illumination; yet, the CCD refers to ‘long-lasting’ (the ENMOD Convention equivalent to long-term) as meaning ‘lasting for a period of months, or approximately a season’.Footnote ¹¹⁰ Again, these are both highly anthropocentric conceptions of time. The notions of ‘months’ and ‘seasons’ are useful ways for humans to gauge the passage of time, but they may have limited bearing on ecosystems. For example, a system of waterways could be temporarily contaminated by the release of toxins from an industrial site. The contamination may clear away in two months (hence not considered ‘long-term’ for the purposes of the ENMOD Convention) but in that time it may affect the breeding cycles of fish and amphibians, or damage their systems, in ways that are hard to detect. It should not matter if the contamination did not last for a few months or a season. Given these difficulties, it would have been preferable for the Panel to attempt a more fully thought-out, ecocentric definition of ‘long-term’.

6.1. The proposed mental element for ecocide

Finally, we come to the mental element of the proposed crime of ecocide. A wide range of possible options could be chosen. At the very strict end of the spectrum, ecocide could require subjective intent to cause environmental damage. At the more relaxed end of the spectrum, ecocide could merely require objective awareness of a risk of environmental damage or, indeed, it could have no mental component at all and would thus be established on the basis of strict liability. Of course, the notion of a mental element is an inherently anthropocentric concept given that it is concerned with what a person thought, or ought to have thought, in the circumstances under review. That said, there is a degree of nuance here as requiring, for example, subjective intent (which places emphasis on what the particular accused thought) is arguably more anthropocentric than, for example, strict liability (which is not concerned with what the accused thought or even what a reasonable person in their position would have thought). Consequently, the selection of the mental element test for ecocide provides an insight into the extent of the Panel's anthropocentrism.

6.2. The default mental element test in the Rome Statute

By default, under the Rome Statute ‘a person shall be criminally responsible … only if the material elements [of the alleged crime] are committed with intent and knowledge’.Footnote ¹¹¹ Intent requires that a person ‘means to engage in the conduct [or] means to cause that consequence or is aware that it will occur in the ordinary course of events.Footnote ¹¹² Knowledge requires ‘awareness that a circumstance exists or a consequence will occur in the ordinary course of events’.Footnote ¹¹³ According to Heller, the ICC interprets ‘knowledge’ in this context very strictly and requires the perpetrator to be aware that their actions are ‘virtually certain’ to result in the unlawful outcome.Footnote ¹¹⁴ Indeed, the Panel was aware that ‘most decisions and commentators have concluded that [the default test] requires an awareness of a near certainty that the consequences will occur’.Footnote ¹¹⁵ In other words, the text of the Rome Statute and the subsequent practice of the ICC take the most rigorous approach possible in assessing the mental element. The individual accused's thoughts will be interrogated, and the outcome of the case will depend on those thoughts. This is an especially anthropocentric approach.

6.3. The Panel's proposal for the mental element of ecocide

In the context of ecocide, the Panel opted to require that the accused has ‘knowledge that there is a substantial likelihood’ of environmental damage occurring.Footnote ¹¹⁶ This appears to fix quite a high threshold for the mental element of ecocide (albeit slightly lower than the default threshold of knowledge of a ‘near certainty’).Footnote ¹¹⁷ The Panel explained that the default test ‘was too narrow and would not capture conduct with a high likelihood of resulting in [environmental damage]’.Footnote ¹¹⁸ Minkova agrees with this approach in principle, noting that ‘environmental harm is often anticipated but not necessarily intended or known with certainty’.Footnote ¹¹⁹ In truth, however, the Panel does not favour a standard based on knowledge at all. Rather, as one panellist, Voigt, put it, the assessment should be of whether there was ‘“awareness” … of the significant likelihood of serious damage in the sense of being aware or [having the potential to be] aware based on publicly accessible information and data’.Footnote ¹²⁰ Indeed, even the Panel's Commentary demonstrates that it hopes and expects that its mental element test will, in fact, be interpreted according to the lower standard of ‘recklessness or dolus eventualis’.Footnote ¹²¹

