4.1 Introduction
International environmental law (IEL) is a fairly recent addition to public international law.Footnote 1 International courts and tribunals so far have had few opportunities to discuss its customary aspects, yet they have, in recent years, ruled on some issues central to IEL. This chapter focuses on the interpretative efforts of, mainly, the International Court of Justice (ICJ) but also looking into ad hoc tribunals when necessary, that is, whenever they venture into customary international environmental law. An analysis of all international environmental law judgments would clearly yield some useful results but it cannot be successfully undertaken within the confines of these chapters. Another choice that limits the scope of the chapter is to look at two customary IEL obligations: the no-harm rule and the obligation to conduct an environmental impact assessment (EIA). Therefore, it is far more useful for our purposes to focus on those rules that have been discussed by the ICJ and by tribunals with general international law jurisdiction
Having delimited the chapter’s content regarding IEL, it is apposite to also sketch the theoretical contours of the chapter regarding interpretation. The starting point on any discussion on interpretation of customary rules is to distinguish it from the process of identification of these rules.Footnote 2 After they identify the existence a customary rule (such as no-harm or EIA),Footnote 3 courts also often – but not always – engage with the determination of its content.Footnote 4 Arguably, this shows that there are two steps in the process: step one is the identification of the existence rule as custom and step two is the determination of its content. This step can be regarded as interpretation of the rule.Footnote 5 The chapter focuses on this second step.
It looks like, at least at first sight, that identification of the existence of the rule (that is the first step) is simply a snapshot of the rule at a particular point in time. Then the question becomes what are the possibilities that a court has to determine regarding its content and what normative weight each of them carries. A first option is the possibility that the court has identified the rule and, in so doing, it has also determined its content. Another, more convincing, possibility is that the court identifies the rule and has to give it some meaning, has to explain it, in order to apply it to the facts of the case. This is the interpretative process.Footnote 6 Such an approach means that there are some elements of the rule that cannot be considered as part of the identity of the rule but as interpretation of the rule, and it is the interpretation that gives the rule its content. One can think of a case where the rules of customary EIA are being fleshed out by courts through references to the rich material to be found in the Conference of the Parties of IEL treaties, for example.Footnote 7 This construct can prove to be particularly useful especially since IEL is a technical and rather fast evolving area of law, and courts can take the opportunity to enrich their enunciation of a customary rule through interpretation that conforms to the accepted IEL standards.
It appears that international courts in the interpretative process of IEL often oscillate between evolutive and regressive interpretations of the relevant customary rules. This requires some explanation. The evolutive interpretation takes into account developments that appear to affect the rule in question and push it to an interpretive point where it catches up with these developments.Footnote 8 Regressive interpretation involves a backward-looking approach where the courts are content with offering an interpretation that diverges from the standards surrounding the rule rendering a more conservative version of it.Footnote 9 Both interpretative methods form, among others, a central finding of the chapter.
The chapter proceeds by accepting the idea that customary international law can be interpreted, for the most part, using the methods of interpretation propounded in the Vienna Convention on the Law of Treaties.Footnote 10 The range of methods the courts can use in interpreting customary IEL is no different than that applicable to other branches of international law.Footnote 11 Of course, the distinct possibility that the development of a rule through time may not simply affect its interpretation but may yield a new rule through fresh practice and opinio juris is acknowledged.
In the second section, the chapter sets the stage by introducing IEL and its peculiarities in relation to customary law. The third section is divided in two parts. The first one discusses customary law interpretation of the no-harm rule while the second part analyzes the interpretation of the obligation to conduct an EIA. The fourth section presents some concluding remarks showing that international courts do not have a consistent approach to the interpretation of international environmental law. They rather oscillate between expansive and restrictive interpretative efforts.
4.2 The Peculiar Case of IEL
Before embarking on the analysis of the ICJ’s attempts at interpretation of customary IEL, it is necessary to place the primary rules of IEL in the framework of international dispute settlement procedures. Every branch of international law (or any kind of law for that matter) tends to be treated as a ‘special case’ by the literature devoted to that branch.Footnote 12 For the purposes of this chapter, it is irrelevant why this happens, maybe it is simply the manifestation of a natural knack for differentiation or for carving a separate niche within the broader picture of public international law. It is most probably a corollary of the development of specific areas of international law. With that caveat in mind, it is submitted that IEL has some attributes that, at least for the purposes of the subject matter of this chapter, distinguish it from most of the other branches of public international law.
