Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-14T09:56:33.602Z Has data issue: false hasContentIssue false
  • English
  • Français

The Obligation to Undertake an Environmental Assessment in the Jurisprudence of the ICJ: A Principle in Search of Autonomy

Published online by Cambridge University Press:  26 April 2017

Elisa RUOZZI
Rights & Permissions [Opens in a new window]

Abstract

The practice of carrying out an environmental impact assessment (EIA) has gained strength in international law and jurisprudence, finding application in the case law of the International Court of Justice (ICJ). If, on the one hand, the ICJ has recognized the customary nature of this principle, on the other its application poses a set of challenges, mainly linked to the autonomy of this obligation from other international environmental law norms. More precisely, the obligation at issue has been applied in connection with the due diligence and notification principles, creating uncertainty about its scope, as well as about its substantive or procedural nature. Likewise, the autonomy of the obligation to perform an EIA has been challenged in relation to the definition of the content and scope of the obligation itself, which in turn is linked to the existence of applicable treaty provisions or of soft law. This article discusses the impact of these elements on the reasoning of the Court in the cases at issue, in order to demonstrate how such lack of autonomy can undermine the coherence of the reasoning itself and, therefore, an effective application of the principle.

I. Introduction

Nowadays, the obligation to conduct an EIA in relation to any project capable of having a significant impact on the environment has been introduced in most national systems, as well as at the international level. Such a normative aspect is rooted in the effectiveness which is almost unanimously associated with EIA: even if it is extremely difficult to quantify the benefits determined by the performance of an EIA in relation to a given project, the doctrine generally recognizes that such a practice tends to improve the quality of decision making and of the design of the plan, mainly in terms of mitigation of negative environmental impacts and of choice of “greener” options.Footnote 1 Further benefits have been identified in enhanced transparency towards civil society and better cooperation between the proponent and public authorities which, in turn, can contribute to a higher acceptance of the project and to reduced litigation costs.Footnote 2 On the other hand, the costs for the proponent appear to be reasonable if not limited: the performance of an EIA has been broadly estimated to be below 1% of the total investment cost;Footnote 3 costs for public administration are more difficult to determine, as EIA often takes place in parallel with other procedures for the same project.Footnote 4 In a similar way, estimates regarding time costs caused by the performance of an EIA differ widely, even though there is no widespread evidence that such a procedure regularly causes significant delays in the completion of the project.Footnote 5

The role played by EIA in the protection of the environment was recognized at the international level as early as 1992, when the Rio DeclarationFootnote 6 established the obligation of States to undertake an EIA when proposing activities that could have an impact on the environment. Article 17 of the Declaration states that: “Environmental impact assessment, as a national instrument, shall be undertaken for proposed activities that are likely to have a significant adverse impact on the environment and are subject to a decision of a competent national authority.”

Notwithstanding the non-binding force of the Declaration, with the insertion of Article 17 its drafters gave a strong signal as to the relevance the principleFootnote 7 was acquiring in international environmental law. Such relevance had already been underlined, though a soft law instrument, by the United Nations Environmental Programme (UNEP), which, in 1987, published the Goals and Principles of Environmental Impact Assessment,Footnote 8 a set of guidelines aiming to set out in detail the modalities through which EIA had to be conducted by States. An even stronger endorsement of the principle came with the adoption, in 1991, of the Convention on Environmental Impact Assessment in a Transboundary Context (Espoo Convention);Footnote 9 the Convention, drafted in the framework of the United Nations Economic Commission for Europe, entered into force in 1997 and was completed in 2003 by a Protocol concerning strategic environmental assessment.Footnote 10 Leaving aside specific instruments dealing exclusively with this procedure, the obligation to undertake an EIA is contained in several international conventions – including the Convention on Biological DiversityFootnote 11 and the Protocol on Environmental Protection to the Antarctic TreatyFootnote 12 – as well as in the International Law Commission Draft Articles concerning Prevention of Transboundary Harm from Hazardous Activities.Footnote 13

In the light of these elements, the obligation to perform an EIA is deemed to have acquired customary nature; however, strong uncertainties persist with regard to its scope and content, as clearly emerges from the jurisprudence of the ICJ. Increasingly faced with cases concerning the protection of the environment and involving the obligation for States to undertake an EIA, the Court has recognized its customary nature, though appearing uncomfortable with some aspects of its application. On the basis of an analysis of the relevant jurisprudence, it is submitted, in particular, that the principle obliging States to undertake an EIA suffers from a lack of autonomy from other international environmental law principles. This, in turn, implies a set of consequences on its application, including a problematic cross-contamination between procedural and substantial aspects, which are capable of undermining the effectiveness of EIA itself.

