The Precautionary Principle and a Fair Allocation of the Burden of Proof in International Environmental Law

Authors


Abstract

Notwithstanding the general acknowledgment of its importance, there are still uncertainties and ambiguities as to the precise meaning and actual application of the precautionary principle in international environmental law. In the context of decision making at the inter-State level, it has been explained that one of the interpretations of this principle means a ‘shift’ in the allocation of the burden of proof. Unsurprisingly, a similar effect of the principle can be, and has been, claimed in dispute settlement. Given these claims, the general aim of this article is twofold: first, to define what a ‘shift’ in the allocation of the burden of proof actually means; and second, to explore and assess whether this theoretical/conceptual understanding of the ‘shift’ is applied when the precautionary principle is at stake, and whether there are good reasons to apply it in dispute settlement and decision-making processes.

INTRODUCTION

The principle1 of precaution (or precautionary principle, PP) started its journey in international environmental law2 twenty years ago in Rio de Janeiro, when it received a very prominent place by the inclusion of a definition in the Rio Declaration on Environment and Development.3 In 1992, the international community committed itself to the idea that scientific uncertainty should not be used to postpone action necessary for the protection of the environment – that is, ‘decisions should err on the side of environmental preservation’4 (in dubio pro natura). However, twenty years later we still cannot convincingly talk about the ‘full application’ of the principle. One of the probable reasons for this is that, notwithstanding the general acknowledgement of its importance, there are still uncertainties and ambiguities as to its precise meaning and actual application.5 Remarkably, the 2012 Rio+20 summit did not contribute to a better understanding of the principle, with the Conference paying only very limited attention to it.6

In the context of environmental decision making7 at the inter-State level, it has been explained that one of the interpretations of the PP implies a shift in the allocation of the burden of proof (BoP): a proponent of a certain action should bear the onus of proving that the activity will not be harmful. Unsurprisingly, a similar effect of the PP can be, and has been, claimed in dispute settlement. State parties to a dispute have also referred to the applicability of this shift in (quasi-)judicial proceedings at the inter-State level; however, they have so far been unsuccessful in this regard. Indeed, collecting relevant scientific evidence as to the existence of (a certain level of) risk is neither a simple nor a cheap exercise, which implies that the allocation of the BoP can have far-reaching consequences for the outcome of a decision-making process, for dispute settlement, and not the least for the environment. So what does this BoP-shifting mean for the PP?

Against this backdrop, the general aim of this article is twofold: first, to define what a ‘shift’ in the allocation of the BoP actually means; and second, to explore and assess whether this theoretical/conceptual understanding of the ‘shift’ is applied when the PP is at stake, and whether there are good reasons to apply it in dispute settlement and decision-making processes. This second purpose includes two related issues. First, in light of the theoretical conceptualizations of the BoP rules, the article analyzes PP-related cases decided by international (quasi-)judicial bodies, and illustrates the importance of BoP arrangements for a successful reliance on the PP. Second, the findings in dispute settlement are adjusted and applied to decision making with regard to which the claim to shift the BoP is generally made. This discussion about the allocation of the BoP does not, however, address the standards of proof that might vary in accordance with the different substantive understandings of the PP.8

The article proceeds on the basis of Franck's understanding of ‘normative fairness’ or ‘distributive justice’,9 according to which, normative rules such as the PP can be regarded as fair when they ensure ‘distributive justice’– that is, when ‘an appropriate or proper allocation of burdens and benefits [is] secured by the rules themselves’.10 Based on this concept, the need to adjust or change a normative rule or its procedural consequences is, accordingly, created by a sense of injustice in the allocation of these burdens and benefits given the purpose of the rule and its actual application. In other words, the reasons for either applying the general rules on the allocation of the BoP or shifting it will be based on fairness11 given the underlying purpose of the PP – that is, the protection of the environment.12 Indeed, harm to the environment is difficult, if not impossible, to cure or reverse, which justifies the prominent place of the principle in this field of law. Fairness in relation to the allocation of the BoP arguably concentrates on the main interest to be protected by the substantive rule at stake – in the present case, the environment – taken in combination with the procedural position of the parties. So, fairness in the sense of defining the procedural consequences does not consider the different interests of the parties, given that such an exercise would result in the substantive balancing of interests, which is related to the proportionality principle, and not to the procedural consequences of the PP.

The article first provides a brief explanation of the PP and the two possible forms thereof that can be invoked in dispute settlement proceedings. These forms are illustrated by the Pulp Mills13 and the Gabčikovo-Nagymaros14 cases of the International Court of Justice (ICJ) as well as the EC-Hormones case15 of the dispute settlement bodies of the World Trade Organization (WTO). Although the latter case does not relate to the field of the environment, the nature and importance of the protected interest (health) can be compared to that of the environment and thus the arguments of the dispute settlement body can provide useful insights.16 Following this explanation, the next section addresses the theoretical conceptualizations of the general and special rules on the allocation of the BoP. As the special allocation of the BoP, which is often described as ‘shifting the burden of proof’, has been conceptually devised in discrimination cases, the discourse relating to this field of law provides significant insights for the present study. These conceptual clarifications are followed by an analysis of the PP in terms of the allocation of the BoP illustrated by the above-mentioned cases. The two forms of the PP are matched with the rules on the allocation of the BoP based on fairness-driven arguments. The article then discusses how the conceptual considerations as elaborated with regard to dispute settlement can be adapted and applied to decision making. The article concludes with some final observations on the PP and the allocation of the BoP.

THE PRECAUTIONARY PRINCIPLE: A CONCISE OVERVIEW

DIFFERENT INTERPRETATIONS OF THE PRECAUTIONARY PRINCIPLE

The concept of precaution is part of our everyday lives and has gained a prominent place in international environmental law as a legal principle.17 The PP has been codified in more than fifty treaties in international environmental law.18 Nevertheless, its status as being part of customary international law has not been unequivocally established.19 There is agreement that its purpose is to protect the environment by providing a strategy for addressing (scientific) uncertainty or uncertain risk: the in dubio pro natura principle.20 A practical consequence of the principle is the need for risk assessment and/or environmental impact assessment (EIA).21 Arguably, not only does the protection of the environment require that an initial risk assessment is carried out when a measure is introduced, but it also implies that the potential risk of a measure to the environment is periodically reassessed. While this position is convincingly defendable from an environmental point of view, such a dynamic interpretation of the principle has not been accepted yet.