The most obvious one is that IEL is mainly law produced through written agreement. Depending on the source, the number of IEL treaties varies, but it is safe to say that it lies in the thousands.Footnote 13 That is counting everything from bilateral treaties regulating the use of a shared resource (e.g. the River Uruguay in Pulp Mills )Footnote 14 to global instruments with near universal acceptance (e.g. Convention on Biological Diversity).Footnote 15 It is true that many of these instruments sometimes eschew clear-cut categorization. Is the Convention on International Trade of Endangered SpeciesFootnote 16 an environmental protection treaty, a trade regulating treaty, or something in between? Be that as it may, it is fair to admit that their sheer number makes it abundantly clear that IEL is predominantly shaped by them. On top of that there are some treaties that, although they do not fall strictly under the rubric of IEL, have had an immense impact on the field like, most importantly, the UN Convention on the Law of the Sea (LOSC).Footnote 17 At the time of its adoption, it was the instrument with the most comprehensive environmental protection section that had ever appeared.Footnote 18
There are also two other places, besides treaties, where the heavy influence of the written word on the formation and content of IEL shows. First, the soft law of the Stockholm and Rio Declarations.Footnote 19 It is undeniable that both Declarations shaped many aspects of IEL.Footnote 20 The Stockholm Declaration was the first document that attempted to provide for a complete account of the principles of international environmental law.Footnote 21 It is obvious that the Declaration is not binding. Yet, it is equally obvious, mainly from the language of the text,Footnote 22 that it aimed at setting the scene for further development of the principles into one or more legally binding documents. The Rio Declaration, concluded twenty years after Stockholm, showed that States had not, even at that time, reached a point where they could agree to a binding document on general IEL. Nonetheless, the precautionary principle, the principle of prevention and the principle of sustainable development were all proclaimed in these documents.Footnote 23 Their role in the formation of both customary international environmental law and in shaping the content of many international environmental treaties cannot be overstated.
The second place where the influence of the written word on international environmental law is undeniable is the work of the International Law Commission (ILC) of the United Nations through the Articles on the Prevention of Transboundary Harm from Hazardous Activities and (ILC Prevention Articles) and the Principles on Allocation of Loss in the Case of Transboundary Harm from Hazardous Activities (ILC Allocation of Loss Principles).Footnote 24 While the ILC Allocation of Loss Principles represent a light prescriptive approach by generalizing the special rules on civil liability,Footnote 25 the ILC Prevention Articles mainly codify customary international environmental law.Footnote 26
Both the declaratory instruments and the work of the ILC are not binding as such on States. They, nonetheless, exert an immense normative pull. The principles contained in the former have found their way in numerous ways in international environmental treaties while the latter, through the authoritative status of the ILC, have been relied upon by courts and States alike.Footnote 27 Parts of both sets of documents codify customary law, some have become customary law, no less because they have been included in these documents, and others still exist in the twilight of lex ferenda.
The second peculiarity of IEL is that it does not have a constituent binding international instrument that sets out the basic general rules. In the field of human rights there are major international conventions as well major regional conventions that share a common vocabulary and, in a way, contain the vast majority of human rights in fairly similar manner.Footnote 28 The LOSC also sets out the bulk of the rules that govern the oceans. In much the same way, the World Trade Organization (WTO) sets out the rules by which States conduct trade across the globe.Footnote 29 This is important because it means that, despite the vast number of IEL treaties, there is not a single place where one can look and identify the main rules that permeate the whole of IEL. This is reasonable. There are numerous components of the environment that merit separate, detailed attention.Footnote 30 At the same time, there are numerous environmentally harmful substances that also have to be regulated separately and in detail.Footnote 31 The level of detail required by most environmental regulation simply cannot be attained through a general instrument.
At the same time, it is submitted that an instrument containing the general principles of IEL and a host of generally accepted procedural, preventive rules that cut across all fields of environmental protection would be a welcome development. First, because it would crystallize relevant customary rules providing a basis for analysis and interpretation by courts and tribunals. This would bypass the process of affirming the existence of a customary rule and then going about defining its content.Footnote 32 Second, it would provide a starting point to courts and tribunals whose jurisdiction is limited to a single convention. The International Tribunal of the Law of the Sea (ITLOS) and the Arbitral Tribunals constituted under the LOSC, or the WTO panels, for instance, are mainly applying and interpreting the instruments that have conferred jurisdiction upon them. When they venture outside the confines of those instruments, they do so in order to interpret conventional rules by reference to other relevant instruments, under Article 31(3)(c) VCLT.Footnote 33 Despite their best efforts, these tribunals have both a built-in bias (since it is their job to apply the convention in question and not to apply IEL) and a constraint (since their jurisdiction is limited by the convention in question).
The number of IEL treaties, the complementing soft law instruments such as the Stockholm and Rio Declarations as well as the work of the ILC appear to cover most aspects of IEL regulation. It is probably, however, the lack of a general IEL treaty that lends IEL customary law its special significance. We are presented here with some sort of a paradox. The proliferation and sheer volume of IEL treaties apparently marginalize customary IEL, yet, at the same time, the lack of a general IEL treaty (let alone the lack of any specialized international court)Footnote 34 brings customary IEL squarely back into the picture. Therefore, if the question is ‘how many customary international law rules exist that regulate international environmental law’, the answer would be ‘very few’. But if the question is tweaked and becomes ‘how important are customary rules in international environmental law’, the answer differs slightly.
