I. Introduction
Environmental impact assessment (EIA) is widely heralded as a central pillar in the protection of the environment in international law. The obligation to conduct EIAs is found in a wide range of multilateral treaties, as well as forming part of the environmental practices of international organisations.Footnote 1 The obligation to conduct EIAs has been identified as a ‘general’ or customary obligation by the International Court of Justice (ICJ),Footnote 2 the International Tribunal for the Law of the Sea (ITLOS),Footnote 3 and the International Law Commission (ILC).Footnote 4 Despite its pedigree as a firmly established requirement in international law, the obligation to conduct EIAs remains controversial as both a conceptual and methodological matter.
Conceptually, EIAs can be viewed in a highly instrumental manner. They are planning tools that are intended to inform decision-makers of the environmental consequences of planned activities. EIA obligations, on this view, are understood as narrow and technical requirements to produce environmental information.Footnote 5 The obligation is procedural in the sense that the EIA process itself does not determine outcomes, but rests on the assumption that decision makers, if fully informed of the environmental consequences of their proposed activities, will arrive at decisions consistent with environmental goals. EIAs can also be understood in much more deliberative terms.Footnote 6 In addition to generating knowledge about impacts, EIAs also provide a process for those potentially affected by the proposal to understand how their interests may be impacted and have their interests accounted for within the decision-making process. This understanding has a less instrumental posture in that the aim of EIAs is to legitimise decisions impacting the environment in the eyes of those impacted and the public more generally by ensuring that decision-makers have appropriately accounted for a wide range of interests. EIA is, on this view, both a means to an end—harm avoidance, but also an end in itself—legitimation.
How EIAs are understood has important practical implications for how EIAs are carried out. For example, a more technical view of EIA is more likely to view consultation and participation narrowly, emphasising expertise over representation. Consultation focuses on the results of the assessment, not the assessment itself. The scope of consultation is likely to be oriented towards those who are directly impacted, constructing a narrow understanding of the nature of interests at stake in the process. A more deliberative view of EIA understands scientific knowledge and its creation as being more open to contestation, requiring input into how the assessment is undertaken. Moreover, whereas the instrumental understanding of EIA rests on an assumption of normative consensus respecting the balancing of environmental and economic goals, the alternative model accepts a more pluralistic understanding of the normative landscape in which decisions respecting planned activities occur. EIA, in its more deliberative conceptualisation, envisages a dialogical process, as opposed to a one-way conveyance of information from the source State to the affected State.
The conceptual debate has methodological links relating to how EIA obligations are formed in international law. As a State duty, EIA implements other general rules and principles of international environmental law by providing a greater degree of specificity respecting the precise actions a State is required to undertake in the face of potential transboundary harm. For example, in the linked disputes between Nicaragua and Costa Rica that were adjudicated before the ICJ in 2015 (referred to collectively here as the Certain Activities/Construction of a Road Case), the Court drew a link between EIA and a State's due diligence obligations to prevent transboundary harm.Footnote 7 This link was also clearly identified by the Court in its earlier decision in the Pulp Mills Case.Footnote 8 While the precise nature of the relationship between the general obligation and more specific customary rules was the source of some debate among the Judges in the Certain Activities/Construction of a Road Case, the underlying idea that EIA obligations implement due diligence, and that the obligation to conduct an EIA must be understand in light of that obligation, was not called into question.
The question the Court does not address is whether EIA implements harm prevention alone or whether the more specific requirements of EIA are responding to other general duties in international law. In addressing this question, this article argues that the ICJ's analysis of EIA obligations is incomplete because it has ignored EIA's role in implementing the duty to cooperate, notwithstanding the latter duty's equally strong standing in international environmental law. The duty to cooperate is triggered by the same condition that triggers the duty to prevent harm; namely, the risk of significant environmental harm, and thus, must also be implemented by States in the same circumstances. Whereas prevention focuses on the obligation of a State to inform itself of the consequences of its activities, cooperation focuses on the obligation to inform other States and to seek their perspectives on an activity that has the potential to affect their interests. It is further argued that ignoring the duty to cooperate pushes international law towards a thin and technical version of EIA, whereas if the duty to cooperate is brought back in, EIA's deliberative and ‘other-regarding’ nature is more clearly seen. Understanding EIA as implementing both the duty of prevention and the duty to cooperate provides a more coherent understanding of the nature of EIA in international law.
This has important implications for the substance of EIA obligations, and provides a basis for further elaboration of the customary duty. The relationship between EIA and cooperation also provides a more conceptually sound basis for developing new EIA obligations in international treaties by treating more seriously the epistemic and normative challenges that face States in their determination of whether to pursue environmentally risky activities.