Although there is an agreement on the main purpose of the principle, it has been interpreted in several ways.22 One of the accepted legal definitions can be found in the Rio Declaration on Environment and Development. Principle 15 of the Rio Declaration provides that:

In order to protect the environment, the precautionary approach shall be widely applied by States according to their capabilities. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation.23

Among the different interpretations, Wiener24 identified three ‘archetypal versions’ of the PP: ‘uncertainty does not justify inaction’; ‘uncertainty justifies action’; and ‘shifting the burden of proof’.25 Whereas the first means that precautionary regulation can be adopted when the risk is uncertain, the second implies that precautionary measures have to be taken even in the absence of full certainty concerning the risks.26 The third version means that an activity that might cause harm to the environment cannot be allowed until the lack of (a certain level of) risk has been proven by the advocate of such activity.27

Without having to take a position as to which approach should be followed, it can be argued that there is a significant difference between the first two and the third one. While the former vers have been identified based on a substantive conception of the PP, the latter refers to a procedural effect or understanding of the principle. Therefore, this article hypothesizes that due to the procedural nature of the BoP rules, the BoP-driven PP can also be seen as a procedural concept. Put differently, it is argued that the procedural understanding is not an alternative, but rather a complementary version of the PP.28 The present study focuses on this procedural understanding (or rather: effect) of the PP – that is, whether or not it has and/or should have an effect on the allocation of the BoP.

APPLICATION OF THE PRECAUTIONARY PRINCIPLE IN DISPUTE SETTLEMENT

The PP has been said to have effect on the allocation of the BoP not only in decision making, but also in dispute settlement. When it comes to the latter, the procedural position of the parties is regulated by relatively clear rules. Given that the allocation of the BoP is closely related to the procedural position of the parties,29 this position of the parties will be used to explore the manner in which the PP was invoked by them. The parties can be either applicants alleging the violation of an obligation, or respondents trying to come up with justification (depending, of course, on whether or not such an exception is enshrined in the applicable regulation).

In the case-law of the ICJ and the WTO panels and Appellate Body (AB), the PP has been invoked in two ways: as an obligation, and as a justification. In the Pulp Mills case,30 for instance, the applicant State (Argentina) argued that Uruguay had not acted in accordance with the PP when it had started the operation of the mills on the Uruguay River. Here, the PP takes the form of an obligation,31 and Uruguay's non-compliance with this was invoked by Argentina:

According to Argentina, Uruguay ha[d] breached its obligation to contribute to the ‘optimum and rational utilization of the river’ by failing to co-ordinate with Argentina on measures necessary to avoid ecological change, and by failing to take the measures necessary to prevent pollution.32

In case-law, PP can also take, and has indeed taken, the form of justification33– good illustrations of which are the Gabčikovo-Nagymaros34 and the EC-Hormones cases.35 The latter was initiated by the United States against the European Communities (EC), in which it was alleged that the EC violated WTO law by restricting or prohibiting imports of meat and meat products from the United States. In this case: ‘The European Communities … invokes the precautionary principle in support of its claim that its measures in dispute are based on a risk assessment.’36 In other words, the EC tried to justify the restrictions on trade by relying on the PP. Similarly, in the Gabčikovo-Nagymaros case, Hungary referred to the PP in order to justify its non-compliance with the obligations flowing from the treaty between Hungary and the Czechoslovak Republic.37

Given that these forms of the PP (as obligation or justification) are closely related to the procedural position of the parties in the dispute settlement, they will be addressed separately in the exploration of the applicable and desirable BoP rules.

RULES ON THE ALLOCATION OF THE BURDEN OF PROOF IN DISPUTE SETTLEMENT

In order to assess the effect of the PP on the allocation of the BoP, first the general rules on BoP allocation will be conceptualized and explained. An important development in evidence law is that in certain instances some factors have led to deviation from these general rules, and special rules have been adopted and applied regarding the allocation of the BoP (often also called ‘a shift in the burden of proof’). To the best knowledge of the author, discrimination law is the only field of law so far where these special rules have found their way into written legislation. Based on the conceptual elements of the rules relating to the allocation of the BoP, the second part of this section will provide a conceptualization of these special rules, and explain the factors that have influenced their adoption into discrimination law. Although implicit, the driving force behind the creation of these special rules is, arguably, the inherent need to achieve fairness. Hence, these factors are also relevant for determining fairness.

GENERAL ALLOCATION OF THE BURDEN OF PROOF

In general evidence law, two forms of BoP can be identified: an evidential or subjective one, also called the ‘burden of production’; and a legal or objective one, also called the ‘burden of persuasion’.38 While the former refers to the ‘burden of going forward with the evidence by producing evidence’,39 the latter means the burden of persuading the fact-finder that the allegations made are true.40 The burden of persuasion is by definition allocated for a specific issue and remains on the same party during the assessment of the matter at stake. On the contrary, the burden of production moves back and forth between the parties because the facts underlying the allegations are generally not established on the basis of the evidence coming only from one party, but they are also refuted, disproved or impaired by the other.41 The importance of this distinction lies in the fact that the party having the burden of persuasion has to make his case. In other words, he bears the risk of not proving the alleged ‘violation’, while the other party only has to make these allegations weaker with the help of his burden of production.

In light of these definitions, the term ‘allocation of the burden of proof’ has two separate meanings. On the one hand, it addresses the question of which party has to face the consequences of the fact-finder's conclusion that the allegations remained unproven.42 In this sense, the BoP pertains to the burden of persuasion (also called the ‘legal burden’). On the other hand, the allocation of the BoP might also describe which party can or must produce evidence at a particular moment during the process. This meaning points to the burden of production. Admittedly, there is no need for the explicit regulation of the burden of production43 since it shifts back and forth between the parties regarding the matter at stake. Accordingly, the ‘allocation of the burden of proof’ is hereafter meant to describe who bears the burden of persuasion in a concrete case with regard to the specific matter.

Pursuant to the commonly known general rule of evidence, onus probandi incumbit actori:44 the party alleging something is obliged to prove it.45 On the basis of the general allocation of the BoP, accordingly, the burden of persuasion rests with the party initiating a procedure (the applicant)46 against the other party (the respondent) concerning a specific issue.47 This also means that it is also the applicant who bears the initial burden of production. When the applicant has complied with his burden of production, this burden shifts to the other party, who has the possibility to submit evidence to disprove the facts provided by the applicant.48 However, the burden of persuasion stays with the applicant because it is him who bears the risk of losing his case. The respondent will only try to weaken the applicant's statements and evidence by challenging them through his burden of production.

SPECIAL ALLOCATION OF THE BURDEN OF PROOF49

In certain instances the BoP is, or is argued to be, allocated in a manner that is different from its general distribution. Similar to the procedural effect of the PP, this special allocation is also often referred to as a ‘shift’ in or ‘shifting’ the BoP. Although there are several interpretations of this ‘shift’, based on evidence law it can only be realized in one specific manner.