These aspects have not yet been specifically addressed by existing literature which has focused, on one side, on the definition and application of the principle in the context of specific geographical areas,Footnote 14 national systems,Footnote 15 actors,Footnote 16 international organizationsFootnote 17 or treatiesFootnote 18 and, on the other, on the recognition of its customary nature.Footnote 19 This article therefore aims to fill this lacuna through the analysis of the jurisprudence of the ICJ, whose case law is of crucial relevance for the assessment of customary international law. A proper understanding of the content of the obligation to carry out an EIA can be of interest for academics as well as for practitioners, in the first place those involved in the drafting of national legislation, given that, as will be seen, the content of an EIA is still mainly based on national law. In addition to this, the topic can be of direct relevance for those who, at the national and international level, are responsible for the authorisation or funding of projects capable of having a transboundary impact on the environment, as failure to carry out an adequate EIA could be invoked in front of a tribunal as a ground preventing the completion of the project.

The Report is structured as follows: after a short overview of the emergence and consolidation, in the jurisprudence of the Court, of the principle imposing an EIA (section II), the issue of the lack of autonomy will be addressed (section III), and its consequences on the jurisprudence of the Court assessed (section IV). Finally, some suggestions and final remarks will follow (section V).

II. The recognition of the principle by the ICJ: from Nuclear Test II to Certain activities in the border area

It was not before 2010 – in the Pulp Mills case – that the principle imposed on States to carry out an EIA found explicit application by the Court. However, its relevance came into play far earlier, starting from the Nuclear Test II case, introduced by New Zealand in 1995 and concerning nuclear experiments carried out by France in the Pacific Ocean.Footnote 20 In its oral statements, New Zealand affirmed the obligation on France to carry out an EIA before any major project that might affect the marine environment, based not only on the Noumea Convention binding the two States,Footnote 21 but also on customary law.Footnote 22

As is known, the Court did not enter into the merits of the subject matter, arguing that the possibility of revising the first judgment issued in 1974 was restricted to tests carried out in the atmosphere, whereas the tests that formed the object of the new claim would take place underground.Footnote 23 Notwithstanding this, in his dissenting opinion, Judge Weeramantry took into account the principle imposing an EIA, which, in his view, was gathering strength and international acceptance, and had reached the level of “general recognition” of which the Court should take notice.Footnote 24 After having recalled the UNEP Principles mentioned above, the Judge affirmed that the magnitude of the problems alleged by New Zealand made the principle applicable in the case at stake.Footnote 25 Moreover, he argued that, based on the “position of special trust and responsibility in relation to the principles of environmental law” occupied by the Court, the Court was entitled to take into account the EIA principle when a prima facie case was made of the possibility of environmental damage.Footnote 26 Finally, the Judge pointed out that a potential finding of the absence of any risk, though plausible, would be reached after the EIA, and not before.Footnote 27

A few years later, the same judge devoted a part of his separate opinion in the case of Gabčíkovo-Nagymaros Footnote 28 to the principle of “continuing environmental impact assessment”. In particular – and despite the fact that the principle imposing an EIA was not invoked by the parties to the dispute or applied by the Court – the Judge stated that the principle played an important role in the case at issue, and underscored its nature as a continuing process that should take place as long as the project is in operation. Such a statement was justified not only by EIA being a “dynamic principle”, but also by considerations of prudence, as any project can have unexpected consequences, especially if great in size and scope.Footnote 29 According to the Judge, the customary nature of continuous monitoring as part of the obligation to carry out an EIA found support in international and domestic practice and, from a theoretical point of view, on an EIA being a specific application of the larger general principle of caution.Footnote 30

In the case at stake, the applicability of the principle was not raised as an issue, as monitoring obligations were contained in the treaty that formed the basis of the jurisdiction of the Court; however, the Judge submitted that the obligation to carry out an EIA should be considered as “in-built” in any treaty that may reasonably be considered to have a significant impact upon the environment, on the basis of environmental law in its current state of development.Footnote 31

As mentioned above, the Pulp Mills caseFootnote 32 was the first time that the principle was explicitly dealt with by the Court. In this case, concerning the construction of two pulp mills on the banks of the Uruguay River, the obligation to perform an EIA was invoked by Argentina, according to which Uruguayan authorities released the authorization to construct the plants without transmitting an adequate EIA.Footnote 33

In its judgment, the Court fully recognized the customary nature of the obligation to perform an EIA by describing it as “a practice, which in recent years has gained so much acceptance among States that it may now be considered a requirement under general international law”.Footnote 34 In the words of the Court, “where there is a risk that the proposed industrial activity may have a significant adverse impact in a transboundary context, in particular, on a shared resource”, an EIA has to be performed.Footnote 35