This article proceeds from the understanding that the content of the EIA obligation can only be understood in light of the general rules it serves. The critical point here is that in the context of transboundary harm, due diligence is accompanied by the duty to cooperate, and a determination of EIA obligations that proceeds without regard to cooperation, as has been the case in the ICJ, is necessarily incomplete.
To address this point, the article first examines in greater detail the duty to cooperate in the context of transboundary harm, which requires States to notify and consult in good faith potentially affected States. While both general obligations are engaged when States propose to undertake environmentally risky activities—and in this regard, both must necessarily mediate between a source State's sovereign right to engage in economic activities within its territory and an affected State's sovereign right to be free from environmental harm—each employs a distinct logic. The duty to prevent harm seeks to identify and delineate the extent of each State's sphere of activity: States do not have to account for or accommodate the interests of other States so long as their activities remain confined to their sovereign domain; they must simply determine the extent of their own rights. The duty to cooperate treats the problem of transboundary harm as a shared problem. It recognises that modern environmental problems are not easily and neatly allocated between States, and instead invites States to balance their interests in light of the interests of others.
The next section explores the implications of the duty to cooperate for the legal construction of the customary obligation to conduct EIAs. Interpreting EIA in light of the duty to cooperate allows for a further refinement, and in some cases reconsideration, of a number of aspects of EIA procedures, including the obligations of States when they disagree over whether an activity raises a risk of significant harm, the timing of notification, the duty to give reasons for decisions respecting planned activities and the potential remedy for a breach of the obligation to conduct an EIA.
The conclusion returns to the nature of EIA obligations themselves and the potential for EIA obligations to promote decisions that better account for and help construct community interests in international law. The ‘other-regarding’ obligation of States to account for the rights and interests of those States affected by their activities has been an issue of long-standing interest to international lawyers,Footnote 9 but less attention has been paid to specific legal processes that implement these obligations. EIA is one such tool, but realising EIA's deliberative potential requires careful attention to its normative antecedents.
II. THE PRINCIPLES UNDERLYING THE DUTY TO CONDUCT AN EIA
The Certain Activities/Construction of a Road Case illustrates some of the difficulties associated with identifying a customary obligation to conduct EIAs. The issue of a duty to conduct an EIA arose in both disputes. In the Certain Activities case, the activity in question was the dredging of a channel by Nicaragua in a disputed area that was subsequently determined to be Costa Rican territory. In the Road Case, the activity was the construction of a road by Costa Rica in their territory, but running along the San Juan River, which forms the border between the States. In both cases, it was alleged that a transboundary EIA was required and was not conducted. There was no existing treaty obligation between the parties that addressed transboundary EIA, so the ICJ was required to determine whether a customary rule existed.
The starting point for the ICJ was its earlier decision in the Pulp Mills case, where the Court was required to determine whether an obligation to conduct EIAs ought to be read into a treaty provision obligating the parties ‘to protect and preserve the aquatic environment’.Footnote 10 In finding that an EIA obligation ought to be read into the duty to prevent environmental harm, which the Court characterises as ‘an obligation to act with due diligence’,Footnote 11 the Court noted that:
… the obligation to protect and preserve, under Article 41(a) of the Statute, has to be interpreted in accordance with 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 to undertake an environmental impact assessment 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. Moreover, due diligence, and the duty of vigilance and prevention which it implies, would not be considered to have been exercised, if a party planning works liable to affect the regime of the river or the quality of its waters did not undertake an environmental impact assessment on the potential effects of such works.Footnote 12
The precise status of the EIA obligation is ambiguous due to the fact that the Court in Pulp Mills was interpreting a treaty, not expressly determining a customary rule. Nevertheless, the ICJ refers to the EIA duty as a ‘requirement under general international law’ and states that such a duty is implied by due diligence (which, in this case, was a treaty obligation but was also recognised as a customary rule).Footnote 13
In elaborating on the substance of an EIA duty, the Court maintained that the specific content is a matter for domestic law but ought to be responsive to the facts of the specific case and the requirement for due diligence.Footnote 14 The Court is not explicit about its methodology for arriving at the conclusion that there exists in ‘general international law’ a duty to conduct EIAs.Footnote 15 The approach appears to be one of evolutionary interpretation, whereby the treaty obligation to prevent harm is read in light of a subsequent practice respecting EIAs.Footnote 16 The Court does not specify that this subsequent practice amounts to a customary obligation, and does not examine State practice and opinio juris in relation to EIA, but rather seems to derive the obligation from due diligence.Footnote 17 In effect, Uruguay is required as a matter of due diligence to protect the environment, this necessitates that it must inform itself of the consequences of its activities, and EIA is the manner by which this is done.