As the reader may have discerned, the Panel's position on the mental element for ecocide is confusing and contradictory. It spans knowledge of a risk, awareness of a risk, recklessness and dolus eventualis. Karnavas summarises it as a ‘confusing if not troubling … casual [conflation]’.Footnote ¹²² Ultimately, though, despite what the Panel says about the mental element of ecocide in its Commentary, the text of Article 8 ter still formally requires the accused's ‘knowledge’ of a substantial risk of environmental harm and not merely their awareness of a risk or their recklessness. Thus, it set one of the highest mental element tests available for prosecutors to satisfy. In this way, not only is there a contradiction between the draft article and the CommentaryFootnote ¹²³ but, as Minkova notes, this human knowledge-based standard ‘could have negative implications for the prosecution of ecocide at the Court’.Footnote ¹²⁴ The ICC would not be able to interpret ‘knowledge’ to mean ‘recklessness’ without violating the principle of nulla poena sine lege.Footnote ¹²⁵ Not only does the Panel's approach reduce the prospects of convictions for ecocide, it also further entrenches anthropocentrism into the concept as any test for assessing mens rea requires considering the state of a person's mind. The only exception to this would be a strict liability approach. However, that is not the option favoured by the Panel and, as Robinson says, ecocide warrants a ‘fault standard appropriate to the gravity and stigma of a serious international crime [which] would rule out … strict liability’.Footnote ¹²⁶ In short, anthropocentrism has influenced the Panel's approach to setting the mental element that the accused must possess to be convicted of ecocide.

7.1. The desire for adoption

Thus far, we have seen that the new crime of ecocide would indeed be a welcome addition to international law owing to the lacuna of environmental protection that persists. However, we have also seen that the blueprint for closing this gap offered by the Panel – the aspiring architects of the crime of ecocide – is blotted by anthropocentric bias. The effect of this bias is to undermine the proposal to such an extent that its potential to have any positive impact at all is doubtful. This begs the question: Why did the Panel take an anthropocentric approach to what ought to have been an ecocentric issue? The answer is a consideration that has stalked ecocide since its earliest days.

In 1970, Galston himself (noted above as the man who coined the term ‘ecocide’) envisaged an inherent connection between the proposed crime and its effects on mankind. In the context of irreversible damage to plant life caused by the war in Vietnam, he emphasised that, because of the interrelationship of all life in an ecological system, such damage could also have long-range effects on animals and humans.Footnote ¹²⁷ He noted, for example, that the contamination of mangroves lining the waterways of Vietnam undermined the viability of habitats for shellfish and migratory fish and went on to highlight that this could affect ‘the health and welfare of South Vietnamese who depend heavily on such marine life for protein’.Footnote ¹²⁸ In other words, when Galston articulated ecocide, its ultimate objective was the protection of human beings rather than the environment itself. This human-orientated justification for ecocide is still broadly the same as that espoused today by Stop Ecocide International, which states that ‘without a healthy Earth, there can be no healthy human beings’.Footnote ¹²⁹ However, it seems likely that environmental champions – from Galston through to the likes of Attenborough and Thunberg – have expressed environmental protection in anthropocentric terms not because they want to but, rather, because they feel they need to in order to attract people's attention and support. If they do not make the connection between environmental damage and the impact on humanity, human selfishness and apathy is likely to result in no action being taken.

The use of anthropocentrism by environmental advocates as a mechanism to ensure public buy-in is effectively confirmed in the context of the Panel's approach to ecocide. Indeed, Voigt acknowledged that the Panel's work was guided by implicit parameters, which she understood to be: ‘1. pragmatism and realism, 2. precedent, 3. deference and respect, 4. environmental integrity, and 5. legal effectiveness’.Footnote ¹³⁰ Voigt conceded that a more robustly ecocentric approach ‘could perhaps have given a stronger environmental signal but might have been detrimental to the likelihood of being adopted’,Footnote ¹³¹ and that the Panel was keen that ecocide ‘would stand a chance to be supported by state parties to the Rome Statute’.Footnote ¹³² In short then, the Panel wanted to produce a definition of ecocide that would be palatable enough for states to be adopted and implemented. Unfortunately, this meant that its proposal – which was intended to act as ecocide's blueprint for incorporation into international law – needed somehow to be compromised. The tool used to achieve the compromise was anthropocentrism. This ‘political palatability’ is why anthropocentric considerations appear time and again throughout what should be an ecocentric document. The Panel thought states might just get on board with the proposal if they were left with enough flexibility to escape its clutches in many cases. This approach is understandable given the Panel's lack of political or diplomatic clout, the absence of which meant that it could not pursue a bolder vision of ecocide. Nonetheless, anthropocentrising ecocide was a mistake.