In discrimination law, special rules on the BoP have been codified at the level of the European Union. Several Directives have been adopted in which these special rules are laid down, and which have subsequently been implemented in the Member States. Pursuant to this rule,

when persons who consider themselves wronged because the principle of equal treatment has not been applied to them establish, before a court or other competent authority, facts from which it may be presumed that there has been direct or indirect discrimination, it shall be for the respondent to prove that there has been no breach of the principle of equal treatment.50

This rule, however, does not explicitly address the question to what extent this allocation of the BoP is different from the general rules. As explained above, in order to establish how the BoP is distributed exactly, and thus to what extent or in which respect its allocation is special, the party who bears the burden of persuasion has to be identified.51 Arguably, the adoption of these special rules would be pointless if the burden of persuasion was not assigned to the respondent. Accordingly, the special allocation of the BoP means that it is not the applicant who bears the burden of persuasion, but instead this burden rests with the person who has to disprove the alleged occurrence of discrimination. Strikingly (and thus somewhat confusingly), unlike in procedures with the general allocation of the BoP, at the beginning of the discrimination procedure the burdens of production and persuasion do not rest with the same person. The applicant bears the initial burden of production without having the burden of persuasion. Nevertheless, his initial burden of production establishes the respondent's burden of persuasion.

There are at least three arguments supporting this interpretation of the special allocation of the BoP. First, from a formal point of view, the burden of persuasion does not shift by definition from one party to the other. It is established at the beginning of the procedure and stays with the same party concerning the particular issue at stake. Second, as demonstrated above, if it were only the burden of production which shifted from the applicant to the respondent and the burden of persuasion stayed with the applicant it would not suggest a shift in the BoP since the burden of production always moves back and forth between the parties.52 As Bindman posits: ‘The burden on the respondent needs to be more than an evidential burden if there is to be a real inducement to avoid discrimination.’53 Third, the Explanatory Memorandum of the Racial Equality Directive also stipulates that the legal BoP should be placed on the respondent.54 These arguments convincingly show that the special allocation of the BoP means that the burden of persuasion rests with the respondent as soon as a presumption (prima facie case) of discrimination is established.

The Danfoss case55 in 1989 was the first judgment in which the ECJ underscored the importance of the special allocation of the BoP in discrimination cases. It declared that in case of

a system of pay which is totally lacking in transparency, it is for the employer to prove that his practice in the matter of wages is not discriminatory, if a female worker establishes, in relation to a relatively large number of employees, that the average pay for women is less than that for men.56

A similar reasoning was given in the Enderby case57 a couple of years later. This view of the Court led to the codification of this rule in the Burden of Proof Directive in 1997.58 Its provision on the BoP has been repeated in the Racial Equality Directive and the Framework Equality Directive,59 as well as in the later Directives on equal treatment.60

The above-mentioned cases reveal that the general allocation of the BoP was considered to cause difficulties for the applicant in discrimination cases,61

particularly in the employment sector where an unequal power relationship between employer and employee exists. The employer tends to be in a considerably stronger position than the employee, principally because much (if not all) of the information about how the allegedly discriminatory decision was arrived at is held by the employer.62

If the burden of persuasion were not on the respondent, the applicant would not be able to make out his case since the evidence is generally in the hands of the respondent.63 In the words of the ECJ, workers

would be deprived of any effective means of enforcing the principle of equal pay before the national courts if the effect of adducing such evidence was not to impose upon the employer the burden of proving that his practice in the matter of wages is not in fact discriminatory.64

At the same time, this means that the obligation to apply the special allocation of the BoP in discrimination cases can be deduced from the general principle of effective enforcement of the prohibition of discrimination.65 According to Kokott, ‘an intimate relationship exists between the effective protection of human rights, on the one hand, and adequate material standards as well as the appropriate allocation of the burden of proof, on the other’.66 This relationship has also been made clear both in the preamble of the Burden of Proof Directive67 and in the preamble of the Racial Equality Directive.68

In short, the arguments for relying on the special allocation of the BoP, which arguably originate from the idea of ‘fairness’, can be summarized as follows: first, lack of transparency in the adoption of the measure at stake; second, unequal power relationships; and third, effective enforcement of the principle aimed to be protected (principle of non-discrimination in casu). In other words, these grounds seem sufficient to justify the application of evidential rules that are more favourable for the person who has allegedly become the victim of a prohibited measure.

It might be clear that the term ‘shifting the burden of proof’ does not properly describe the phenomenon that actually takes place in the course of providing evidence on the basis of the special rules concerning the allocation of the BoP. In actual fact, there is no ‘shift’ in the BoP between the parties. The burden of persuasion does not shift from one party to the other during the procedure since it is placed on the party charged with having committed discrimination at the beginning of the procedure regarding the specific issue. Accordingly, the term ‘special allocation of the burden of proof’ would more accurately denote the essence of these evidentiary rules in that it makes it clear that these rules differ from the general rules on the distribution of the BoP. As argued above, this difference can manifest itself only in one specific manner – namely placing the burden of persuasion on the other party.

THE PRECAUTIONARY PRINCIPLE AND THE ALLOCATION OF THE BURDEN OF PROOF IN DISPUTE SETTLEMENT

The claim has been made that (one interpretation of) the PP consists of shifting the BoP. This section examines this claim with respect to dispute settlement. The conceptualization on the evidentiary rules outlined above serve as a guide for examining whether and in what way the BoP is allocated when precaution is referred to as obligation and justification, respectively. In addition, it examines whether fairness-based arguments, as indicated above, justify the application of the special rules on the allocation of the BoP.

PRECAUTION AS OBLIGATION AND THE ALLOCATION OF THE BURDEN OF PROOF: THE PULP MILLS CASE

In the Pulp Mills case, the Court established that the State was ‘obliged to use all the means at its disposal in order to avoid activities … causing significant damage to the environment of another State’.69 Here, this meant that Uruguay was under the obligation not to take any action that could cause (significant) damage, among others, to the river. There are, arguably, two factors that essentially influenced the outcome of the case, which, in turn, was interpreted as the ICJ's rejection of the applicability of the PP. First, in accordance with the general allocation of the BoP, and as was also argued by Uruguay and accepted by the Court, the applicant had to bear the burden of proving that the respondent did not comply with this obligation – that is, that the respondent avoided activities ‘causing significant damage to the environment’. Argentina tried to make the argument that the BoP ‘should not be placed on Argentina alone’ claiming that ‘the 1975 Statute imposes an equal onus to persuade’.70 The Court, however, relied on the general rules concerning the allocation of the BoP when it explained that ‘in accordance with the well-established principle of onus probandi incumbit actori, it is the duty of the party which asserts certain facts to establish the existence of such facts’.71 So the burden of proving that Uruguay did not comply with the PP as obligation had to be borne by Argentina. Hence, the fact that a State did not take measures amounting to careful considerations had to be proven by the State challenging this.

Second and closely related to the previous point, the ICJ seemed to have accepted that the absence of precautionary measures could be proven by the occurrence of harm to the environment – namely, the Uruguay River. The possibility of proving the lack of precaution based on the existence of harm is significant from at least two perspectives. On the one hand, this approach is somewhat inconsistent with the logic of the PP that focuses on preventing any harm from occurring – that is, being ‘forward-looking’ in nature rather than ‘backward-looking’.72 A backward-looking approach implies that a violation of the PP can only be established if harm is caused/has occurred, which might be problematic in terms of proving a violation of the PP in cases where there is no actual harm yet even though it seems that the PP has not been complied with. On the other hand, proving the existence of harm was probably the only possibility available for the applicant to prove his case; otherwise Argentina would not have been able to submit any related evidence.