The court in Pulp Mills also elaborates on the role of EIA in the context of a State's duty to notify and consult. Here again, this duty (to notify) was treaty-based. On this point, the Court notes that where there is an obligation to notify another State respecting potential transboundary harm, the EIA forms the basis of the notification, since only by conveying the full details of the assessment can the affected State assess how it is impacted.Footnote 18 As such, the EIA must be conveyed prior to any approval.Footnote 19 The duties, however, are presented as being independent and sequential. Notification and consultation are only triggered where an EIA discloses a risk of significant transboundary harm, although in the Pulp Mills Case, Uruguay was under an additional treaty obligation to notify the river commission, CARU, once it had some preliminary basis to believe its activity might cause significant harm to another State.Footnote 20
The ICJ in the main judgment of the Certain Activities/Construction of a Road Case cites its reasoning from Pulp Mills, and again links the duty to conduct EIAs directly to a State's due diligence obligations.Footnote 21 While the decision has certainly been interpreted as treating EIA as custom,Footnote 22 the Court maintains its ambiguity respecting the customary status of EIA duties, and again does not examine State practice and opinio juris. The reasoning is almost identical to that employed in Pulp Mills:
Thus, to fulfil 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 EIA.Footnote 23
There was disagreement among the Judges of the Court as to the precise nature of the EIA obligation, with Donoghue and Owada viewing EIA simply as evidence of due diligence, but not forming the basis of an independent obligation, while Dugard was more emphatic that EIA obligations were independent and supported by State practice and opinio juris. Despite their methodological differences, the various judges arrived at the same basic set of requirements. Because EIA obligations implement the requirement for due diligence by responding to the need for greater specificity, construction of the obligation must be made with recourse to the underlying principle. In this regard, due diligence plays an interpretive role in providing the underlying purpose for conducting EIAs, but it also plays a deeper, more substantive role by providing the standard by which the adequacy of EIAs will be assessed. Very clearly EIA only addresses certain aspects of the broader requirement to prevent harm; namely, it applies to the possible harms that may arise from planned physical undertakings, and it applies to the planning stages of a project.Footnote 24
What the Court does not turn its attention to, however, is whether EIA requirements only implement the duty to prevent harm or whether EIA also is intended to implement other general duties that arise in relation to planned activities. If, however, the duty to prevent harm is triggered, those same conditions—the decision to undertake a project that has a risk of causing significant transboundary harm—also trigger the duty to cooperate. Like the harm principle, the duty to cooperate is central to mediating the competing sovereign interests of States when one State proposes an activity that can harm another. Birnie, Boyle and Redgwell note, ‘if due diligence is the first rule of transboundary environmental risk, cooperation is the second’.Footnote 25
The duty to cooperate is equally well established in international environmental law as the duty to prevent harm. The duty is present in Principle 24 of the Stockholm Declaration,Footnote 26 and subsequently in Principle 27 of the Rio Declaration,Footnote 27 as well as numerous treaties, including the United Nations Convention on the Law of the Sea (UNCLOS),Footnote 28 the Convention on Biological DiversityFootnote 29 and the Ozone Convention.Footnote 30 The duty to cooperate has both a general formulation (as seen in Principle 27 of the Rio Declaration), which is open-ended in its structure, but includes matters such as general obligations to cooperate with other parties and international organisations, and to promote and cooperate in scientific and technical matters.Footnote 31 There is also a more specific obligation to cooperate in the context of shared resources and transboundary harm, which follows the dictum in the Lac Lanoux caseFootnote 32 and is set out in Principle 19 of the Rio Declaration:
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.Footnote 33
The obligation to cooperate through notification and consultation is enshrined in numerous treaties involving watercoursesFootnote 34 and transboundary pollution,Footnote 35 and has been accepted as a customary rule of international law by various codification bodiesFootnote 36 and tribunals,Footnote 37 including the ICJ in the Certain Activities/Road Case. Footnote 38
The structure of the duty to cooperate, which originates in the Lac Lanoux arbitration, is that a source State is obligated to notify and consult a State potentially affected by a proposed activity, but the affected State does not have a right to exercise a veto over the proposal.Footnote 39 The obligation requires the source State to take the interests of the affected State into consideration. The trigger for the duty to cooperate is the same as the no-harm principle. As such, both duties are engaged when there is a risk of significant transboundary harm. EIA processes, which govern when notification should occur, who should be notified, what information should be exchanged and structures the consultations themselves, are the primary means of implementing the duty to cooperate in relation to planned activities.