4.3 Interpretation of Customary IEL
The starting point in this discussion is the issue of the applicable methods of interpretation of IEL. These methods should roughly correspond those enumerated in Articles 31–33 of the VCLT. Among these methods, some doubt remains as to the applicability of the ‘object and purpose’ of the rule, at least within the context of IEL.Footnote 35 It is submitted that it is extremely difficult, and possibly fruitless, to ascertain such an object and purpose to customary IEL. All the more so that most, if not all, IEL rules can be certainly read in multiple ways, pertaining to the protection of the environment within a given context: development, trade, the normal continuation of economic activity and the like. Therefore, most environmental rules can be seen as having contrasting objects and purposes.Footnote 36 This is the main reason why most environmental rules have by their inception an inherent constraint with regards to teleological interpretation as they yield to a number of considerations that have little to do with what could be assumed to be the object and purpose of an environmental rule, that is, simply the protection of the environment. The question that comes first is which are those rules with which international courts are faced. Bearing that in mind, two preliminary observations are in order.
First, most – if not all – IEL cases that have reached an international court or tribunal are one way or another connected to an international treaty. This may happen because the said court or tribunal is being set up by a treaty. The ITLOS, for instance, in the Advisory Opinion on the Deep Seabed analyzed the no-harm obligation in the context of the deep seabed as it appears in the LOSC.Footnote 37 It did not go about making far-reaching proclamations outside the scope of its jurisdiction. Similarly, the Arbitral Tribunal set up for the South China Sea dispute interpreted (in a very rigorous manner) the LOSC, at times looking also at other conventional instruments.Footnote 38 Yet, the fact remains that it was the LOSC that was the focus of the analysis. Second, customary IEL rules are, as often is the case with custom, rather vague and general in nature. This is not necessarily bad. On the contrary, it leaves plenty of room to the courts and tribunals to elaborate on the rule presenting it each time in its more up to date and fullest form.
In the cases where courts and tribunals have discussed customary IEL, the two customary obligations that feature prominently are the prohibition of transboundary environmental harm (or no-harm rule) and the obligation to conduct an environmental impact assessment (EIA). The no-harm rule is possibly the most basic obligation in IEL. The starting point for any discussion involving transboundary environmental harm is the content of the obligation of the state from whose jurisdiction or control the harm emanates.Footnote 39 The ILC managed to formulate the main aspects of the obligation in the Prevention Articles.Footnote 40
With regard to the obligation to conduct EIA, as I have argued elsewhere,Footnote 41 it is an obligation of dual nature. On the one hand, it is a constitutive element of the broader no-harm obligation and, on the other hand, it is a free standing obligation of result The no-harm rule is an obligation of conduct and can be seen as containing a number of elements and the obligation to conduct an EIA is one of them.Footnote 42 On the other hand, the obligation to conduct an EIA is also an obligation of result generating its own content.Footnote 43 This is one way of looking, through a unifying concept, at the two obligations: the substantive obligation (no-harm) being analyzed in a series of procedural obligations, one of them being the obligation to conduct an EIA. Both rules are general in nature, yet not open ended, and they have evolved through time. This is evident both from the extensive discussions in the ILC, as well as the conclusion of numerous treaties that include either one or both rules. What is more, when it comes to the obligation to conduct an EIA, there is substantial state practice, as well as at least one dedicated treaty to the subject, namely the Espoo Convention.Footnote 44 It is, therefore, evident that when the courts have to apply either rule they have a variety of tools before them so as to proceed with the identification and elaboration of their content. The following section will look at the way the courts have approached these rules.