The position is confirmed in recent casesFootnote 36 concerning the occupation of Costa Rican territory by Uruguay, including dredging and canalization activities on the San Juan RiverFootnote 37 and the violation of Nicaragua’s territorial integrity by the construction of a road in the border area.Footnote 38 The obligation to perform an EIA was invoked by both parties: according to Costa Rica, Nicaragua should have conducted such an assessment before carrying out dredging works,Footnote 39 but according to the counterparty, Costa Rica breached the same obligation in relation to the construction of the road.Footnote 40

Recalling its statements in the Pulp Mills case, the Court confirmed the customary nature of the principle obliging States to undertake an EIA, pointing out that, even if the Pulp Mills ruling referred to industrial activities, the principle applies generally to proposed activities that may have a significant adverse impact in a transboundary context.Footnote 41

On the basis of this jurisprudence, it can, therefore, be concluded that the principle imposed on States to perform an EIA before commencing activities that may cause damage to the environment has been definitely recognized by the Court as part of customary international law.Footnote 42

This said, the recognition of this obligation as a customary one merely forms the premise of the real object of this analysis, i.e. the scope of the obligation itself and its status as an autonomous legal principle of international environmental law.

III. The duty to undertake an EIA: an autonomous principle or an ancillary obligation?

The autonomous or ancillary nature of the principle obliging States to perform an EIA can be analysed, first of all, in its relation to other procedural and substantive environmental obligations.Footnote 43 This issue is clearly exemplified by the Pulp Mills case, where the principle is taken into account in the paragraph concerning the breach of procedural obligations and, more precisely, in the subparagraph relating to the duty to notify the plans concerning a new activity to the other party. Such a duty is established by Article 7 of the 1975 Statute linking the two parties: according to this provision the notification shall describe the main aspects of the work and any other technical data that will enable assessment of the impact of the activity on the régime of the river.Footnote 44 In this context, the EIA is, therefore, seen not as an autonomous obligation, but as part and parcel of the obligation to notify, as confirmed by the statement of the Court according to which the EIA mandated by Article 7 is intended to enable the notified party to participate in the process of ensuring that the assessment is complete.Footnote 45 Argentina was found to be in breach of the duty to notify not by reason of the content of the EIA, but because the EIA was notified to Argentina after (and not before) the activity was undertaken.Footnote 46

Similarly, in the same case, the duty to perform an EIA is linked to substantive environmental obligations: in the words of the Court, the subparagraph specifically devoted to EIA deals with “the relationship Footnote 47 between the need for an environmental impact assessment … and the obligations of the Parties under Article 41”, i.e. the obligation to adopt adequate domestic measures that aim to protect the aquatic environment.Footnote 48 Such a relationship is described in two ways: on one side, the obligation to protect and preserve the environment established by Article 41 of the 1975 Statute has to be interpreted in accordance with the obligation to undertake an EIA.Footnote 49 On the other, “due diligence, and the duty of vigilance and prevention which it implies, would not be considered to have been exercised” if an EIA is not duly performed.Footnote 50

Once again, therefore, the obligation to conduct an EIA is not recognized as having full autonomy, but is analysed in relation to another obligation – in this case, the duty to prevent environmental harm, which, in turn, is contained in both treaty (the 1975 Statute) and customary obligations. In this context, the principle works both as an interpretative tool with respect to the duties established by Article 41 and as a conditio sine qua non with respect to the principles of due diligence, vigilance and prevention. It is, therefore, plausible to conclude, using the words of the Court, that the analysis is focused not on the principle in itself, but on the “role of the environmental impact assessment in the context of procedural obligations” and “in the fulfilment of the substantive obligations” of the parties.Footnote 51

It is true that the obligation to perform an EIA was not, as such, contained in the 1975 Statute, which could explain the choice to deal with it in relation to other obligations more as a necessity linked to the applicable law than as a deliberate approach. However, at least two elements suggest this second hypothesis. First of all, the relationship, established by the Court, between an EIA and customary international norms such as due diligence, vigilance and prevention expands its analysis beyond treaty law. Second, the fact that the recognition by the Court of the obligation to perform an EIA as a customary norm would have enabled it, regardless of the treaty provisions, to consider it independently from other obligations.