The purpose of the duty to cooperate is instrumental in the sense that it is oriented towards the prevention of harm.Footnote 40 But cooperation should not be viewed as merely implementing due diligence. In the environmental context, the duty to notify arises out of the duty to warn, identified in the Corfu Channel Case as a means to avoid affecting the interests of other States, and the need to jointly manage shared natural resources.Footnote 41 Thus, for example, in the Pulp Mills Case, cooperation was identified as a means to ‘jointly manage the risks of damage to the environment’.Footnote 42 Whereas the structure of harm prevention is directed towards delineating spheres of territorial integrity, cooperation arises from circumstances involving shared interests.Footnote 43 The mediating role of cooperation is evident in the ILC's Draft Articles on Prevention of Transboundary Harm from Hazardous Activities, which draws on the law of transboundary watercourses in identifying an obligation to ‘seek solutions based on an equitable balance of interests’.Footnote 44 The implication being that the determination of rights in the context of transboundary harm is contingent and involves a degree of political give and take.
Fundamentally, the duty to cooperate is oriented towards reconciling the competing sovereign interests that arise when one State potentially impacts the environment of another. Judge Wolfrum, in the Mox Plant Case, emphasised the shift towards community interests that cooperation entails in these terms:
The duty to cooperate denotes an important shift in the general orientation of the international legal order. It balances the principle of sovereignty of States and thus ensures that community interests are taken into account vis-à-vis individualistic State interests.Footnote 45
If cooperation is understood as a handmaiden to prevention, it may be too easily argued that States acting reasonably are not required to notify and consult other States, so long as they take reasonable steps to avoid the harm. Due diligence is not necessarily ‘other-regarding’, whereas the duty to cooperate necessitates taking steps to understand the impacts from the perspective of the affected State. Since the duty to cooperate operates alongside due diligence, the two duties ought to be interpreted in a manner that is complementary and mutually reinforcing.
The obligations that arise in the face of potential impacts is framed by Benvenisti in terms that very much match the structure of the duty to cooperate:
The obligation to weigh the interests of foreign stakeholders does not necessarily imply an obligation to succumb to those interests, and does not even require full legal responsibility for ultimately preferring domestic interests in balancing various opposing claims. It does not necessarily imply that sovereign discretion should be subject to review by third parties such as foreign or international courts, ones that would replace the sovereign's discretion with their own. What it does imply as a minimum, however, is that sovereigns—whenever they are considering the adoption and pursuit of policies that potentially affect foreign stakeholders or, more generally, global welfare—give due respect to those foreign and global interests.Footnote 46
The obligation to ‘give due respect’ is operationalised by providing potentially affected States with sufficient information to enable them to understand how their interests may be affected and to provide them with an opportunity to express their views and have them accounted for in the decision-making process. The requirement for good faith, which has been identified as governing the duty to cooperate,Footnote 47 has been described as entailing ‘a genuine intention to achieve a positive result’Footnote 48 and as requiring something more than ‘mere formalities’.Footnote 49 EIA addresses good faith by providing processes for consultation and, in some circumstances, the provisions of reasons that are responsive to the concerns raised.Footnote 50
The duty to cooperate on its own does not give rise to EIA, but it does provide a further basis from which a customary EIA rule derives its content. As such, treaties containing EIA instruments contain requirements to notify and consult on the basis of the EIA. The particular modalities of notice and consultation vary depending upon the context and the degree of specificity in the instrument. The Espoo Convention, for example, requires notification of listed activities that are likely to cause a significant transboundary impact ‘as early as possible’ and links notification to the obligation to consult.Footnote 51 Consultation occurs on the basis of the EIA documentation and is aimed at the elimination or reduction of the impacts.Footnote 52 The source State is not required to resolve the concerns or necessarily accommodate the affected State, but it must account for the outcome of the consultations.Footnote 53 Even those treaties that contain unelaborated obligations to conduct EIAs (that is, the requirement simply requires an EIA be conducted where there is a likelihood of environmental harm, but does not provide details of the required procedures in the same manner as the Espoo Convention or the Antarctic (Madrid) Protocol), contain explicit requirements to notify and consult.Footnote 54
If EIAs are the central mechanism by which the duty to cooperate is implemented, then the relationship between EIA and the duty to cooperate should be viewed in a similar manner as the relationship between EIA and due diligence. From an interpretive standpoint, the duty to cooperate supplies a quite different rationale for conducting EIA that relates not to harm prevention, but to the right of States to be treated with ‘due respect’. Bringing cooperation back into the EIA obligation provides further insights into several key areas of continuing controversy and ought to provoke some reconsideration of the existing rules in order to satisfy the requirements of cooperation.