4.3.1 No-harm Obligation
This section explores the stance of international courts towards the customary rule of no-harm in IEL. The approach followed is a chronological one because in this way it is easier to follow the movement of the interpretative efforts of the jurisprudence. Ironically enough, the first proclamation on non-conventional IEL is to be found in the Trail Smelter case between the US and Canada regarding air pollution where the Arbitral Tribunal famously stated that
no state has the right to use its territory or permit the use of its territory in such a manner as to cause injury by fumes in or to the territory of another or the properties or persons therein, when the case is of a serious consequence and the injury is established by clear and convincing evidence.Footnote 45
The irony lies in the methodology of the Tribunal because it did not make that proclamation after reviewing the previous case law of international courts since such case law did not exist. The Tribunal made the point that it could not even draw an analogy with water pollution since there was no international case law on that issue either.Footnote 46 Therefore, it was forced to come up with a different solution since the applicable law on the case was international law, relevant rules of the law of the US, and the Tribunal was also directed to ‘give consideration to the desire of the high contracting parties to reach a solution just to all parties concerned’.Footnote 47
The Tribunal came up with the first iteration of the no-harm rule after reviewing the jurisprudence of the US Supreme Court on transboundary intra-state pollution. Ιt used as a legal basis US law, and it declared that the same principles could be part of international law by analogy especially since there was no international rule contradicting them.Footnote 48 Moreover, the Tribunal, in the process of establishing the no-harm rule, also made explicit reference to the desire of the parties to reach a just solution. It appears that it took this to mean that it could go beyond existing international law offering an extra justification to its decision to draw an analogy between US and international law. Therefore, the Tribunal propounded the above dictum on the basis of the ‘principles of international law’ as well as the law of the United States. It is not clear how the Tribunal uses the term ‘principles’ in the text. It could be that it wanted to denote that there is a general principle of lawFootnote 49 or, more probably, it just used the term ‘principle’ interchangeably with the term ‘rule’. The language in the pertinent paragraph indicates that it probably meant to give the term a normative value. The Tribunal refers to the ‘right’ of States, and it also refers to ‘injury’. It, crucially, also includes the threshold of the application of the principle by stating that the consequences of the harm must be serious, and it also declares a standard of proof. While all this could be just the only wording the Tribunal had at its disposal, it most probably means that it viewed the principle as a legal rule. This becomes even clearer a few lines further down the text where the Tribunal holds that, on the basis of the dictum, it finds Canada responsible in international law.Footnote 50
The crucial question is whether the Tribunal simply established the existence of the rule, or it also determined its content. It is obvious that the Tribunal did both, at least to some degree and with a twist. This is not a case where the judges look at the two elements of customary law, state practice and opinio juris and then decide whether a rule has come to existence. The Tribunal decided that a rule existed without looking at the two elements (they were not there) or even mentioning its nature, leaving it to be inferred from the wording of the decision. Then, it went on to shape its contours. Therefore, it did interpret the rule. Based on its reading of the jurisprudence of the US Supreme Court, it did set the threshold of application of the rule (serious consequence) and it did set the evidentiary standard required for the rule to be applied.
This is a striking dictum in that what can be seen as interpretation of the no-harm rule also serves as evidence that the Tribunal indeed meant to establish an international customary rule that does not come from international law as such. Rule identification and content determination, or rather interpretation, are intertwined in the dictum and still there is a way to distinguish them. For both identification and interpretation, though, there is no methodological consistency or clarity in the decision. The identification of custom is unorthodox to say the least, and the interpretation also comes without any strong backing in practice, opinio juris, or other elements that could lead to evolutive or systemic interpretation. The Tribunal looked for an interpretative solution outside the confines of classic customary international law and therefore it worked its way through the analogy with US intrastate law. The award remains a testament to the efforts of the Tribunal to come up with a viable solution to a pressing problem while, at the same time, it brought the no-harm rule to life. And this is no small feat.
The general application of the rule was affirmed, albeit in a context outside IEL, by the ICJ in Corfu Channel.Footnote 51 It is submitted that it cannot be safely assumed that the ICJ had in mind the Trail Smelter judgment when formulating the dictum in Corfu Channel. In Corfu Channel, the ICJ held that every State has an ‘obligation not to allow knowingly its territory to be used for acts contrary to the rights of other States’.Footnote 52 This is a less nuanced repetition of Trail Smelter. There are two points of contact between Trail Smelter and Corfu Channel dicta. First, it was clear in the Trail Smelter dictum that the rule can be breached through omission; and in Corfu Channel, the ICJ applied exactly this aspect of the rule since the harm was indeed caused by omission. Second, and more importantly, the rule in Corfu Channel becomes one of general application and is not confined to cases of injury by fumes. These points show that the ICJ did not move towards interpretation but that it simply generalized the Trail Smelter no-harm iteration. It also, importantly, did not adopt the extra elements of the Trail Smelter dictum, namely the requirement of seriousness and the evidentiary standard. This is not to say that there was any kind of judicial dialogue at play. It is merely a reading of the two cases with the benefit of hindsight. At the same time, the ICJ did include a criterion of knowledge that was not there in Trail Smelter. It is hard to see how this is an interpretative move at least not in the sense of refining the Trail Smelter dictum. The ICJ simply stated clearly what was implicit in Trail Smelter thereby deducing an obligation of notification. In Trail Smelter, the facts of the case did not leave any doubt as to the element of knowledge; therefore, it was only natural that the Tribunal would not linger on the issue. The operation in Trail was known to both parties and what was left to be determined was responsibility and remedies. In Corfu Channel, the issue of knowledge was central to the facts and the ICJ gave more prominence to it.Footnote 53
What seemed like a promising discussion regarding the no-harm rule in the 1940s became a non-issue in the following decades. When the issue resurfaced before the ICJ almost fifty years later in Nuclear Weapons, IEL had already exploded.Footnote 54 Both the Stockholm and Rio Declarations were adopted, the ILC had been working on the topic of injurious consequences arising out of acts not prohibited by international law for over a decade and most major environmental conventions were in place.Footnote 55
Despite the developments in IEL, there was no actual environmental litigation before the ICJ or any other forum with general international law jurisdiction. It was only in 2010 in Pulp Mills that the ICJ had a chance to revisit the no-harm rule. In Pulp Mills, the ICJ took a surprising course that not only left the content of the rule opaque but also sparked some unneeded confusion. The dispute between Argentina and Uruguay concerned the possible environmental impact of the construction of pulp mills on the river Uruguay, as the river was a shared resource between the two States. The legal basis for the dispute was the Statute of the River Uruguay (CARU) that laid down the rules for the management of the river by the two parties before the Court. At the same time, however, the ICJ recognized that in order to interpret the obligations arising out of the Statute, it had to resort to general international law as well as subsequent agreements between the parties.Footnote 56 It was in this vein that the Court went on to discuss customary IEL.