The issue of the autonomous or ancillary nature of the principle arises for a second time in the joint cases involving Costa Rica and Uruguay where, as mentioned above, the customary nature of the principle is recognized and where the ICJ states that: “to fulfill its obligation to exercise due diligence in preventing significant transboundary environmental harm, a State must, before embarking on an activity having the potential adversely to affect the environment of another State, ascertain if there is a risk of significant transboundary harm, which would trigger the requirement to carry out an environmental impact assessment.”Footnote 52

That statement implies that the obligation to perform an EIA, far from being considered as an autonomous one, is dependent upon two different factors: due diligence and prevention obligations and the ascertainment of a risk of significant harm to the environment. In relation to the first factor, similarly to what was observed in the Pulp Mills case, the obligation to undertake an EIA is framed in terms of a means-end relationship with respect to due diligence and prevention obligations. Even if it is not possible, in this context, to enter into details of the complex issue of the definition of due diligence,Footnote 53 suffice it to say that such a definition in turn affects the relationship between due diligence and the duty to perform an EIA, especially in relation to the qualification of due diligence as an obligation of conduct or as an obligation of result.Footnote 54

Finally, a further aspect of the autonomy of the obligation to conduct an EIA concerns the definition of its content, which in turn is linked to the existence of an international law source that can work as applicable law in the assessment of parties’ conduct. In the Pulp Mills case, the Court ruled out the application of the Espoo Convention on the basis of the fact that neither State was party to it, but observed that the UNEP Guidelines, though not binding, had to be taken into account “as guidelines issued by an international technical body”.Footnote 55 Based on that, the Court established that: “it is for each State to determine in its domestic legislation or in the authorization process for the project, the specific content of the environmental impact assessment required in each case, having regard to the nature and magnitude of the proposed development and its likely adverse impact on the environment as well as to the need to exercise due diligence in conducting such an assessment.”Footnote 56 Such an approach is confirmed in the Certain activities and Construction of a Road cases, whereas no mention is made of international instruments.Footnote 57

If, therefore, this jurisprudence seems to confirm the non-autonomous character of the principle, the analysis of the impact of such an approach on the reasoning of the Court constitutes a further, interesting step.

IV. The lack of autonomy of the principle and its impact on the reasoning of the court

If, as illustrated above, the performance of an EIA is generally deemed to bring about environmental benefits, an autonomous application of the corresponding normative principle – the obligation to perform an EIA – is of the essence. What it is argued, therefore, is that the lack of autonomy, far from being a mere formal aspect of the legal reasoning, can impact on the effectiveness of the principle and jeopardise its benefits.

A first, relevant consequence of the lack of autonomy of the obligation to undertake an EIA is related to the distinction between procedural and substantive obligations. As has been shown, the obligation to perform an EIA, being an instrumental one, does not qualify as procedural or substantive per se. This determines a certain contamination, in the reasoning of the Court, between these two dimensions and, more precisely, the fact that compliance with what is ultimately a procedure might become dependent on factual aspects. It is submitted that this could undermine the very rationale of EIA, i.e. the identification of future environmental harm, to the extent that the presence of harm, instead of being the outcome of the process, becomes one of its constitutive elements. This, in turn, could undermine some of the main advantages associated with EIA, such as the improvement of the design of the project and the choice of more environmentally-sound options.

The problems stemming from establishing a link between the performance of an EIA and substantial aspects have been underlined by Judge Dugard, who argued that the “danger” of viewing the due diligence obligation as the source of the obligation to perform an EIA is that it allows a State to argue, retrospectively, that because no harm has been proved at the time of the legal proceedings, no duty of due diligence arose.Footnote 58 According to the Judge, such a “backward looking approach” was applied in the Certain activities case, where the Court confirmed the finding – already contained in the study carried out by Nicaragua in 2006 – that the dredging programme would have no impact on the river environment, which, in turn, allowed it to state that Nicaragua was not required to carry out an EIA.Footnote 59

The mixing of procedural and substantive issues can be seen, first of all, in the Pulp Mills case. Here the analysis of the compliance – in the broader framework of the violation of substantive obligations – with the duty to carry out an EIA in relation to the choice of the site of the mill is split in two further questions: whether Uruguay failed to act with due diligence in conducting the EIA – particularly with respect to the choice of the location – and whether the location chosen was actually unsuitable or could damage the environment of the riverFootnote .60 Whereas the first point is assessed by taking into account the methodology followed (in particular, the consideration of alternative sites),Footnote 61 as regards the second issue the Court focuses its analysis on substantial aspects, i.e. the capacity of the waters of the river to receive, dilute and disperse discharges of effluent from the plant.Footnote 62 The reasoning of the Court is therefore logically flawed to the extent that it is based on the correct performance of the procedure and, at the same time, on the outcome that such a procedure should have brought about.