7.2. The flaw in logic of the anthropocentric compromise

The decision of the Panel to design a politically palatable, anthropocentric blueprint of ecocide was a mistake. This is because, by compromising its design, the Panel's vision of ecocide is now unlikely to appeal to either (i) states that want to see change (the ‘progressive’ states) or (ii) states that wish to preserve the status quo (the ‘regressive’ states). It should be acknowledged before proceeding that the idea of progressive and regressive states sitting in a neat dichotomy is an oversimplification: the attitudes in different countries to environmental protection is a complex, varied and constantly evolving phenomenon.Footnote ¹³³ Nonetheless, the bifurcation is a useful shorthand for present purposes.

Turning first to the progressive states, we could say that in some corners the battle to have environmental issues taken seriously has now, finally, been won. States routinely make proclamations now about the dangers of global warming, the threats posed by the climate emergency, loss of biodiversity on the planet, pollution of the oceans with plastics, and a long list of other issues.Footnote ¹³⁴ There is widespread recognition that ‘fundamental transformations’ need to be effected to mitigate humankind's impact on the planet – even steps that cause significant inconvenience and financial cost with the potential to change how we work, travel, eat, shop, clothe ourselves, and so on.Footnote ¹³⁵ Against this backdrop, the Panel's approach will be seen as unduly conservative. Indeed, according to Voigt, the Panel was clear that its definition of ecocide ‘should not venture beyond the boundaries of legal concepts which states are familiar with in international law’.Footnote ¹³⁶ Moreover, the Panel ‘carefully analysed and drew inspiration from the jurisprudence of international courts including the ICC, ICJ, ITLOS, existing international treaty law and custom’.Footnote ¹³⁷ In other words, the Panel actively eschewed taking a more radical (and probably more ecocentric) approach and preferred instead to stick to familiar, anthropocentric standards. Again, this conservatism is understandable when one considers that the Panel was acting under the auspices of a NGO with limited international influence. Nevertheless, it will not make the Panel's proposal satisfying reading for states that are keen to see real action against plastic pollution, sewage dumping in seas, and so on. Such states will wish to see bolder proposals.

Turning to the regressive states, many powerful countries engage in environmentally destructive activities while pursuing their economic goals, with the likes of China, India, Russia and the United States being obvious examples.Footnote ¹³⁸ Those states are likely to resist change. Equally, some states in the global south are also likely to resist ecocide as it might hamper their ability to catch up economically with the developed world through the same sorts of industrialised process – a consideration of which the Panel was also aware.Footnote ¹³⁹ As a consequence, the weight of academic opinion is that the Panel's proposed ecocide amendment to the Rome Statute will not be adopted by the state parties. Ambos believes that adoption is ‘not very likely’.Footnote ¹⁴⁰ For Minha, ‘amending the Rome Statute so that it includes the crime of ecocide will not be an easy task’.Footnote ¹⁴¹ Karnavas speaks a little more bluntly, saying ‘let's not mince words … Article 8 ter is a non-starter’.Footnote ¹⁴²