The ICJ stated that ‘that the principle of prevention, as a customary rule, has its origins in the due diligence that is required of a State in its territory’.Footnote 57 It then went on to cite Corfu Channel and Nuclear Weapons in support of this statement and it held that ‘[a] State is thus obliged to use all the means at its disposal in order to avoid activities which take place in its territory, or in any area under its jurisdiction, causing significant damage to the environment of another State’.Footnote 58 To begin with, the wording of the no-harm obligation as it appears in Pulp Mills is different than the wording used in previous decisions. While the content remains the same, it is unnecessarily confusing. It is not clear why the Court referred to due diligence. Whether it did so because it considered due diligence a general obligation, rule, or principle is not clear. Due diligence speaks to the nature of an obligation denoting that it is an obligation of conduct and not of result.Footnote 59 It simply characterizes the obligation in question. In this instance, the no harm obligation is an obligation that requires due diligence on behalf of the State of origin of the activity. Second, the ICJ does not explain the relationship between the principle of prevention and due diligence. It could be that the Court really meant that there is a general obligation of due diligence, irrespective of the existence of a primary obligation like the no-harm rule in the context of international environmental law. It could be that the Court pointed to the dictum in Corfu Channel as establishing such a general principle. The fact that the Court cites Corfu Channel in the preceding sentence could point towards that direction. It is highly unlikely, however, that the Court would proceed to such an important statement without offering any support. Moreover, and just like in Nuclear Weapons, the ICJ did not make any use of the actual development and refinement of the customary obligation so as to interpret it in a new light. It could either decide that the rule has changed through state practice and opinio juris, it could interpret dynamically the rule and bring it up to modern standards, or it could look at the development of IEL as a whole and go for a systemic interpretation. The ICJ did not look at international practice, it did not take into account the work of the ILC, it did not look at relevant treaties or the work of their treaty bodies. It neither seemed to bother too much with the current understanding of environmental harm in science and technology.
Nonetheless, the ICJ did follow the developments in IEL in one significant respect by introducing the threshold of ‘significant harm’. The ILC in its Prevention Articles had already lowered the threshold from ‘serious consequence’ to ‘significant harm’ and it has not been seriously challenged before international courts. The ILC was clear that the threshold is that of ‘significant harm’.Footnote 60 The ILC based its decision on the Brundtland Commission’s Article 10, the Commission being a group of experts under the UN that produced the seminal ‘Our Common Future’ report.Footnote 61 The ILC went on to further define ‘significant’ as meaning ‘something more than detectable but need not be at the level of “serious” or “substantial”’.Footnote 62 The ILC makes a clear choice between the two thresholds. The Commentary to Article 2 acknowledges that the threshold of ‘serious’ is used both in Trail Smelter and in Lake LanouxFootnote 63 arbitrations.Footnote 64 Both judgments were handed down decades before the ILC adopted the Articles on Prevention; and in between, IEL had developed in an explosive manner covering almost, if not all, areas of human activity and its connection to the environment. The ILC follows the development of the law towards a more progressive path.