Thus, the application of the general rules on the allocation of the BoP made the Court interpret the PP as a result-oriented obligation. This interpretation required the applicant to carry out costly scientific assessments of the Uruguay River, which the respondent was initially obliged to do. Hence, it can be convincingly argued that the applicability of the PP (as obligation) was in actual fact accepted by the ICJ.73 However, due to the application of the general evidentiary rules it became difficult for the applicant to prove the respondent's non-compliance with this obligation, and the principle became a result-oriented obligation.

The question can then be raised as to whether fairness-based arguments that aim to ensure the protection of the environment could or would justify a different allocation of the BoP? Precaution as obligation, the violation of which was alleged by the applicant, can arguably be compared with the obligation not to discriminate as far as the reasons for ‘shifting’ the BoP are concerned.74 First, the manner in which the debated measure was adopted is unclear – what kind of research was carried out, what were their results and so on? (lack of transparency). Second, the party under obligation is generally in possession of the necessary information,75 and given the sovereign nature of the parties, the applicant could not have access to any such information, which, in turn, created unequal procedural positions. Similarly, in order to prove his case the applicant had to carry out expensive and time-consuming impact or other assessments, which was or in any event would have been the task of the respondent. Third, and in light of these two previous reasons, effective enforcement of the PP and the protection of the environment would seem to require that the party under the obligation to observe this principle bore the burden of proving that the measures were not harmful. Should the applicant bear the BoP, the respondent can easily escape from responsibility for not observing this principle, which would, in turn, encourage non-compliance with the PP. Finally, the principle of in dubio pro natura would create a somewhat alleviated position for the party who aims to protect the environment. This assessment implies that the BoP should not be placed on the party who argues in terms of environmental protection, but on the one who claims the non-harmful nature of the measure at stake. As also argued in the dissenting opinion by Judge Weeramantry in the Nuclear Test case (1995): ‘The law cannot function in protection of the environment unless a legal principle is evolved to meet this evidentiary difficulty, and environmental law has responded with what has come to be described as the precautionary principle.’76

Based on the above considerations, it can be argued that fairness-based arguments could have the effect of imposing the burden of persuasion on the respondent as soon as the applicant establishes a prima facie case (presumption) of violation.77 The application of these rules would thus mean that the party under an obligation to observe the PP bears the BoP that the measures do not constitute78 harm to the environment. In other words, ‘[t]he proponents of potentially harmful activities’ have to provide proof that such activities are ‘environmentally acceptable’.79 This allocation of the BoP would, in turn, make these parties carry out impact assessments before introducing a measure.

PRECAUTION AS JUSTIFICATION AND THE ALLOCATION OF THE BURDEN OF PROOF: THE GABČIKOVO-NAGYMAROS AND EC-HORMONES CASES

Although the ICJ did not explicitly elaborate on the rules relating to the allocation of the BoP in the Gabčikovo-Nagymaros case, it can still be observed that the burden of proving that the wrongfulness of the conduct could be justified by the PP was imposed on Hungary.80 The absence of any BoP-related discussion can probably be attributed to the fact that the allocation was done in accordance with the general rules: the party invoking a rule should prove it. As is commonly known, the justification provided by Hungary was rejected by the Court. Arguably, the combination of conceptual and evidentiary considerations contributed to a great extent to this outcome, which does not, however, necessarily mean that the Court also rejected the whole idea of the PP and relying on it as justification.81

There are two main conceptual reasons why Hungary's justification was not accepted. First, the explanation relating to the PP was placed within the context of the generally accepted exception of ‘state of necessity’. The application of this concept changes the approach towards the PP: the Court was focusing on the existence of a ‘peril’. As the Court argued,

however … serious though these uncertainties might have been they could not, alone, establish the objective existence of a ‘peril’ which is a component of state of necessity. The word ‘peril’ certainly evokes the idea of ‘risk’; that is precisely what distinguishes ‘peril’ from material damage.82

Still, the Court was not satisfied with the level at which the peril was proved. Second and closely related to this previous point, the Court continued its argument that ‘a state of necessity could not exist without a “peril” duly established at the relevant point in time; the mere apprehension of a possible“peril” could not suffice in that respect’.83 This claim of the Court actually goes against the whole idea behind the PP – namely reacting to scientific uncertainty and risk. It is exactly this focus on the existence of a ‘peril’, which does not entirely fit the concept of precaution since the PP centres on the idea of uncertainty as to the possible damage and, in case of lack of certainty, the benefit of the doubt should go to the protection of the environment (in dubio pro natura). Moreover, it also reflects the view on precaution as a result-oriented obligation.

In addition to these conceptual difficulties, the allocation of the BoP arguably also contributed to the rejection of Hungary's justification. The Court relied on the general rules of evidence law, and required the party invoking the legal ground to prove it. So when it comes to justification, it is not the party arguing that there was a violation but the party invoking the ground of justification who bears the BoP in accordance with the general rules. In other words, the application of the general rules means that it is for the ‘applicant’ to prove that there was a violation, and the ‘respondent’ bears the burden of proving that this violation can be justified. According to the Court, in order to justify its violation of treaty obligations Hungary had to prove that the compliance with these obligations would have resulted in harm to the environment – that is, that there was a (certain level of) risk. Had the special rules been applied, it would have sufficed if Hungary raised the presumption of a (certain level of) risk, which would have placed the burden of persuasion on the respondent.

The situation was somewhat different in the EC-Hormones case.84 In addition to the substantive grounds in terms of the WTO Agreement on Sanitary and Phytosanitary Measures (SPS Agreement), two procedural reasons can also be discerned that contributed to the rejection of the EC's arguments. The first procedural reason is that the Panel and the AB rejected the argument that the PP could override Articles 5.1 and 5.5 of the SPS Agreement. They both argued that the PP could not be used as an extra ground of justification of the trade restriction in addition to the provisions of the Agreement. Remarkably, this decision was reached without defining the legal status of PP in customary international law, particularly with regard to the protection of health, which was at stake in this case. While the Panel did not find it important to address it, the AB did devote a couple of words to the legal status of PP in international law – not least to conclude that ‘it is unnecessary, and probably imprudent, for the Appellate Body in this appeal to take a position on this important, but abstract, question’.85

Second, uncertainty around the allocation of the BoP has also had an impact on assessing the matters to the benefit of the applicants. This uncertainty could be related to the unclear position of the PP: whether it was an obligation or justification. As noted by several scholars,86 there was some disagreement between the Panel and the AB in relation to the conceptualization of the relationship between Articles 3.1 and 3.3 of the SPS Agreement. The Panel argued that an obligation–exception relationship existed between these two articles, which had the effect of ‘shifting’ the BoP to the EC under Articles 3.3 and 5 of the Agreement. In other words, in the view of the Panel, the EC had to prove that ‘its measures in dispute meet the requirements imposed by the SPS Agreement’87– namely that the import ban was necessary for the purposes of protecting health (there is scientific justification that a higher level of protection is needed than foreseen by international standards). One cannot but agree with the Panel that, indeed, such a relationship existed between these two rules. The identification of this relationship also changed the allocation of the BoP: the respondent, who invoked the justification, had to bear the burden of persuasion in this regard. However, this change did not result in ‘shifting’ the BoP. Actually, the Panel simply established, even though using a somewhat different language, which party in their interpretation invoked a rule and applied the general BoP rules accordingly.