One organisation where the Panel's proposal has received support is the European Union (EU). The EU Parliament has recommended that member states support the Panel's proposal and adopt ecocide as a fifth crime under the Rome Statute.Footnote ¹⁴³ However, the EU parliament is not the executive decision-making body of the EU and it does not speak directly for EU member states – certainly not in the context of the Rome Statute to which only states are parties.Footnote ¹⁴⁴ Thus, it is free to make positive noises about the adoption of ecocide in the knowledge that they do not amount to action. It is instructive that, as yet, no states have signed up to the Panel's call to add ecocide to the Rome Statute, although some states have since made progress towards criminalising ecocide in domestic law (for example, Belgium, France and Italy).Footnote ¹⁴⁵ Even if some states actively support the Panel's proposal in the future, any amendment of the Rome Statute could be frustrated quite easily by other states as a result of the treaty's rules on amendment. Any state party may propose amendments to the Rome Statute;Footnote ¹⁴⁶ however, the adoption amendments can occur only through consensus or, in the absence of consensus, through the support of a two-thirds majority of states.Footnote ¹⁴⁷ Given this threshold, it seems unlikely that the ecocide amendment will be adopted. The consequence of all this is that, given the way in which the Rome Statute works and the politics at play, the Panel's proposal is unlikely to be adopted.

If the Panel had acknowledged this reality, it would have been free to be more radical (more ecocentric) in its approach and to devise a proposal that more robustly provided for the protection of the environment. Initially, such a bold stand may have been merely symbolic. However, over time, states that favoured a rigorous approach to ecocide could have said so officially, both domestically and in international fora such as the UN, with the effect that a body of state practice could have begun to form around this tougher stance. Courts may then have begun to refer to such state practice and, in time, new customary international law might have formed around it. This is a much slower and more speculative approach than that taken by the Panel. Arguably, though, it would have produced a better way forward in the long run than the palatable but diluted proposal currently on offer.

7.3. Summary

The Panel's lack of standing and influence pushed it into producing a compromised definition of ecocide that is undermined by anthropocentrism. For states that genuinely want to see ecocide as an international crime which can hold individuals to account, the Panel's proposal is too weak. Conversely, for states that wish to preserve the status quo, the Panel's proposal remains anathema as it will present too much risk to their current operations. Either way, the Panel's anthropocentric proposal is doomed to fail. A bolder but slower route could have been taken by the Panel in the form of arguing for a more rigorous approach to ecocide in the hope of contributing to the evolution of customary international law; but the Panel chose not to take that path.

8.1. Overview

For the reasons outlined above, the Panel's blueprint should not be used to construct a new international crime of ecocide. The Panel, while acting with the best of intentions, was working from a position of weakness, and felt compelled to compromise ecocide using the tool of anthropocentrism in the ultimately futile hope that its design would be politically palatable to states. However, this does not mean that the concept of building a new crime of ecocide is wrong. After all, as discussed above, the addition of ecocide would be hugely positive by filling a large gap in the context of environmental protection. Instead, it simply means that – as eminent as the individual panellists undoubtedly are – the Panel as a collective entity was simply not well placed to act as the architect for this project.

The principal reason for this is that NGOs such as Stop Ecocide International and, by extension, the Panel, lack ‘authority’. As Karen Alter and co-authors note:Footnote ¹⁴⁸

With regard to the legal formalist approach, bodies may derive ‘authority from the act of delegation from member states’.Footnote ¹⁴⁹ The Panel benefits from no such delegated authority as, rather, it endeavours to influence states from the outside. Regarding the normative approach, power here ‘comes from a sense that the law or the institution applying it is legitimate’.Footnote ¹⁵⁰ It is unlikely that many will view the Panel has having a legitimate lawmaking capacity and, instead, it seems more likely that it will be seen as an advocacy group pursuing an agenda.

The sociological legitimacy approach requires that ‘an institution justifies its actions in the perception of audiences, co-opting or becoming part of their beliefs about legitimacy’.Footnote ¹⁵¹ Here, focus would be on the audience's reaction to the Panel's proposals rather than the institutional standing of the Panel. However, it would be difficult to properly define the Panel's ‘audience’ and even more difficult to measure accurately its reaction to its proposals – the use of surveys would be untenable – thus, authority is unlikely to be bestowed under this heading. Finally, the compliance approach involves measuring the compliance of states with the actions of different bodies.Footnote ¹⁵² However, for Alter and co-authors, ‘[c]ompliance is an insufficient measure of de facto authority because compliance can [co-exist despite] widespread rejection of that body’.Footnote ¹⁵³ For example, in the future states might well reduce environmentally destructive practices. However, this may be the result of a correlative, rather than a causative, relationship with the views of the Panel, which they may have rejected or ignored.