The Court accepted the progress signalled by the ILC regarding the threshold of environmental harm, but it proceeded to introduce fresh confusion in the content of the no-harm rule. The ICJ cryptically stated that there is a ‘functional link’ between the substantive obligations of states (that presumably include the no-harm rule) and the procedural obligations of states (that include notification, EIA, the obligation to monitor the relevant activity, to consult and to cooperate, etc.).Footnote 65 The Court came to this conclusion interpreting the Statute of the River Uruguay and not a customary law obligation. Nonetheless, it is submitted that it is useful to look at the interpretation of the Statute and the distinction between substantive and procedural obligations therein as significant because the same distinction is reflected in the customary obligation of no-harm. There is the general substantive no-harm rule that prohibits States from using their territory so as to cause harm to another State.Footnote 66 And then, there are procedural obligations that provide guidance to States as to the steps they must take to avoid causing such harm.Footnote 67
The Court’s dictum on the functional link between these two types of obligations is apparently promising in that it looks like it interprets the no-harm rule in a nuanced manner. However, it probably means, according to the judgment, that if the substantive obligation is not breached, a breach of a procedural obligation (in Pulp Mills, the obligation to notify)Footnote 68 is not important from the point of view of reparation. Indeed, the Court declared that the recognition of the breach in the judgment itself constitutes satisfaction. Yet the judgment does not explain in detail what this functional link is – a fact not lost on judges Al Khasawneh and Simma who criticized this lack of any explanation in their joint dissent.Footnote 69 The dissent also criticizes the view of the Court on the procedural obligations as unsatisfactory because it renders the procedural obligations essentially unimportant.Footnote 70 It is only natural that the ‘functional link’ statement has not been followed up in the case law that followed Pulp Mills. On the contrary, Judge Donoghue directly rejected it in her separate opinion in Costa Rica/Nicaragua stating that even ‘[i]f at a subsequent phase, the failure of a State to exercise due diligence in the implementation of a project causes significant transboundary harm, the primary norm that is breached remains one of due diligence, but the reparations due to the affected State must also address the material damage caused to the affected state’ therefore it is not ‘[u]seful to draw distinctions between “procedural” and “substantive” obligations’.Footnote 71
Pulp Mills was indeed a rare opportunity both from the standpoint of the substance of the case as well as from the standpoint of the evolution of the methodology of the ICJ in identifying and elaborating upon customary international law. The Court followed a path wherein it simply repeated its previous jurisprudence and declined to explain, at the very least and even as obiter, what the functional link between procedure and substance is. This is even more frustrating if one takes into account the fact that Argentina claimed the breach of both substantive and procedural rules. When the ICJ went on to look at whether the mills had already caused harm to the river Uruguay it did so solely with reference to the Statute of the river as its legal basis.Footnote 72 It essentially dismissed all of Argentina’s claims based on lack of proof of harm. The outcome of the decision is not unrelated to the lack of elaboration of the no-harm principle. A better enunciated rule could have been more useful in properly assessing the impact of the mills.Footnote 73
The ICJ did not offer any more significant insights in its very brief mention of the no-harm rule in Costa Rica/Nicaragua where again it simply affirmed its existence.Footnote 74 It did not even look at the more elaborate definition of the rule and the insight to some of its content by the ITLOS in its Deep Seabed Advisory Opinion.Footnote 75 The ITLOS Special Chamber went on at some length in discussing the no-harm rule and it made a few interesting points along the way. First, it admitted that the content of due diligence is difficult to pin down.Footnote 76 Second, it gave an idea of the mechanics of the rule. The Chamber claimed that the ‘obligation becomes stricter or more lenient depending on the activity’.Footnote 77 Despite not breaking new ground, the Chamber did interpret the no-harm obligation as one that does not remain constant in its application, but it must be applied differently depending on the circumstances. This is, according to the Chamber, how the nature of the obligation – that is, due diligence – affects the concrete application of the obligation. The ICJ did not engage with the Advisory Opinion and, like it had done in Pulp Mills, ignored the other developments in IEL. The only notable, and important, difference between Pulp Mills and Costa Rica/Nicaragua is that the ICJ dropped, without explanation, its idea of a functional link between procedural and substantive obligations, and that was a positive step.
What emerges from this picture is that the ICJ did not feel the need to expand upon, analyze, or identify the content of one of the cornerstones of IEL. Why the court made that choice is a matter of speculation. Especially when there is a rich customary as well as treaty background available to the ICJ. The outcome therefore can be described as regressive interpretation of the rule. This concept comes from the realm of human rights. The European Court of Human Rights (ECtHR) has since its early days treated the European Convention on Human Rights as a ‘living instrument’, that is, it interprets the Convention by adapting its content to the current political and social conditions.Footnote 78 In the following years, the ECtHR has interpreted the Convention following the ‘living instrument’ doctrine, but the interpretation has not been unidirectional towards the expansion of human rights. Research has convincingly shown that the ECtHR has also adopted a regressive interpretative approach overturning previous ‘expansionist’ decisions.Footnote 79 Methodologically speaking, there are significant differences between the two fields when it comes to studying the interpretation of the respective rules. The ECtHR has an impressive number of judgments that offer ample samples for qualitative and quantitative research.Footnote 80 The cases that involve IEL before the ICJ, or arbitral tribunals for that matter, are few and far between. Moreover, the ECtHR has the benefit of interpreting a Convention instead of customary international law; therefore, it has a more secure footing from the start. The ICJ, on the contrary, often identifies the customary IEL rule and interprets it at the same time.Footnote 81 The concept of regressive interpretation is used for what it actually is: a concept. It is not intended to serve as an analogy to human rights law or the human rights jurisprudence of the ECtHR. With few decisions and fewer dissenting or separate opinions, the concept is tested, for the most part, against settled doctrine.