The AB departed from the Panel's position, arguing that the panel ‘erred in law’ as far as this identification and its effect on the allocation of the BoP were concerned. According to the AB, each of these provisions of the SPS Agreement should be seen as imposing an obligation, the violations of which were alleged by the applicants. This approach to the provisions also influenced the AB's allocation of the BoP – namely, the AB did not share the Panel's view that the burden should entirely be borne by the EC. It argued that:

In accordance with our ruling in United States – Shirts and Blouses, the Panel should have begun the analysis of each legal provision by examining whether the United States and Canada had presented evidence and legal arguments sufficient to demonstrate that the EC measures were inconsistent with the obligations assumed by the European Communities under each Article of the SPS Agreement addressed by the Panel… . Only after such a prima facie determination had been made by the Panel may the onus be shifted to the European Communities to bring forward evidence and arguments to disprove the complaining party's claim.88

Although this language of prima facie case (presumption) seems to indicate that the applicant's BoP was alleviated and thus the special rules on the allocation of the BoP were used, the explanation of the AB on the meaning of a prima facie case shows that this was not case.89 The term ‘prima facie’ seems to have been used to pinpoint who had the burden of persuasion. So the AB placed the burden of persuasion on the United States and Canada with regard to both provisions (Articles 3.1 and 3.3 of the SPS Agreement), given that the PP was also seen as obligation. Even though conceptually incorrect, this approach of the AB might actually be seen as advantageous for the EC since the United States and Canada had to bear the burden of persuasion. However, the fact that the PP was seen as obligation also changed the emphasis of the examination of the case: the focus being on the claims as presented by the applicants rather than as brought forward by the EC.

Although the PP was not successfully relied on in these two cases, some conclusions for the allocation of the BoP can still be drawn from the assessment of these judicial bodies. Based on the theoretical considerations on the general allocation of the BoP, the fact that the PP is used as justification/exception would require that the party invoking it needs to prove that his precautionary approach justifies the non-compliance with his other obligations. This use of the PP, at the same time, also reverses what needs to be proven: instead of the lack of risk, the party needs to provide evidence as to the existence of a potential risk to the environment or human health.90 Given that it is the respondent invoking this principle, according to the general allocation of the BoP, the respondent needs to prove the existence of a (certain level of) risk.

Again, it is worth exploring whether fairness-based arguments would justify the application of the special rules on the allocation of the BoP. It could be argued that the principle of in dubio pro natura would require that the procedural position of a party acting ‘on behalf of’ the environment be facilitated.91 In a similar manner, such an approach would also be required by the effective enforcement of the PP in that it emphasizes the need of reassessing the sustainability of the measure at stake, which can be triggered by such justificatory arguments. In other words, a dynamic interpretation92 of the PP would also favour the application of the special rules concerning the allocation of the BoP. Indeed, the party wanting to continue the application of, or introduce, a certain measure is under an obligation to ensure that his measures are environmentally friendly. Accordingly, it is not unreasonable to place the burden of proving that his measures are not harmful to the environment (or health in the EC-Hormones case) on this party.

THE PRECAUTIONARY PRINCIPLE AND THE ALLOCATION OF THE BURDEN OF PROOF IN DECISION MAKING

Having discussed the allocation of the BoP in dispute settlement, the next question is: what do the above conclusions and arguments mean for decision making and the allocation of the BoP? In other words, who proves what, and what is the effect of fairness-based arguments on the allocation of the BoP in decision making when PP is at stake?

Imagine the following two situations: a company wanting to introduce a new pesticide; and an environmental nongovernmental organization alleging that a certain product or measure causes danger to the environment and thus should be revoked or amended. Who bears the burden of proving that there is/is not (a certain level of) risk of harm to the environment in these cases?

These two scenarios were also sketched in the communication of 2 February 2000 on the PP of the European Commission,93 which concluded that: ‘Measures based on the precautionary principle may assign responsibility for producing the scientific evidence necessary for a comprehensive risk evaluation.’ Regarding the first scenario, the Commission assessed that

Community rules and those of many third countries enshrine the principle of prior approval (positive list) before the placing on the market of certain products, such as drugs, pesticides or food additives. This is one way of applying the precautionary principle, by shifting responsibility for producing scientific evidence.94

Concerning the second scenario, the communication indicated that:

In other cases, where such a prior approval procedure does not exist, it may be for the user, a private individual, a consumer association, citizens or the public authorities to demonstrate the nature of a danger and the level of risk posed by a product or process.95

While in the first scenario the company wants to introduce a new product, in the second scenario the claim relates to the environmental effect of a product or measure that has already been introduced (i.e., there is no prior approval). In terms of the allocation of the BoP, the assessment of the first scenario accepts the application of the special rules relating to the allocation of the BoP, and thus is in accordance with the fairness-based assessment discussed above. In other words, first, the party who wants to introduce a new measure has to carry out a risk assessment and prove that there is no (serious) risk,96 in accordance with the special allocation of the BoP (PP as obligation). If the general evidentiary rules were applicable, those having doubts about the effect of these measures on the environment should prove that there is a (certain level of) risk. In the second scenario, the Commission assessed that the general rules on the allocation of the BoP should be applied: the party referring to the potential harm of an already existing measure needs to prove it. In such situations, the Commission states, the party alleging that an already existing measure has a harmful effect on the environment needs to provide evidence of the existence of a (certain level of) risk, in accordance with the general evidentiary rules (PP as justification). If one accepts that fairness-based arguments justify the application of the special rules concerning the allocation of the BoP in dispute settlement, then it seems difficult to reject the very same arguments for decision making – that is, when the environmental compatibility of a measure or product is challenged, the burden of proving that there is no such risk should be borne by the party having introduced and/or using a certain measure or product.