In short, it is very difficult to identify a solid basis of authority upon which the Panel might construct a new crime of ecocide. The question, therefore, becomes whether any alternative architects have a better, more authoritative, grounding. Three possible options will now be considered: (i) the international courts, (ii) the drafters of legal manuals, and (iii) the ILC.

8.2. The international courts

One alternative mechanism for the creation of ecocide begins with Minha's view that the crimes already enumerated in the Rome Statute provide a ‘ready-made solution’.Footnote ¹⁵⁴ Ambos agrees with this and expands upon it, saying that it is more sensible ‘to further develop in case law those elements in the existing international core crimes’ that touch on environmental destruction than to begin afresh.Footnote ¹⁵⁵ This approach involves taking the law as it currently stands and changing how it is interpreted through a process of judicial activism in the international courts, such as the ICC or the International Court of Justice (ICJ). This should be possible, as an essential feature of international courts, according to Cesare Romano and co-authors, is that they are ‘composed of independent adjudicators’ who are not beholden to any state or executive.Footnote ¹⁵⁶ Moreover, courts in general have a significant degree of authority – formal, normative, sociological and performative – as discussed above.Footnote ¹⁵⁷ In essence, then, this option involves casting the international courts in the role of the architects of ecocide. Note that domestic courts do not offer an attractive alternative for the creation of ecocide because, for example, many countries do not even give courts the capacity for common law-style judicial activism. Most obviously, civil law system states such as France – where codified statutes predominate and stare decisis is inoperable – confine judges to a more limited role.Footnote ¹⁵⁸

The notion of an organic, judge-led, emergence of ecocide has its merits. Certainly, it would remove the need for state sponsorship – upon which the Panel's proposal for Article 8 ter is utterly reliant. Free from the need to seduce states, judicial framing of ecocide could lead to something bolder and more ecocentric than the Panel's proposal. Court decisions also carry a good deal of authority and – while they typically lack the capacity to create formally binding precedent, as can be seen from the Statute of the International Court of Justice (ICJ Statute)Footnote ¹⁵⁹ – they nonetheless carry significant norm-creating influence. This influence can allow court decisions effectively to crystallise into customary international law and thus bind even the most recalcitrant states. Jurisprudence also evolves over time, meaning that a judge-led approach to ecocide has the potential to become more robust in the future. Consider, for example, the attitude of the ICJ towards the use of force, which has become increasingly restrictive through cases such as Corfu Channel, Nicaragua, Oil Platforms and Armed Activities in the Congo.Footnote ¹⁶⁰

However, there are significant downsides to relying on judicial activism. Firstly, it is probably not feasible to suggest that the anthropocentric limitations in each of the existing crimes (the environmental war crime, crimes against humanity and genocide) could merely be interpreted away. Indeed, any court that attempted to do so would be likely to find itself in violation of the principle of nulla poena sine lege.Footnote ¹⁶¹ Secondly, while the law on the use of force has arrived at a suitably robust standard (compare with the recent Russian invasion of Ukraine, which highlights the weaknesses in enforcement), it has taken the ICJ more than half a century to hone the law on the use of force into its current form (Corfu Channel was decided in 1949 and Armed Activities in the Congo was decided in 2005). It is undesirable to wait 50 years for ecocide to be hewn into solid form – especially as so many of the problems facing the environment are said to be urgent (the phrase ‘one minute to midnight’ is often heard).Footnote ¹⁶² Thirdly, judicial activism is not uncontroversial and can be corrosive for courts as institutions. An example of this is the Appellate Body of the World Trade Organization. The Appellate Body has recently been perceived by the United States to be taking an unacceptably activist position on key issues. The effect of this is that the US has suspended nominations to the Appellate Body, which is now without a quorum and unable to do its work. James Bacchus describes its potential loss as ‘devastating’.Footnote ¹⁶³