III. REASSESSING CUSTOMARY LAW IN LIGHT OF THE DUTY TO COOPERATE
Since the relationship of the duty to cooperate to EIA is similar in its structure to due diligence, the approach to reassessing the customary rule follows the same approach. At a minimum, customary EIA rules ought to be interpreted so as to give effect to the more general requirements of cooperation (and harm prevention). Treaty-based EIA obligations similarly provide instantiations of the international community's understanding of how the duty to cooperate ought to be implemented in the context of planned activities, and thus, while not binding, still retain interpretive relevance.
A. The Threshold
A fundamental difficulty with EIA is that the threshold condition to trigger the obligation to conduct an EIA—the risk of significant environmental harm—must itself be determined through some form of assessment—thus, the very process that determines the level of harm is only triggered when that level is exceeded. The difficulty is compounded by the fact that the threshold, while intended to provide an objective standard,Footnote 55 is notoriously vague, and because the determination is a function of risk, it involves a level of subjectivity. Unsurprisingly, whether a proposed activity triggers an obligation to conduct an EIA is at the root of many interstate disputes respecting EIA. For example, both disputes in the Certain Activities/Road Case involved disagreements over whether the threshold of a risk of transboundary harm had been met.
On the facts of the Certain Activities dispute, the Court found that the conditions to trigger an EIA were not present, whereas in the Road Dispute, those conditions were present. The complication in both disputes was that the obligation requires an ex ante consideration of risk, but the case arose after the projects in both cases had been constructed. Thus, the parties tended to offer evidence of no actual harm as proof that the risk was not present. The distinguishing feature between the disputes is the Court's finding of evidence that Nicaragua turned its mind to whether a risk was present (by conducting a domestic impact assessment, which disclosed no risk), while Costa Rica had undertaken no such preliminary inquiry. But the Court is not clear whether a lower threshold than risk of significant transboundary, which the EIA in intended to assess, is required.
Where States disagree over whether an activity presents a risk of significant harm, due diligence imposes no additional obligation on the parties. In order to avoid such an impasse, the Espoo Convention identifies a set of listed activities, for which a potentially affected State may trigger a third party inquiry process in the event that they disagree with the source State's determination that the activity does not present a risk.Footnote 56 The process, conducted by an inquiry commission, investigates and advises the parties on whether the significance threshold is met.Footnote 57 Even for unlisted activities, a State remains under an obligation to ‘enter into discussion’ with a potentially affected State on whether an activity requires a transboundary EIA.Footnote 58 In other words, in the absence of agreement over whether an activity triggers an assessment, a source State must still cooperate in good faith to resolve disagreements over whether transboundary EIAs must be undertaken. The approach of the ILC similarly requires a source State to provide a ‘documented explanation setting forth the reasons for such finding’ of no significant impact and, if that response fails to satisfy, to enter into consultations, during which the source State should take steps to minimise any potential harm to the affected State.Footnote 59 However, the ILC also makes it clear that an affected State must have a ‘serious and substantiated’ belief that significant harm may arise before a duty to consult arises.Footnote 60 The approach gives effect to the idea of ‘due respect’ by imposing obligations of justification on both sides.
These obligations in the face of disagreement over the presence of risk are further supported by the practice of ITLOS, which, in the context for requests for provisional measures, has required Parties to cooperate by continuing to exchange information and consult with one another, notwithstanding that the threshold requirement for harm had not been proven.Footnote 61
The significance of these examples is not that the duty to cooperate lowers the threshold for EIAs or notification, but rather that the duty to cooperate is engaged when the source State has knowledge that its proposed activities may affect another State. That knowledge may arise by virtue of its own investigations or by being informed of a potential impact by an affected State. However, once the source State has this knowledge, it cannot simply ignore the affected State's claim, as to do so would not be to act in good faith. (Good faith is, of course, reciprocal, which explains why the affected State is bound to only raise ‘serious and substantiated’ claims.) Whereas due diligence, which is rooted in the presence of a bio-physical threat, triggers obligations in an all or nothing fashion (either the threat exists or it does not), the duty to cooperate, which is rooted in mutual respect, informs the relations between States throughout their engagement on the issue.
It is important to recognise at a stage prior to an EIA being conducted, neither the source State nor the affected State has full evidentiary basis to determine the presence of a risk of significant harm. There is no compelling reason why the source State's determination of risk should prevail in the absence of an assessment. If they have evidence that the activity poses no threat, then it ought to be shared. And if they do not have evidence, it is in keeping with their obligations to treat other States with respect, to hear and weigh the concerns raised by a potentially affected neighbour.