Two additional observations are worth mentioning with regard to the allocation of the BoP in decision making. First, it is remarkable that in the case of the PP as obligation, the application of special rules seems acceptable, is regulated in many instances and is also implemented in practice,97 while it still has not found its way to dispute settlement. Second, when the PP is used as justification, the application of the general rules on the allocation of the BoP seems commonly accepted in decision making as well as in dispute settlement. Rejecting the application of the special rules in the latter case might be attributed to the wish to avoid any abuse of the possibility to challenge the compatibility of existing measures/products with environmental protection. However, if the concept of prima facie case of the PP98 were also to be used in decision making, it would, similar to discrimination cases, prevent ‘environmentalists’ from unreasonably challenging those measures.

All in all, similar to the approach in the Commission's Communication, one might legitimately expect that a decision-making process follows the evidentiary rules applicable in dispute settlement: first, it should be assessed whether a reference to the PP concerns an obligation or justification; and second, fairness seems to justify in these processes, too, the reliance on the special rules when allocating the burden of proving.

CONCLUDING OBSERVATIONS

In dispute settlement the form in which the PP is relied on has consequences for the allocation of the BoP. This analysis revealed that the uncertainty surrounding the concept of the PP as well as the applicable BoP rules mutually reinforce each other to the detriment of protecting the environment in that dispute settlement bodies find it difficult to accept precaution-based arguments. In this light, it explored whether a fairness-based interpretation of the allocation of the BoP could lead to the application of the special rules when the PP is referred to both as obligation and justification. It has been revealed that the essence of the principle – namely in dubio pro natura– would arguably be challenged should the special rules not be applied. Such a BoP arrangement would also in part eliminate the negative effect of seeing the PP as a result-oriented obligation, which stems from the fact that dispute settlement is inherently backward-looking. The need for a forward-looking approach when the environment is at stake was also emphasized in the Pulp Mills case, where the Court argued that ‘it has to take a forward-looking, prospective approach, engage in a comprehensive risk assessment and embrace a preventive rather than compensatory logic when determining what this risk might entail’.99

As explained, this fairness-based argumentation can also be applied to the allocation of the BoP in decision-making processes. In order to properly and consistently allocate the BoP, one first has to identify in which sense the PP is referred to – that is, as obligation or justification. Interestingly, as the Commission's Communication showed, the special evidentiary rules have gathered more ground in decision making than in dispute settlement, and also when the PP is used as obligation than as justification. In other words, while to some extent the application of special evidentiary rules constitute lege lata in decision making, in inter-State dispute settlement they are ‘only’lege ferenda. This might be attributable to the fact that in dispute settlement (quasi-)judicial bodies are rather hesitant to allocate the onus of proof differently than required by the general rules due to the strict procedural rules applicable. Nevertheless, if one accepts that the ‘shift’ in the BoP can be applied in decision making, it seems difficult to argue that (quasi-)judicial bodies should follow different rules. In a similar vein, fairness-based arguments could justify the application of the special rules on the allocation of the BoP when the PP obtains the form of justification.

The analysis illustrated that a BoP-shifting interpretation is a procedural understanding of the principle and can thus be attached to any substantive conceptualization of the PP. The conceptual consistency provided by such an understanding and the fairness achieved by the special BoP arrangements would, in turn, contribute to the legitimacy of the exercise of public power both in decision-making processes and in dispute settlement, and would be indispensable for an effective protection of the environment. The protection of the environment can thus be facilitated not only by the adoption and application of substantive rules, but also by appropriate procedural arrangements.

This observation on the relationship between the substantive rules and their procedural consequences also highlights an important aspect of what Franck described as the tension100 between ‘normative fairness’ and ‘procedural fairness’. The article confirms that the tension between these two sets of rules can be dissolved, and fairness achieved, if account is taken of both of them – more specifically, if procedural arrangements are made in the light of the substantive rules.

Dr Monika Ambrus works as assistant professor at the Department of Public International Law at the Erasmus School of Law, Erasmus University Rotterdam, The Netherlands. She studied law and linguistics in Budapest, Hungary. She obtained her PhD in 2010 at the Erasmus School of Law; her PhD research was related to the enforcement mechanisms of the prohibition of racial discrimination in the European Union. Currently and in general, she focuses on the legitimacy of judicial decision making and international water law, and in particular, she researches the following topics: genocide, precautionary principle, burden of proof and standard of proof. The author is grateful for the very helpful comments of Professor Ellen Hey on a draft of this article, which improved it considerably, as did the comments provided by the anonymous referee of RECIEL. Any errors that remain are, of course, the author's own responsibility.

Footnotes

  • 1

    See, inter alia, D. Bodansky, ‘Deconstructing the Precautionary Principle’, in: D.D. Caron and H.N. Scheiber (eds.), Bringing New Law to Ocean Waters (Brill, 2004), 381, at 382.

  • 2

    It has been included in numerous multilateral environmental agreements (MEAs). See A. Trouwborst, Evolution and Status of the Precautionary Principle in International Law (Kluwer Law International, 2002), at 63ff.

  • 3

    Rio Declaration on Environment and Development, in Report of the UN Conference on Environment and Development (UN Doc. A/CONF.151/26/Rev. 1 (Vol. I), 14 June 1992), Annex, Principle 15.

  • 4

    See A. Trouwborst, n. 2 above, at 362.

  • 5

    For a deconstruction of the principle, see D. Bodansky, n. 1 above, at 389–391; and A. Arcuri, ‘Reconstructing Precaution, Deconstructing Misconceptions’, 21:3 Ethics and International Affairs (2007), 359.

  • 6

    The PP is ‘only’ mentioned with regard to oceans and seas in the main outcome document of the conference. See United Nations Conference on Sustainable Development, The Future We Want (UN Doc. A/CONF.216/L.1, 19 June 2012), at paragraphs 158 and 167.

  • 7

    The term ‘decision making’ is understood here as including not only decision making in individual cases, but also policy making in certain instances. Although it can also be framed as decision making, dispute settlement by (quasi-)judicial bodies is not included in this term for the purposes of this article. ‘Decision making’ has also been referred to as ‘implementation’. See, e.g., A. Trouwborst, ‘The Precautionary Principle in General International Law: Combating the Babylonian Confusion’, 16:2 Review of European Community and International Environmental Law (2007), 185, at 192.

  • 8

    Arguably, the strong opposition of accepting a ‘shift’ in the allocation of the BoP stems from seeing this ‘shift’ as creating a very low standard of proof. See, e.g., J.S. Gray, ‘Integrating Precautionary Scientific Methods into Decision-making’, in: D. Freestone and E. Hey (eds.), The Precautionary Principle and International Law (Kluwer Law International, 1996), 133, at 136.