8.3. Drafters of legal manuals

Another potential group of architects that could be tasked with creating a new international crime of ecocide are the drafters of legal manuals. Manuals are non-binding documents drafted by experts drawn from the likes of academia and the military, and are often well regarded and frequently quoted. According to Heather Harrison Dinniss, manuals ‘play a valuable role in the development of international law, by influencing the decisions of policy makers, treaty negotiators and others, particularly in an era characterised by lack of formal law making processes in emerging domains’.Footnote ¹⁶⁴ Manuals have indeed been used to good effect in international law – often in a humanitarian law context – and a number of examples have been produced; these include the Oxford Manual (on warfare on land),Footnote ¹⁶⁵ the San Remo Manual (on warfare at sea),Footnote ¹⁶⁶ the Harvard Manual (on air and missile warfare),Footnote ¹⁶⁷ the (original) Tallinn Manual (on cyber attacks),Footnote ¹⁶⁸ Tallinn Manual 2.0 (on cyber operations)Footnote ¹⁶⁹ and the currently developing Woomera Manual (on international space law).Footnote ¹⁷⁰ Manuals have a role to play in the formation and interpretation of international law given that, according to the ICJ Statute, ‘the teachings of the most highly qualified publicists … [are a] subsidiary means for the determination of rules of law’.Footnote ¹⁷¹ However, manuals also come with a major drawback.

The trouble with taking the manual route in the context of ecocide is that, historically, the remit of manuals has been merely to capture and consolidate the current lex lata rather than to comment on and develop lex ferenda. For example, the preface to the Oxford Manual states thatFootnote ¹⁷²

Similarly, in the context of the Tallinn Manuals, the role of the experts was to ‘examine how extant legal norms apply to this new form of [cyber] warfare’.Footnote ¹⁷³ In Tallinn 2.0 the experts were clear that ‘it is not a “best practices” guide, does not represent “progressive development of the law” and is policy and politics-neutral’.Footnote ¹⁷⁴ Michael Schmitt, who directed the drafting processes, observed that the experts ‘assiduously avoided including statements reflecting lex ferenda’.Footnote ¹⁷⁵ Most recently, in the context of the emerging Woomera Manual, the drafters state that they will ‘maintain a strict focus on the law as it is (lex lata), not on the law as we might wish it to be (lex ferenda)’.Footnote ¹⁷⁶ The reason for the focus of manuals on the lex lata is captured well by Lianne Boer, who notes:Footnote ¹⁷⁷

Moreover, according to the drafters of the Woomera Manual, the lex ferenda ‘is more appropriate for official, intergovernmental efforts in drafting new international instruments’.Footnote ¹⁷⁸ In short, then, the drafters of legal manuals do not set out explicitly to change the law, merely to capture it.

It should be noted, however, that many doubt that the drafters of manuals truly confine themselves to capturing the lex lata and suspect that manuals are an attempt to quietly evolve the law. As Boer put it, ‘the experts are “really” doing lex ferenda rather than lex lata’.Footnote ¹⁷⁹ Dan Efrony and Yuval Shany make a similar point and agree that manuals have a more significant role that the drafters usually wish to admit, saying that, for example, the Tallinn Rules ‘attempted to flesh out an existing regulatory framework’.Footnote ¹⁸⁰ Likewise, they note that drafters are ‘opting to extend the law by way of interpretation and analogy’.Footnote ¹⁸¹ Thus, they conclude that there is a ‘gap … between [s]tate practice and some key Tallinn Rules’.Footnote ¹⁸² Statements by senior officials in countries such as the US and UK ‘also cast doubt on whether [the Tallinn Rules always] coincide with contemporary opinio juris’.Footnote ¹⁸³ This provides a further indication that the experts have done more than merely restate the law as it stands. Even if this is the case, though, and if the experts who draft manuals are really attempting to develop the law rather than collate it, it seems unlikely that they would make the best choice of architects for a new international crime of ecocide. This is because experts are usually only able to play on the grey areas at the edges of the law in order to encourage its evolution. For example, as Efrony and Shany put it, even the Tallinn Manuals ‘shape only to a limited degree state practice’.Footnote ¹⁸⁴ On the other hand, the creation of an entirely new international crime would amount to a revolution in the law. A change as significant as that cannot simply be slipped in through the back door in the hope that no one notices.