The same applies to the Road in Costa Rica and Certain activities cases because of the fact that, as established by the Court, the obligation to undertake an EIA only arises once a preliminary assessmentFootnote 63 of the existence of a risk has been carried out. Once again, the performance of an EIA is made contingent upon an element – the existence of risk – which should be the result of EIA itself. In addition to this, the application of a probabilistic logic inherent in any risk assessment clashes with the very nature of EIA, which tends to be conceived as a quasi-automatic procedure. In any case, it would be hard to establish which threshold of risk is the one applied in the context of the preliminary assessment and which one in the context of the EIA.Footnote 64 These thresholds, in turn, do not correspond to the one applied to ascertain risk after the activity has taken place.Footnote 65

A further element that can undermine the effectiveness of the obligation to carry out an EIA lies in its relationship with the obligation to notify and to consult the other party. Whereas in the Pulp Mills case the EIA is ancillary to the obligation to notify, in the other cases the relationship is inverted as, according to the Court, such an obligation arises once the EIA has been carried out.Footnote 66 Such an approach raises some perplexities; first of all because it is not consistent with the most frequent formulations of the principle of consultation and notification, which consider it as a self-standing obligation.Footnote 67 As observed by Judge Donoghue, States’ practice does not justify such a formulation, as due diligence may call for notification at a different stage in the process, e.g. before the EIA has taken place.Footnote 68 Moreover, from a practical point of view, the benefits in terms of transparency and public participation that generally stem from EIA would be reduced if the obligation to perform the EIA was only triggered when an obligation to notify exists, as this obligation does not always apply. On the other side, the opposite relationship would give rise to a paradoxical effect to the extent that failure to comply with the obligation to perform an EIA would exempt the defendant from the obligation to notify.Footnote 69

Finally, the statements by the Court with respect to the scope and content of the obligation to carry out an EIA raise a set of uncertainties. First of all, it is not clear whether with its pronouncement the Court meant to make a full-fledged renvoi to national legislation.Footnote 70 Even if this was the case, it remains to be seen how national provisions (of which of the parties in dispute?) could be reconciled with international instruments that, as in the case of UNEP Guidelines, just have to be “taken into account”. At the same time, the Court did set some criteria that could be interpreted as a sort of benchmark in case of judicial review and, therefore, point to the formation of a customary norm. The lack of a clear position in relation to this issue not only prevents a definition of the content of the principle, but also blurs the assessment of States’ conduct to the extent that the question as to what an EIA should in theory include merges with that as to what the EIA actually included. Such a problem is exemplified in the Pulp Mills ruling, where the Court first has to deal with the question of whether the assessment should have, “as a matter of method”,Footnote 71 necessarily considered alternative sites. However, its reasoning takes into account both whether the examination of alternative sites was mandated by UNEP Principles and whether Uruguay actually included this kind of information in its analysis.Footnote 72 This démarche is even more evident in relation to the point concerning the consultation with affected populations, with respect to which the Court affirms to deal (and actually deals) with the point whether these populations should have, “or Footnote 73 have in fact, been consulted”.Footnote 74 The uncertainty in the identification of the precise content of the obligation to carry out an EIA is likely to give birth to a vicious circle, to the extent that the lack of legal certainty might prompt judicial organs to rely, in the last resort, on national law and that no incentive will arise for States where legislation regarding EIA is still weak to adapt it to an international standard.

V. Conclusions

The jurisprudence analysed above enables us to make some remarks about the way the obligation to perform an EIA has been applied so far by the ICJ and, more precisely, about its independent status as a principle of international environmental law.

As was underlined at the outset, the jurisprudence of the Court leaves no doubt as to the customary status of the principle. On the other hand, the obligation to perform an EIA struggles to find a definite shape, both in relation to other customary principles of public international law and in relation to existing treaty law.

With respect to the first aspect, it has been remarked how the Court tends to consider the obligation to perform an EIA in conjunction with other environmental obligations. It is here submitted that, to achieve greater effectiveness in the ultimate aim of these principles (the protection of the environment), a decoupling would be recommendable. In the first place, the obligation to perform an EIA should be applied independently from due diligence and prevention obligations, not only as the definitions of these obligations are (especially the first one) not completely clear, but also to avoid a “backward looking approach”. As illustrated above, such an approach allows a determination that, where it is demonstrated that the activity does not pose any risk, the EIA is deemed to be not necessary. Likewise, the obligation to perform an EIA should also be assessed independently from the notification obligation; even if, in practical terms, the EIA often forms part of the documents that have to be notified to the other party, a separate application would allow the judge to hold the State responsible for two distinct breaches. This could be particularly useful in those circumstances (as the Pulp Mills case) where applicable treaty provisions are specific on notification obligations but more vague in relation to the performance of an EIA.