B. Timing of Notification
EIA, as practised in domestic and many international contexts, is conducted as a collaborative exercise between the proponent and affected stakeholders. To this end, notification of intent to undertake an EIA is provided at the start of the project, and consultations are engaged in during the early phases of the EIA when the scope of the study is being determined. Providing for input during the scoping stage recognises that affected or interested persons ought to have an opportunity to express their views respecting the scope and design of the study.Footnote 62 Early consultation can promote acceptance of the study's approach and substance by providing those affected with a degree of authorship over the process. More generally, early consultation during the scoping stage is understood as a best practice.Footnote 63
EIA, when understood as a function of due diligence alone, leads to a sequential relationship between assessment and notification and consultation. This point is made at several junctures in the Certain Activities/Road Case. In the Certain Activities dispute, the ICJ held that because Nicaragua was not under an obligation to carry out an EIA (due to the absence of a risk of significant harm), it was not required to notify or consult with Costa Rica.Footnote 64 In the Road Case dispute, the ICJ is explicit about the sequential nature of the obligations:
The Court reiterates its conclusion that, if the environmental impact assessment confirms that there is a risk of significant transboundary harm, a State planning an activity that carries such a risk is required, in order to fulfil its obligation to exercise due diligence in preventing significant transboundary harm, to notify, and consult with, the potentially affected State in good faith, where that is necessary to determine the appropriate measures to prevent or mitigate that risk.Footnote 65
Here a clearer picture of the relationship between due diligence and notification arises. Notification serves the obligation to prevent harm insofar as when the risk is confirmed and measures must be taken to avoid the harm, and these measures ought to be determined collectively. The effect of this sequential understanding of EIA and notice and consultation is that the EIA process remains entirely source-State driven. The source State determines the scope of the study, whether and which alternatives are assessed and even how the project itself may be identified. The affected State has no role in the EIA until it is complete.
However, the duty to cooperate shares the same trigger as the duty to conduct an EIA. On that basis, the obligations that flow from cooperation should be engaged at the same time the EIA commences. The importance of cooperation at the early stages of a potential impact upon another State's interests is recognised in Principle 19 of the Rio Declaration, which requires notification and consultation ‘at an early stage and in good faith’.Footnote 66 The Espoo Convention adopts an approach consistent with Principle 19, whereby it is anticipated that notification will be undertaken by the source State ‘as early as possible and no later than when informing its own public’.Footnote 67 The source State is required to provide relevant information to the affected State and clearly anticipate that the notice is intended to allow the affected State to determine whether they wish to participate in the EIA.Footnote 68
Judge Donoghue acknowledges that there may be circumstances where notification at a stage prior to the completion of the EIA may be beneficial, such as where ‘input from a potentially affected State may be necessary in order for the State of origin to make a reliable assessment of the risk of transboundary environmental harm’.Footnote 69 Donoghue goes on to note that such a circumstance arose on the facts of the Construction of a Road dispute, since only the affected State, Nicaragua, had access to the San Juan River, the main ecosystem component of concern.Footnote 70
Donoghue's point here is that the flexibility of locating the obligation in due diligence, not in EIA, is preferable given the contextual demands of assessment, notification and consultation. However, looking at the reasons for early notification from a due diligence perspective may limit that right to instances, such as the Construction of a Road Case, where involving the affected State facilitates some informational deficit. As a function of cooperation, the purpose of early notification is broader and less instrumental. EIA processes typically provide the flexibility that due diligence and cooperation require, but are likely better able to signal more clearly how those discretionary decisions ought to be exercised and, therefore, create more predictable and stable expectations.
Donoghue identifies a further reason why affected State participation in the EIA process may be desirable, noting that it may be important to receive the affected State's views on the sensitivity of the environment or procedural details of the EIA. Here her portrayal of EIA more closely fits how the process tends to unfold in domestic settings. The substantive requirement, avoidance of significant environmental harm, is contested due to the ambiguity of the standard and the presence of uncertainty, particularly in predictive processes. There is no ‘right’ outcome, except at the poles, where significant harm is clearly found to be present or not. Instead, decision-makers must balance the competing perspectives, which will be informed by the respective interests of the parties. Participation by affected States in EIAs facilitates this more dialogical process. It acknowledges that decisions must account for the perspectives and interests of affected States.Footnote 71 This justification, as Donoghue herself notes, is not rooted in due diligence, but rather flows from the demands to due respect.Footnote 72
C. The Duty to Give Reasons
The obligations on source States once an EIA (including consultations) is completed are not clear. Ultimately, a decision must be made and conveyed to the affected State. The procedural nature of EIAs does not require that the source State mitigate harm to the satisfaction of the affected State. But is the affected State entitled to receive reasons that are responsive to its concerns? As a due diligence obligation, such a right is doubtful, as giving reasons will only indirectly influence whether harm occurs or not. As a function of due diligence, EIA is not oriented towards justification. The requirement of reasonableness that inheres in due diligence relates to reasonable steps to avoid harm, as opposed to reasonable steps to account for and treat seriously the views of other States.