In short, a manual would not work in the context of ecocide because, officially at least, manuals are intended simply to restate the lex lata and, of course, there is currently no crime of ecocide to restate. Furthermore, even if most manuals unofficially attempt to influence the lex ferenda, it seems that the creation of an entirely new crime is simply too significant a change to effect in this way.

8.4. The International Law Commission

A final actor that might be considered as the architect of a new international crime of ecocide is the ILC. Under the ILC Statute,Footnote ¹⁸⁵ the Commission may consider proposals for the ‘progressive development’ of international law, by which is meant ‘the preparation of draft conventions on subjects which have not yet been regulated by international law or in regard to which the law has not yet been sufficiently developed in the practice of states’.Footnote ¹⁸⁶ Such proposals may be referred to the ILC by either (i) the United Nations General Assembly (UNGA)Footnote ¹⁸⁷ or (ii) other entities, such as states, other principal organs of the UN, specialised agencies or official bodies established by intergovernmental agreement.Footnote ¹⁸⁸ If such a referral were to be made, it has the potential to lead to a new treaty, or indeed the amendment of an existing treaty, to cover ecocide. As Robinson notes (writing in the context of a multilateral, rather than ILC-led effort), a treaty that is accepted by a critical mass of interested states could ‘start a “snowball effect” [and could ultimately] call upon [the] vastly greater enforcement machinery and expertise’ of states as opposed to international courts.Footnote ¹⁸⁹

Asking the ILC to design a new crime of ecocide is not a new idea: it was attempted between 1984 and 1996. In particular, the ILC considered including ecocide in the Draft Code of Crimes Against the Peace and Security of Mankind (the Draft Code),Footnote ¹⁹⁰ which eventually became the Rome Statute. Indeed, in 1984 the ILC set out to discuss the criminalisation of acts causing serious damage to the environment and, at one point, Article 26 was planned to criminalise, in war or in peacetime, ‘an individual who wilfully causes or orders the causing of widespread, long-term and severe damage to the natural environment’.Footnote ¹⁹¹ However, the word ‘ecocide’ was not used explicitly and Article 26 was gradually diluted during deliberations until it crystallised into the environmental war crime (discussed above). Ultimately, efforts to include a free-standing crime of ecocide were abandoned altogether by the ILC, and the Rome Statute – with which we are familiar – was produced.

The Human Rights Consortium (HRC), an alliance of NGOs, produced a detailed report in 2013, which attempted to explain the cause of the failure of the ILC attempt to criminalise ecocide.Footnote ¹⁹² The report notes that disagreements emerged in 1986 over whether environmental damage should be a crime of intent and, indeed, even over whether the term ‘ecocide’ was appropriate.Footnote ¹⁹³ In 1995, a working group was established to consider the issue and it later recommended pursuing one of three alternatives: (i) retaining environmental crimes as a distinct and separate provision, (ii) including environmental crimes as an act of crimes against humanity, or (iii) including environmental crimes as a war crime.Footnote ¹⁹⁴ However, according to the HRC, ‘despite this document, none of [the working group's] recommendations were followed up. Worse still, in 1996, at a meeting of the ILC, the then Chairman, Ahmed Mahiou, ‘unilaterally decided to remove the crime of ecocide completely as a separate provision … [w]ithout putting it to a vote’.Footnote ¹⁹⁵ While one other member of the working group, Szekely, is known to have objected, the deletion of ecocide (or indeed any purely environmental crime that occurred outside the context of war) was carried.Footnote ¹⁹⁶ The HRC report goes on to say that the precise reason for the ILC's decision to drop ecocide is not ‘well-recorded’ but that ‘one comment by the Special Rapporteur of the Code, Mr Thiam of Senegal, [was] that the removal was due to comments of a few governments [that were] largely opposed to any form of inclusion of Article 26’.Footnote ¹⁹⁷ Ultimately, though, it concludes that ecocide was removed ‘somewhat mysteriously’ from the Draft CodeFootnote ¹⁹⁸ and that it remains the ‘missing … 5th international crime against peace’.Footnote ¹⁹⁹