  • 9

    T. Franck, Fairness in International Law and Institutions (Oxford University Press, 1998).

  • 10

    I. Scobbie, ‘Tom Franck's Fairness’, 13:4 European Journal of International Law (2002), 909, at 910.

  • 11

    For considering ‘procedural fairness’ as a basis for allocating the BoP in precaution-related cases, see C. Foster, Science and the Precautionary Principle in International Courts and Tribunals: Expert Evidence, Burden of Proof and Finality (Cambridge University Press, 2011), at 240–241; C. Foster, ‘International Adjudication – Standard of Review and Burden of Proof: Australia-Apples and Whaling in the Antarctic’, 21:2 Review of European Community and International Environmental Law (2012), 80, at 86. Although put in different terms, Hepburn et al. also argued that: ‘The distribution of risk is also important, as harms rarely fall equally on those affected. Perhaps most important, due to precaution, the reversal of the BoP means that proponents of risky activities need to take potential impacts into account in their planning, without leaving this aspect solely to those who will be most affected.’ See J. Hepburn, with M.-C. Cordonier Segger and M. Gehring, The Principle of the Precautionary Approach to Human Health, Natural Resources and Ecosystems (Center for International Sustainable Development Law, 2005), at 4.

  • 12

    In other fields of law, where the PP is applied to protect other interests (e.g., health), fairness needs to be assessed in the light of that particular protected aim of the principle. Arguably, however, the analysis provided here equally applies to those areas.

  • 13

    ICJ 20 April 2010, Pulp Mills on the River Uruguay (Argentina v. Uruguay), [2010] ICJ Rep. 14 (Pulp Mills).

  • 14

    ICJ 25 September 1997, Gabčíkovo-Nagymaros Project (Hungary v. Slovakia), [1997] ICJ Rep. 7 (Gabčíkovo-Nagymaros).

  • 15

    WTO DS 18 August 1997, EC – Measures Concerning Meat and Meat Products, WT/DS26/R/USA (EC-Hormones, Panel); WTO AB 16 January 1998, EC – Measures Concerning Meat and Meat Products, WT/DS26/AB/R (EC-Hormones).

  • 16

    On the applicability of the PP to health-related issues see, inter alia, International Law Association, ILA New Delhi Declaration of Principles of International Law Relating to Sustainable Development (UN Doc. A/CONF.199/8, 9 August 2002); Communication from the Commission of 2 February 2000 on the Precautionary Principle, COM (2000)1 final.

  • 17

    See J.B. Wiener, ‘Precaution’, in: D. Bodansky, J. Brunnée and E. Hey (eds.), The Oxford Handbook of International Environmental Law (Oxford University Press, 2007), 597, at 599. For the development of the PP, see D. Freestone and E. Hey, ‘Origins and Development of the Precautionary Principle’, in: D. Freestone and E. Hey (eds.), n. 8 above, 3, at 3–15; J. Cameron and J. Abouchar, ‘The Status of the Precautionary Principle in International Law’, in: D. Freestone and E. Hey (eds.), n. 8 above, 29 and W.Th. Douma, The Precautionary Principle: Its Application in International, European and Dutch Law, PhD Thesis (2002).

  • 18

    See J.B. Wiener, n. 17 above, at 601.

  • 19

    Ibid. See also Gabčíkovo-Nagymaros, n. 14 above; Pulp Mills n. 13 above; EC-Hormones, n. 15 above.

  • 20

    E. Hey, ‘Considerations Regarding the Hormones Case, the Precautionary Principle and International Dispute Settlement Procedures’, 13:1 Leiden Journal of International Law (2000), 239, at 244; J.B. Wiener, n. 17 above, at 598.

  • 21

    C. Tinker, ‘State Responsibility and the Precautionary Principle’, in: D. Freestone and E. Hey (eds.), n. 8 above, 53, at 70.

  • 22

    On the importance of the underlying ‘theories of administrative constitutionalism’ for the interpretation of the PP, see E. Fisher and R. Harding, ‘The Precautionary Principle and Administrative Constitutionalism: The Development of Frameworks for Applying the Precautionary Principle’, in: E. Fisher, J. Jones and R. von Schomberg (eds.), Implementing the Precautionary Principle: Perspectives and Prospects (Edward Elgar, 2006), 113.

  • 23

    Rio Declaration, n. 3 above, Principle 15. See also the Bergen Declaration of Ministers, reproduced in: R. Churchill and D. Freestone (eds.), International Law and Global Climate Change (Springer, 1991), at 344.

  • 24

    Compare with D. Bodansky, n. 1 above, at 383–386.

  • 25

    For a clear opposition of this third version, see J. Ellis and A. FitzGerald, ‘The Precautionary Principle in International Law: Lessons from Fuller's Internal Morality’, 49:3 McGill Law Journal (2004), 779, at 782–783.

  • 26

    See also A. Trouwborst, n. 7 above, at 191–192.

  • 27

    See J.B. Wiener, n. 17 above, at 604–606. See also C. Tinker, n. 21 above, at 55–56; D. Bodansky, n. 1 above, at 390–391; and P.H. Sand, ‘The Precautionary Principle: A European Perspective’, 6:3 Human and Ecological Risk Assessment (2000), 445, at 448.

  • 28

    This position is contrary to the generally accepted view, which holds that ‘[s]ome versions of the precautionary approach would reverse’ the traditional allocation of the BoP. D. Bodansky, n. 1 above, at 390–391 (emphasis added). See also C. Tinker, n. 21 above, at 55–56; J. Jones, ‘Regulatory Design for Scientific Uncertainty: Acknowledging the Diversity of Approaches in Environmental Regulation and Public Administration’, 19:3 Journal of Environmental Law (2007), 347, at 354–355; N.M. Sachs, ‘Rescuing the Strong Precautionary Principle from Its Critics’, 2011:4 University of Illinois Law Review (2011), 1285. The position taken in this article seems to be supported by Trouwborst, who discussed the three legs of the PP, with the third leg being precautionary action which incorporates the BoP rules. However, he also explains that the ‘shift’ in the BoP is not always applied. A. Trouwborst, n. 7 above, at 192.

  • 29

    For acknowledging this distinction between a ‘claim’ and ‘exception’, and the difficulties surrounding exceptions, see also C. Foster, Science and the Precautionary Principle, n. 11 above, at 244; and C. Foster, ‘International Adjudication – Standard of Review and Burden of Proof’, n. 11 above, at 87.

  • 30

    Pulp Mills, n. 13 above.

  • 31

    Although it has been argued in the context of State responsibility that it is difficult to establish ‘the existence of a clear legal duty or obligation’, the article proceeds on the basis of the understanding that a substantive understanding can be attached to the concept, and focuses only on the procedural consequence of PP as obligation. In other words, seeing PP as an obligation in this context is relevant for the discussion on the PP's procedural aspect, but this discussion does not seek to assess whether or not indeed there is a clear obligation established by the PP or determine its possible content. See C. Tinker, n. 21 above, at 57.

  • 32

    Pulp Mills, n. 13 above, at paragraph 170 (emphasis added). See also ibid., Separate Opinion of Judge Cançado Trindane, at paragraph 108.