This leads us to the last point, i.e. the content of the EIA and the existence of a customary norm that, regardless of applicable treaty provisions (e.g. the Espoo Convention), can identify a noyeau dur of information the EIA has to contain. In relation to this point, in his separate opinion, Judge Bhandari posited that, despite its regional nature, the Espoo Convention could be used as a “standard that nation States should strive towards”.Footnote 75 Based on this idea, he suggested what he deemed could constitute a “lowest common denominator”Footnote 76 while conducting an EIA, including e.g. a description of the activity and of potential alternatives.Footnote 77

Fascinating though such an idea may be, it is the opinion of the author that such a detailed approach is incompatible with the general nature of obligations that characterize customary norms and, in any case, with the current practice and opinio juris of States. It is, therefore, submitted that, at the current stage of development of international law, the content of an EIA cannot but be determined by the international treaties that States are party to, which, in turn, will impact on their domestic law. This is why the adoption of soft law instruments (such as the UNEP Principles), as well as the ratification of the Espoo Convention, will be of the essence for the advancement of the practice to undertake an EIA.

Footnotes

*

Assistant Professor in International Law, University of Turin; email: elisa.ruozzi@unito.it.

1 Institute for Environmental Studies, Cost and Benefits of the EIA Directive (2007) 12.

2 Ibid, 13–14.

3 Ibid, 8.

4 Ibid, 9.

5 Ibid, 10.

6 1992 Rio Declaration on Environment and Development, UN Doc. A/CONF.151/26 (vol. 1), 14 June 1992.

7 On the distinction between principles and rules in international environmental law see Daniel Bodansky, Jutta Brunnée and Ellen Hey, International Environmental Law (Oxford: Oxford University Press 2007) 429 et sqq. It is important to remark, however, that some scholars consider EIA as a legal requirement or an implementing measure (Alexandre Kiss and Dinah Shelton, Guide to International Environmental Law (Leiden: M. Nijhoff 2007) 113 et sqq).

8 Goals and principles of environmental impact assessment, UNEP Res. GC14/25, 14th Sess. (1987).

9 Convention on Environmental Impact Assessment in a Transboundary Context, Espoo, 25 February 1991, in force 10 September 1997.

10 UNECE Protocol on Strategic Environmental Assessment to the Convention on Environmental Impact Assessment in a Transboundary Context, Kyiv, 21 May 2003, in force 10 July 2010.

11 Convention on Biological Diversity, Nairobi, 22 May 1992, in force 29 December 1993, Art. 14.

12 The Protocol on Environmental Protection to the Antarctic Treaty, Madrid, 4 October 1991, in force 14 January 1988, Art. 8.

13 International Law Commission, Prevention of Transboundary Harm from Hazardous Activities, 2001, Art. 7.

14 Gunnar Sander, “International Legal Obligations for Environmental Impact Assessment and Strategic Environmental Assessment in the Arctic Ocean” (2016) 31 International Journal of Marine and Coastal Law 88 et sqq.

15 Mariachiara Alberton, Public Participation in Environmental Decision-making in the EU and in China: the Case of Environmental Impact Assessment (Baden-Baden: Nomos 2014).

16 Jean-Roger Mercier, “The World Bank and Environmental Impact Assessment” in Kees Bastmeijer and Timo Koivurova (eds), Theory and Practice of Transboundary Environmental Impact Assessment (Leiden: Nijhoff 2008) 291 et sqq.

17 Ludwig Krämer, “Impact Assessment and Environmental Costs in EU Legislation” (2014) 11 Journal of European Environmental & Planning Law 201 et sqq.

18 Kees Bastmeijer and Ricardo Roura, “Environmental Impact Assessment in Antarctica” in Bastmeijer and Koivurova, supra note 16, 175 et sqq; Timo Koivurova, “Could the Espoo Convention become a Global Regime for Environmental Impact Assessment and Strategic Environmental Assessment?” in Robin Warner and Simon Marsden (eds), Transboundary Environmental Governance: Inland, Coastal and Marine Perspectives (Oxford: Ashgate 2012) 323 et sqq; Laura Pineschi, “The Duty of Environmental Impact Assessment in the First ITLOS Chamber’s Advisory Opinion: towards the Supremacy of the General Rule to protect and preserve the Marine Environment as a Common Value?” in Nerina Boschiero and others (eds), International Courts and the Development of International Law: Essays in Honour of Tullio Treves (The Hague: Asser Press 2013) 425 et sqq.

19 MA Fitzmaurice, “International protection of the environment” (2001) 293 Recueil des Cours de l’Académie de Droit International 280 et sqq.