Despite this failure, there is no reason in principle why the ILC could not resurrect its discussions of ecocide. Clearly, the ILC path is not a straightforward one. Not only does it come with the usual level of bureaucracy that is associated with the UN but, given that the ultimate decision on whether to adopt any draft is decided by the UNGA, it can become politicised too. Consider, for example, the Draft Articles on State Responsibility.Footnote ²⁰⁰ It took from 1953 until 2001 for the ILC to produce its final draft and yet, even today, the UNGA has not formally adopted the document and merely ‘commends [the draft] to the attention of Governments without prejudice to the question of their future adoption’.Footnote ²⁰¹ Moreover, the geopolitical climate seems less conducive to major changes today than it was in 2001. Indeed, the drafters of the Woomera Manual say that ‘it is unlikely that the diverse interests of states in the current geo-political environment will coalesce around any new international instruments on space security’.Footnote ²⁰²

However, recent events in Ukraine notwithstanding, there is more cause for hope when it comes to the creation of a new international crime of ecocide. There is a great deal of momentum at the UN today for environmental protection. For example, the 2015 UN Sustainable Development Goals include Goal 15 by which states must ‘protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss’.Footnote ²⁰³ The UN Secretary General, Antonio Guterres, stated in 2020 that ‘humanity is waging war on nature … biodiversity is collapsing … one million species are at risk of extinction … ecosystems are disappearing before our eyes … [however] … human action can help to solve it’.Footnote ²⁰⁴ With luck, it might be possible to leverage that sort of momentum if the issue of ecocide were to be put before the ILC. Moreover, the ILC has much higher standing than the Panel, and therefore has greater clout, which it can use to produce a bolder vision of ecocide. This clout may help to place further political pressure on states – in the form of the UNGA – to take action. Also, unlike in the 1980s and 1990s, the ILC no longer needs to worry about balancing ecocide against the other crimes as they are already safely listed in the Rome Statute.Footnote ²⁰⁵ Further, in purely practical terms, ‘much of the background work required for developing [ecocide] has already been done [as previous] discussions within the UN lasted over a decade’.Footnote ²⁰⁶

Of course, if there is such momentum for environmental protection, one may legitimately ask why this should not simply manifest as state support for the Panel's proposal (rather than being directed towards an ILC-led effort). However, the deficiencies outlined above (such as persistent anthropocentrism and confusion over the mens rea) rule out the Panel's proposal as a viable offering in terms of its content. Further, the Panel's lack of heft on the international stage (discussed above in terms of its lack of formal, normative, sociological or performative authorityFootnote ²⁰⁷) undermine the Panel's offering in terms of its form. A proposal from the ILC would be likely to receive a warmer welcome given that states nominate individuals to membership of that body; they had no say in the formation of the Panel.Footnote ²⁰⁸

In short, given the greater political awareness of environmental issues, the greater standing of the ILC than the Panel, and the fact that much of the work has already been done, perhaps it is time to give the ILC a second attempt at designing the blueprint for ecocide. As Robinson observed, the current proposals put forward by the Panel and others are all ‘imperfect’ and there may be some undiscovered ‘missing link’ out there somewhere.Footnote ²⁰⁹ Given its expertise, the ILC is arguably best placed to find it.

8.5. Summary

Alternative architects are available besides the Panel for creating an international crime of ecocide. However, in terms of the international courts, there is a limit to what judges can realistically achieve without losing the confidence of states through claims of ‘judicial activism’ or, more importantly, without violating the principle of nulla poena sine lege. Similarly, it seems unlikely that the drafters of an international law manual on ecocide would be able to bring about the necessary change in the law given that manuals have historically been wedded, at least formally, to collating and clarifying the lex lata. The remaining option, the ILC, is daunting both in terms of process and politics. Nonetheless, if the UN machine is prepared to throw its support behind the effort, the ILC has the potential to be a good architect for change, given its standing and prominence. Moreover, consideration by the ILC could give ecocide a fuller airing and is less likely to result in the sort of pre-emptively defanged proposal that the Panel managed to achieve when it anthropocentrised its own proposal in the hopes of making it palatable to the parties to the Rome Statute.