With regressive interpretation in mind, the prehistory of IEL interpretation, traced back to Trail Smelter, offers some insights. The Arbitral Tribunal in Trail Smelter took very bold steps: it had to come up with analogies and essentially ‘invent’ a rule of international law that did not exist at the time. While it took time for IEL to develop and grapple with the difficulties of the no-harm rule, eventually the development did come about through the work of the ILC, through conventions that reflect the customary rule, as well as through elaborate doctrinal work. And then the ICJ started moving almost backwards in what can be seen as regressive interpretation. It did not interpret the rule in a way that captured the changes that had occurred in international law. It did not touch upon the threshold of the obligation; it did not offer insights regarding the nature of the rule (the declaration on the due diligence nature aside); and it certainly did not add anything to the rule. On the contrary, the ICJ added unnecessary confusion with the employment of the ‘functional link’ concept and the decoupling of the procedural obligations from the overall obligation to not cause transboundary harm. Overall, it seems that it regressed to something that falls behind the Trail Smelter dictum – let alone further developments in IEL – in terms of boldness and inventiveness.
4.3.2 The Obligation to Conduct an Environmental Impact Assessment
The situation regarding the treatment of the customary obligation to conduct an EIA is even more complex than that surrounding the no-harm rule. This section shows that the case law slowly, but surely, adopted an evolutive interpretation on two levels: first, on the level of the relevant legal framework (national/international) within which it must be conducted and, second, regarding the content of the obligation and the extent to which it can be accurately sketched by international courts.
The obligation to conduct an EIA features prominently in several IEL treaties, and it has become a rule of customary law.Footnote 82 ΕΙΑ is the platform upon which states usually concentrate their disputes before international courts and tribunals.Footnote 83 Almost every environmental dispute that finds its way before a court has at least one of its aspects dedicated to EIA. What is more, the ITLOS has specifically instructed parties to disputes to cooperate in order to assess the risk of the activityFootnote 84 or to appoint experts to study the environmental impact of the activity.Footnote 85
The beginning of the courts’ involvement with the obligation to conduct an EIA was not promising. In Gabčikovo/Nagymaros, the ICJ had an opportunity to discuss EIA in detail. Despite the fact that the Court shied away from this discussion, it provided some insight as to how international environmental rules should be interpreted. The ICJ suggested that obligations geared towards monitoring transboundary risk are amenable to evolutive interpretation and that they cannot be treated as static norms.Footnote 86 This finding is important enough in and of itself and without any connection to the EIA or the no-harm rule. Yet, beside this dictum (better classified as obiter), the ICJ did not go any further. This attitude, even at the time of the Gabčikovo/Nagymaros judgment, is peculiar. It is not as if the parties to the dispute had not stressed the EIA, quite the contrary.Footnote 87 While the Espoo Convention lays down detailed conditions for the conduct of an EIA, the numerous treaties that contain the obligation do not go into such a detail. At the same time, the ILC was not clear at all in its commentary to the Prevention Articles.Footnote 88 It follows that the snapshot the courts attempt to take of the obligation to conduct and EIA is of great importance as it informs the content of a ubiquitous obligation that seeks clarification.
The ICJ in Pulp Mills affirmed the customary nature of the obligation to conduct EIA coupled with an ongoing obligation to monitor the operation of the activity. After it set out the ground rule of customary law, the ICJ went on to discuss the obligation to conduct an EIA. The ICJ states that the CARU (the Statute of the River Uruguay) does not provide any content to the obligation.Footnote 89 It follows that up by stating that the scope and content of the obligation is not defined in general international law either.Footnote 90 This grossly inaccurate statement is the only way the ICJ can reach its conclusion 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’.Footnote 91 To be fair, this is the only time that Pulp Mills seems to follow the ILC in its assessment that each State determines the process and content of the EIA. But this is not entirely true. Indeed, the Commentary to Article 7 of the Prevention Articles of the ILC states that the specifics of what ought to be the content of the EIA are to be decided by the State that conducts the assessment.Footnote 92 This statement is however qualified further down in the Commentary. The ILC clarifies that the assessment must include the effects of the assessment. It goes on to add that the content of the assessment is dependent by the type of energy used for the activity, the activity’s size, as well as its location.Footnote 93
Therefore, the meaning of the Pulp Mills dictum on EIA merits closer attention. If the ICJ only meant that it is through domestic legislation that each State gives effect to its international customary obligation to conduct an EIA, then there is nothing further to discuss. This is true, yet it is also self-evident. If the ICJ meant just that, then it simply did not add anything new to the discussion. It is also true, however, that if national legislation is to indeed give effect to the international obligation, it cannot merely dictate the conduct of an EIA without any kind of prerequisites. If this was the reasoning of the ICJ behind this proclamation, then this is obviously unsatisfactory for two main reasons. First, the ICJ does not bother to look either to the ILC Articles or at the numerous international instruments that provide for an EIA and at the same time provide for a rudimentary framework. It is only natural that the technical details should be determined in the domestic legislation, but the overarching framework and broad directions could, relatively easily, be inferred from the ILC Articles, IEL conventions and practice. Second, the conduct of the EIA itself is highly complex. It can easily lead to disagreements between the State of origin and the potentially affected State meaning that a deadlock in the discussions is not unprecedented.Footnote 94 A clear framework could at least anticipate and address some of the most obvious obstacles in the process.