20 Request for an Examination of the Situation in Accordance with Paragraph 63 of the Court’s Judgment of 20 December 1974 in the Nuclear Tests (New Zealand v France) Case, 22 September 1995, Order (1995) ICJ Reports 288 et sqq. According to para. 63 of the judgment rendered in 1974, the same kind of conduct by France established the possibility for the applicant to request an examination of the situation if the basis of the judgment “were to be affected” (Nuclear Tests (New Zealand v France), Judgment, 20 December 1974, (1974) ICJ Reports 457 et sqq., at para. 63). This provision was relied upon by New Zealand following the resumption, on the part of France, of nuclear testing in the same region.

21 Convention for the Protection of the Natural Resources and Environment of the South Pacific Region, Noumea, 24 October 1986, in force 22 August 1990.

22 Request for an examination, supra note 20, at para. 35.

23 Ibid, at para. 63.

24 Ibid, Dissenting opinion of Judge Weeramantry, at 344.

25 Ibid, at 345.

26 Ibid.

27 Ibid.

28 Gabčikovo‐Nagymaros Project (Hungary/Slovakia), Judgment, 25 September 1997, (1997) ICJ Reports 1 et sqq, at 7. As is known, the case concerned the joint construction, by Hungary and the Czech Republic, of a system of locks. Hungary invoked the suspension of the Treaty binding the two States to carry out the activity on the basis of the environmental risks involved (at para. 37). See, among others, John Fitzmaurice, “The Ruling of the International Court of Justice in the Gabcikovo-Nagymaros Case: A Critical Analysis” (2000) 9 European Environmental Law Review 80 et sqq.

29 Gabčikovo‐Nagymaros Project, Separate opinion of Judge Weeramantry, at p. 111.

30 Ibid, at 113.

31 Ibid, at 112.

32 Pulp Mills on the River Uruguay (Argentina/Uruguay), Judgment, 20 April 2010, (2010) ICJ Reports 14 et sqq.

33 Ibid, at paras. 22, 118.

34 Ibid, at para. 83.

35 Ibid, at para. 104.

36 Certain Activities Carried out by Nicaragua in the Border Area (Costa Rica/Nicaragua), Judgment, 16 December 2015. The proceedings were joint with Certain Activities carried out by Nicaragua in the Border Area (Costa Rica v Nicaragua).

37 Ibid, at para. 1.

38 Ibid, at para. 9.

39 Ibid, at para. 100.

40 Ibid, at para. 146.

41 Ibid, at para. 104.

42 It is to be noted, however, that not all judges concur with this idea. It is the case, for example, that Judge Donoghue, in his separate opinion, states not to be confident that State practice and opinio juris support the existence of such a specific rule (ibid, Separate opinion of Judge Donoghue, at p. 4).

43 As regards the relationship between these two categories of obligations, in the Pulp Mills case, parties posited different conceptions of such a link. Whereas, according to Uruguay, procedural obligations form a means to achieve the respect of substantive ones (Pulp Mills, supra note 32, at para. 73), according to Argentina, a breach of procedural obligations automatically entails a violation of substantive ones (ibid, at para. 72). The Court took an intermediate position in this respect, by arguing that a functional link exists between the two categories of obligations, which can in any event also be seen separately (ibid, at para. 79).

44 Ibid, at para. 112.

45 Ibid, at para. 119.

46 Ibid, at para. 121.

47 Emphasis added.

48 Pulp Mills, supra note 32, at paras. 190, 203.

49 Ibid, at para. 204.

50 Pulp Mills, supra note 32, at para. 204.

51 Ibid, at para. 206.

52 Certain activities, supra note 36, at para. 104.

53 According to the definition given by the Seabed Disputes Chamber of the International Tribunal for the Law of the Sea, the due diligence obligation is an obligation of conduct that aims to ensure the prevention of environmental harm. Such an obligation consists not only of the adoption of legislative and administrative measures, but also of undertaking efforts to ensure that private operators comply with these norms (Responsibilities and Obligations of States Sponsoring Persons and Entities with Respect to Activities in the Area, 1 February 2011, at paras. 110–111).