Nonetheless, international EIA obligations often require the provision of reasons. For example, the UNEP EIA Goals and Principles require that final decisions be given in writing and provide reasons.Footnote 73 The Espoo Convention requires not only reasons, but that those reasons take ‘due account’ of the EIA report and the comments received on the report by affected persons and States.Footnote 74 As noted, reasons are central to good faith as the responsiveness and adequacy of reasons are the basis by which an affected State can assess whether their concerns were taken seriously.Footnote 75 The duty to give reasons arises where a State is required to account for the position of another State in a non-arbitrary way. For example, the Appellate Body in the Shrimp/Turtle case identified the US government's failure to give reasons for its certification scheme as contributing to their finding that the scheme amounted to ‘arbitrary discrimination’.Footnote 76
A fundamental concern in EIA practice is the potential for the entity conducting the EIA to pay lip-service to the concerns raised by affected persons or States. Good faith stands at the centre of cooperation because in order for consultations to be meaningful, the parties must engage one another with a genuine intention to achieve a mutually satisfying result. Nevertheless, because the source State retains a high degree of discretion to proceed with a project in the face of objections, reasons provide the basis upon which an affected State (and third parties) can police genuineness.
The duty to give reasons aligns the international requirement to conduct EIAs with recognition that discretionary decisions in the international legal system must attend to due process considerations.Footnote 77 In domestic EIA jurisprudence, US Courts have emphasised that the obligation on decision-makers in EIA processes to demonstrate, through ‘sufficient discussion of relevant issues and opposing viewpoints’, that they have arrived at a ‘reasoned decision’.Footnote 78 Similarly, in domestic administrative law contexts, citizens are often entitled to decisions from administrative officials that adhere to minimum requirements of rationality.Footnote 79 The point here is not that domestic standards of administrative law ought to be transplanted into international practice, but that the contexts share a common rationale in ensuring that the views of those affected by discretionary decisions are properly accounted for.Footnote 80 The obligation to give reasons guards against arbitrariness, which goes to the heart to ‘due respect’. While procedural protections are less entrenched in general rules of international law, responsiveness to legal criteria is central to the legitimation of decisions governed by international law.Footnote 81
D. Remedy for Breach
One challenging aspect of the decisions in Pulp Mills and the Certain Activities/Road Case was the question of what remedy to provide in the face of a breach of procedural obligation. In Pulp Mills, the Court found that Uruguay had in fact breached its obligation to notify and engage in good faith negotiations with Argentina. Argentina sought the dismantling of the mill, which had been constructed by the time the matter came before the ICJ. Argentina based this request on its entitlement to restitution for the unlawful act, which in its view required the flawed decision to proceed with the project to be voided, placing Argentina in the position they would have been in had the breach not occurred. The Court in denying this request notes that the breach was procedural in nature and did not result in substantive harm to Argentina. On that basis, Argentina was not entitled to a remedy beyond the declaration of the wrongful conduct by Uruguay. In the Road Case dispute, the Court relies on this reasoning to deny a substantive remedy to Nicaragua for Costa Rica's failure to conduct an EIA.