In Nicaragua/Costa Rica, the Court changed course in a spectacular way. It interpreted the obligation of an EIA in a rather open and progressive way, breaking away from Pulp Mills. It held that several factors are relevant in examining the proper adherence of the State to the obligation to conduct and EIA. The ICJ included the magnitude of the project, its context, the scale, location, as well as the geographical conditions.Footnote 95 This is clearly an interpretation of the obligation that incorporates elements found in environmental agreements, albeit without any such acknowledgement whatsoever. The Espoo Convention, being the most comprehensive international instrument regulating the conduct of EIA, offers significant guidance. Appendix II of the Convention sets out a number of criteria that determine the conduct of an EIA. The Appendix includes the size and the location of the activity as well as the nature and kind of damage that may occur. The ILC Commentary also adheres to these criteria.Footnote 96 They are very similar, practically identical to those propounded by the ICJ. In other words, the Court caught up with the evolution of international environmental law in order to define the scope and content of the obligation. Whether the ICJ moved in that direction because it looked into relevant environmental law instruments, state practice, or other developments is not possible to know.
It is the lack of elaboration on behalf of the ICJ that leaves behind a question mark as to whether this can be classified as evolutive interpretation, or it is merely the application of the obligation to the facts of the case. On the one hand, the ICJ in Nicaragua/Costa Rica quoted (not with disapproval) the Pulp Mills dictum that the State determines the content of the EIA obligation in its domestic legislation. At the same time, the ICJ also looked at the fact that Costa Rica’s domestic legislation provided for an emergency provision exempting the government from the obligation to conduct an EIA.Footnote 97 It determined that the emergency provisions could not trump the obligation of Costa Rica to carry out an EIA under international law.Footnote 98 This might be a hint of differentiation from Pulp Mills. Be that as it may, had the ICJ merely applied the Pulp Mills dictum as a definitive statement on the status as well as on the content of the obligation, it would have deferred any criteria on the content of the EIA to the relevant domestic legislation in question. The ICJ did something different. It interpreted the bare obligation as found in Pulp Mills. It gave flesh to its bones by interpreting it as containing specific criteria that the State of origin has to take into account when it transposes the international obligation to its domestic legal order. This, of course, is only one of the reasonable readings of the judgment, yet the most convincing. State practice, international agreements and the work of the ILC were all there when the ICJ decided Pulp Mills. It may be true that, rather than evolution, it seems more like catching up a few years later; the judgment is an example of expansive interpretation at least compared to the Court’s past jurisprudence.
4.4 Conclusions
The first instance in international jurisprudence of identification and concurrent interpretation of an EIL norm was Trail Smelter. The Tribunal came up with the first definition and, concurrently, interpretation of the no-harm rule, and it definitely set the groundwork for what followed. It was then the turn of the ICJ to try its hand at the interpretation of IEL norms. The ICJ has taken a conservative approach in identifying the rules of customary IEL. Not only that, but it seems that the ICJ has been out of step with the state of development of IEL at the time it decided Pulp Mills, resorting to regressive interpretation. The Court failed to explain the content of the no-harm obligation and also declined to explain what the ‘functional link’ between substantive and procedural obligations is. Moreover, it seemed to hamper the effective application of the no-harm rule by disconnecting the breach of the obligation to notify from the substantive component of the obligation to not cause transboundary environmental harm. In Nicaragua/Costa Rica, the ICJ did not manage to shed further light on the content of the rule.
Regarding the obligation to conduct an EIA, the ICJ changed course from one case to the next. In Pulp Mills, the Court resorted again to regressive interpretation, falling behind the position of EIL on EIA at the time. The ICJ in Costa Rica v Nicaragua/Nicaragua v Costa Rica showed an imbalance in defining the threshold of the triggering of the obligation to conduct an EIA between the two proceedings. In Costa Rica/Nicaragua, the ICJ did not elaborate much on EIA. Nonetheless in Nicaragua/Costa Rica the Court did in fact develop the content of the obligation through evolutive interpretation. It caught up with, at least, the minimum content of the obligation as it appears in international texts and as it probably stands in state practice.
Overall, international courts and tribunals have interpreted IEL in expansive as well as restrictive ways. Sometimes they have even oscillated between the two. Starting from an expansive interpretation, as was the case in Trail Smelter, moving to regression in Pulp Mills and then correcting the course in Nicaragua/Costa Rica. Even though the lineage does not lie within the ambit of the jurisdiction of a single court, it is submitted that it is telling of the difficulties with which courts must grapple. Today, that IEL is not a novelty anymore, international courts will have the chance, and probably take it, to develop the content of customary IEL more freely and with a steadier hand.
Open access