54 The uncertainties relating to the definition of the obligation of due diligence clearly emerge in the separate opinions attached to the Certain Activities and Construction of a Road cases. According to Judge ad hoc Dugard, the duty of due diligence is a standard of conduct required to implement the principle of prevention, which, in turn, is implemented though a number of specific, independent obligations, which include the duty to carry out an EIA (Certain activities, supra, note 36, Separate opinion of Judge Dugard, at paras. 7–9). The autonomy of the obligation to undertake an EIA is instead opposed by Judge Owada, who states this obligation is one element in the “holistic process” that emanates from the obligation of States to use due diligence to avoid environmental harm, “rather than a separate and independent obligation standing on its own under general international law” (ibid, Separate opinion of Judge Owada, at para. 18). In the same sense, see also the opinion of Judge Donoghue, which defines due diligence as the “governing primary norm” (ibid, Separate opinion of Judge Donoghue, at para. 9). As has been underlined, Art. 12 of the Convention on the Law of the Non-navigational Uses of International Watercourses (New York, 21 May 1997, in force 17 August 2014) while imposing a notification duty on State parties, actually falls short of obliging them to conduct an EIA, which has to be included only if “available”. However, it has been suggested that the “due diligence” standard contained in the same Convention places States under an obligation to provide a certain level of information (Alistair Rieu-Clarke, “Notification and Consultation on Planned Measures Concerning International Watercourses: Learning Lessons from the Pulp Mills and Kishenganga Cases” in Ole Kristian Fauchald, David Hunter and Xi Wang (eds), Yearbook of International Environmental Law (Oxford: Oxford University Press) 102 et sqq., 109.

55 Pulp Mills, supra note 32, at para. 205.

56 Ibid, para. 205.

57 Certain activities, supra note 36, at para. 104.

58 Ibid, at para.10.

59 Ibid.

60 Pulp Mills, supra note 32, at para. 209.

61 Ibid, at para. 210.

62 Ibid, at para. 211.

63 Certain activities, supra note 36, at para. 154.

64 It is interesting to note how a similar problem arose in relation to the application of Art. 5.7 of the Agreement on Sanitary and Phytosanitary measures (SPS measures) (allowing Members to adopt a precautionary approach, and therefore to adopt trade-restrictive measures in the absence of scientific certainty) and the general obligation to ensure that SPS measures are based on a risk assessment contained in Art. 5.1 of the Agreement. In an attempt to clarify the relationship between the two provisions, the Appellate Body has affirmed that “Article 5.7 is concerned with situations where deficiencies in the body of scientific evidence do not allow a WTO Member to arrive at a sufficiently objective conclusion in relation to risk” and that the same provision, Art. 5.7 “contemplates situations where there is some evidentiary basis indicating the possible existence of a risk, but not enough to permit the performance of a risk assessment” (United States – Continued Suspension of Obligations in the EC – Hormones Dispute, Report of the Appellate Body, WT/DS320/AB/R, 16 October 2008, at paras. 677–678).

65 Certain activities, supra note 36, Separate opinion of Judge Dugard, at para. 19.

66 Ibid, at para. 104.

67 See, for example, Art. 19 of the Rio Declaration (supra note 6), according to which “States shall provide prior and timely notification and relevant information to potentially affected States on activities that may have a significant adverse transboundary environmental effect and shall consult with those States at an early stage and in good faith”. See also Art. 15.2(b) of the Protocol on Environmental Protection to the Antarctic Treaty (supra note 12) and Article 14.1(c) and (d) of the Convention on Biological Diversity (supra note 11).

68 Certain activities, supra note 36, Separate opinion of Judge Donoghue, at para. 21.

69 Certain activities, supra note 36, at para. 168 where the Court established that: “the duty to notify and consult does not call for examination by the Court in the present case, since the Court has established that Costa Rica has not complied with its obligation under general international law to perform an environmental impact assessment prior to the construction of the road.”

70 In this regard, see the separate opinion of Judge Donoghue, who excludes that the words of the Court can be read in terms of incorporating national legislation (ibid, Separate opinion of Judge Donoghue, at para. 15). Contra see the opinion of Judge Bhandari, according to whom this possibility is actually allowed, even if the scarcity of guidance from the Court and other sources of international law lead one to think that there are presently no minimum binding standards in this regard (ibid, Separate opinion of Judge Bhandari, at para. 29). An intermediate position is the one held by Judge Dugard, who, on the basis of the works carried out by the ILC, affirms that that same matters are inherent in the nature of an EIA and cannot, therefore, be left to national legislation (ibid, Separate opinion of Judge Dugard, at para. 18). In the sense that the content of the rule must be set by domestic law, see Pierre-Marie Dupuy and Jorge E Viñuales, International Environmental Law (Cambridge: Cambridge University Press 2015) 70.

71 Pulp Mills, supra note 32, at para. 206.

72 Ibid, at para. 210.

73 Emphasis added.

74 Certain activities, supra note 36, at para. 206. In this regard, the decision of the Court not to find a legal obligation to consult with the public has been defined as “surprising” in the light of the emphasis put by several modern treaties on this point (Cymie R Payne, “Pulp Mills on the River Uruguay” (2011) 105 AJIL 94 et sqq., at 100).

75 Certain activities, supra note 36, Separate opinion of Judge Bhandari, at para. 33.

76 Ibid, at para. 41.

77 Ibid, at paras. 45–46.

* Assistant Professor in International Law, University of Turin; email: .