The assumption that underlies the Court's approach is that, as long as the substantive rights of a State remain unassailed, then the aggrieved party has lost nothing, or at least very little. If process is simply a means to an end, then so long as the end is maintained, the importance of the means falls away.Footnote 82
In their separate reasons in the Pulp Mills case, Judges Al-Khasawneh and Simma, provide a more nuanced understanding of the relationship between process and substance:
A final observation: in matters related to the use of shared natural resources and the possibility of transboundary harm, the most notable feature that one observes is the extreme elasticity and generality of the substantive principles involved. Permanent sovereignty over natural resources, equitable and rational utilization of these resources, the duty not to cause significant or appreciable harm, the principle of sustainable development, etc., all reflect this generality. The problem is further compounded by the fact that these principles are frequently, where there is a dispute, in a state of tension with each other. Clearly in such situations, respect for procedural obligations assumes considerable importance and comes to the forefront as being an essential indicator of whether, in a concrete case, substantive obligations were or were not breached. Thus, the conclusion whereby non-compliance with the pertinent procedural obligations has eventually had no effect on compliance with the substantive obligations is a proposition that cannot be easily accepted. For example, had there been compliance with the steps laid down in Articles 7 to 12 of the 1975 Statute, this could have led to the choice of a more suitable site for the pulp mills. Conversely, in the absence of such compliance, the situation that was obtained was obviously no different from a fait accompli.Footnote 83
As a matter of due diligence, the approach of the main judgments is logically sound. In the absence of harm, there is no material loss, and therefore no basis for reparations. However, understood as a breach of the duty to cooperate, loss of the affected State looks slightly differently. The loss may be characterised as a deliberative one—in the sense that where a State fails to be notified and consulted, then it has lost an important right to understand how its interests might be affected, to have its views accounted for, and to affect the outcome. Second, it follows that the affected State also loses the opportunity to have the project proceed as it might have done had it been properly consulted. To assess the loss, a court must actually look at the degree or quality of the procedural breaches to assess the counterfactual. The absence of good faith in the dealings between the parties ought to influence this assessment.
Under a due diligence standard, the loss of a better outcome is not relevant as the affected State is only entitled to be free from harm, as objectively determined. But the duty to cooperate treats the ambiguity and subjectivity of the substantive right seriously. Not by accepting the affected State's position, but by treating the determination of harm as a matter to be determined collectively, or at least not without due regard for each side's understanding.
The problem is that courts cannot easily undo what has been done, particularly in cases where the activity in question has already been undertaken. Argentina's suggested remedy of dismantling the mill is not without precedent. The ICJ had opportunity to note in another case that:
if it is established that the construction of works involves an infringement of a legal right, the possibility cannot and should not be excluded a priori of a judicial finding that such works must not be continued or must be modified or dismantled.Footnote 84
Dismantling, however, suffers from a proportionality problem, a consideration in granting restitution, particularly because the loss is difficult to quantify. The Court has acknowledged in other proceedings that reparations must be determined in light of the particular circumstances of the case.Footnote 85 In this regard, the lack of substantive harm is one factor to be considered but should not be determinative of the issue of adequate reparation.
The difficulty in remedying procedural breaches places considerable importance on the granting of provisional measures. In this case, provisional measures were denied on the basis that the procedural or substantive breaches would not result in irreparable harm since physical harm to the environment is potentially compensable.Footnote 86 But viewed as a matter of cooperation, the harm that arises if a planned activity proceeds in the face of inadequate consultation may be viewed as a substantive loss that is much harder to cure through compensation, militating in favour of the application of provisional measures to preserve the affected State's right to be treated with due respect.
IV. CONCLUSION
The shift in international law, identified by Judge Wolfrum, towards community interests,Footnote 87 is perhaps best seen as an extension of an existing approach that governs shared resources to issues, such as transboundary harm, traditionally governed with reference to sovereign interests. The shift, however, is a modest one. Instead of requiring States to abandon self-interest in favour of international solidarity, it simply requires that they take steps to understand and account for the interests of those States potentially affected by their activities. EIA is best understood as a procedural mechanism that both reflects and gives effect to this shift.
Viewing EIA solely through the lens of harm prevention provides an incomplete, and out-of-date, understanding of the role and structure of EIA in international law. Out-of-date in the sense that an international EIA obligation rooted in harm prevention alone holds on to epistemic and normative presumptions that no longer hold sway in either domestic or international settings. Determining spheres of sovereignty with reference to vague standards such as a ‘significant harm’ has proven difficult, especially in light of the high degree of scientific uncertainty that characterise complex ecological systems. As such, the harm principle's more formal and deterministic approach of delineating areas of unfettered State autonomy has given way to reconciling competing interests through reasoned consultation. The point here is not that it is futile to assess activities and determine whether they will cause significant environmental impacts, but rather to recognise that those determinations often involve a measure of political choice, in the sense that science and law cannot provide a complete answer.Footnote 88
In order to legitimise decisions on planned activities made by one State that may impact the interests on another, the duty to cooperate requires, at a minimum, that those interests be taken into account. The solution is not perfect by any stretch. Adherence to the procedural requirements of cooperation will not in and of itself render the underlying decision acceptable to the affected State. Moreover, the duty to cooperate, which is an obligation owed to States, does not address the extent to which obligations of consultation may extend to the public of the affected State or, indeed, elsewhere.Footnote 89
But if we are to take the idea of community interests—which are by their very nature socially determined—seriously, this entails a commitment to construct those interests dialogically. Bringing cooperation back into EIA processes provides more specific procedural protections to give effect to both harm prevention and due respect of affected State interests.