  • 33

    Although not formulated in such terms, the argument of Hepburn et al. also refers to precaution in the form of justification. In their view, ‘precaution shifts the burden of proof from those proposing measures that will reduce uncertain risks to these systems, to those supporting development’. See J. Hepburn, with M.-C. Cordonier Segger and M. Gehring, n. 11 above, at 5. See also L. Hancher, ‘EC Environmental Policy: A Pre-cautionary Tale?’, in: D. Freestone and E. Hey (eds.), n. 8 above, 187, at 202–203.

  • 34

    Gabčikovo-Nagymaros, n. 14 above. On the application of the PP by the ICJ in the Gabčikovo-Nagymaros case, see also W.Th. Douma, n. 17 above, at 164–171.

  • 35

    EC-Hormones, n. 15 above. On the application of the PP in this case, see also W.Th. Douma, n. 17 above, at 184.

  • 36

    EC-Hormones, Panel, n. 15 above, at paragraph 184.

  • 37

    Gabčikovo-Nagymaros, n. 14 above, at paragraph 40. It was referred to as a ‘state of ecological necessity’.

  • 38

    J. Kokott, The Burden of Proof in Comparative and International Human Rights Law: Civil and Common Law Approaches with Special Reference to the American and German Legal Systems (Kluwer Law International, 1998), at 9–10. See also C. Bosse, Bewijslastverdeling in het Nederlands en Belgisch Arbeidsrecht (Kluwer, 2003), at 12–14.

  • 39

    See J. Kokott, n. 38 above, at 9.

  • 40

    Ibid. See also U. Erdal, ‘Burden and Standard of Proof in Proceeding under the European Convention’, 26:1 European Law Review (2001), 68, at 77.

  • 41

    Although scholars in general do not specify which BoP they refer to, they generally discuss the burden of persuasion. Due to the non-specification the two meanings of the BoP are often mixed, leading to uncertainty concerning the proper allocation of the BoP.

  • 42

    See also Bosse, n. 38 above, at 15.

  • 43

    Laying down rules for the allocation of the burden of production is only necessary in terms of the specific course of a particular procedure – that is, when the word is given to one or the other party.

  • 44

    This rule is also referred to as the rule of ei incumbit probatie qui dicit, non qui negat (‘the burden of proof lies with who declares, not who denies’).

  • 45

    See, inter alia, C. Bosse, n. 38 above, at 16; J. Kokott, n. 38 above, at 11; C. Foster, ‘International Adjudication – Standard of Review and Burden of Proof’, n. 11 above, at 85.

  • 46

    An applicant can also become respondent with regard to allegations of violating other rules by the original respondent. The discussion here focuses on one specific matter, with regard to which an applicant and a respondent can be identified.

  • 47

    See, inter alia, K. Monaghan, Equality Law (Oxford University Press, 2007), at 563.

  • 48

    See also U. Erdal, n. 40 above, at 82.

  • 49

    For more details, see M. Ambrus, Enforcement Mechanisms of the Racial Equality Directive and Minority Protection: Theory and Four Case Studies (Eleven International, 2011).

  • 50

    See, inter alia, Council Directive 97/80/EC of 15 December 1997 on the Burden of Proof in Cases of Discrimination Based on Sex (‘Burden of Proof Directive’), [1998] OJ L14, Article 4; Council Directive 2000/43/EC of 29 June 2000 Implementing the Principle of Equal Treatment between Persons Irrespective of Racial or Ethnic Origin (‘Racial Equality Directive’), [2000] OJ L180, Article 8.

  • 51

    See also J. Kokott, n. 38 above, at 2.

  • 52

    For a position according to which it is ‘only’ the evidential BoP which switches to the respondent see, e.g., INTERIGHTS, Written Comments in the case of Nachova and others v. Bulgaria (2005), found at: <http://www.justiceinitiative.org/db/resource2/fs/?file_id=15404>.

  • 53

    G. Bindman, ‘Proof and Evidence of Discrimination’, in: B.A. Hepple and E.M. Szyszczak (eds.), Discrimination: The Limits of Law (Mansell, 1992), 50, at 58 (emphasis added).

  • 54

    Proposal for a Council Directive Implementing the Principle of Equal Treatment between Persons Irrespective of Racial or Ethnic Origin, COM (1999)566 final, at 9. See, inter alia, ECJ, Case C-127/92, Dr Pamela Mary Enderby v. Frenchay Health Authority and Secretary of State for Health, [1993] ECR I-5535, at paragraph 13 (Enderby).

  • 55

    ECJ, Case C-109/88, Handels- og Kontorfunktionærernes Forbund I Danmark v. Dansk Arbejdsgiverforening, acting on behalf of Danfoss, [1989] ECR 3199 (Danfoss).

  • 56

    Ibid., at paragraph 11 (emphasis added).

  • 57

    Enderby, n. 54 above, at paragraph 13.

  • 58

    Burden of Proof Directive, n. 50 above.

  • 59

    Council Directive 2000/78/EC of 27 November 2000 Establishing a General Framework for Equal Treatment in Employment and Occupation, [2000] OJ L303.

  • 60

    Council Directive 2004/113/EC of 13 December 2004 Implementing the Principle of Equal Treatment between Men and Women in the Access to and Supply of Goods and Services, [2004] OJ L373; Directive 2006/54/EC of the European Parliament and of the Council of 5 July 2006 on the Implementation of the Principle of Equal Opportunities and Equal Treatment of Men and Women in Matters of Employment and Occupation (‘Recast Directive’), [2006] OJ L 204. This latter Directive replaced the one on the BoP with effect from 15 August 2009.

  • 61

    See, inter alia, C. Bosse, n. 38 above, at 133–134; J. Luxton, ‘Equality and Sex Discrimination in the European Union: Is Shifting the Burden of Proof the Answer?’, 17:2 Dickinson Journal of International Law (1999), 357, at 358.

  • 62

    F. Palmer, ‘Re-dressing the Balance of Power in Discrimination Cases: The Shift in the Burden of Proof’, 4 European Anti-Discrimination Law Review (2006), 23, at 24 (emphasis added). See also K. Monaghan, n. 47 above, at 563–564 and 571; and E. Ellis, EU Anti-discrimination Law (Oxford University Press, 2005), at 98 and 110.

  • 63

    See E. Ellis, n. 62 above, at 98.

  • 64

    Enderby, n. 54 above, at paragraph 13 (emphasis added). See also ECHR 6 January 2005, Hoogendijk v. Netherlands, No. 58641/00.

  • 65

    The case law of, among others, the European Court of Human Rights on Article 14 clearly shows how important the appropriate allocation of the BoP is for the effective enforcement of human rights. See, inter alia, INTERIGHTS, n. 52 above, at 1; J. Kokott, n. 38 above, at 57; and C. Bosse, n. 38 above, at 140–141. The special allocation of the BoP is, arguably, also applied in cases other than discrimination. See, e.g., UN Human Rights Council (HRC) 2 July 1994, Mukong v. Cameroon, No. 458/1991; and UN HRC 4 December 1995, Ribitsch v. Austria, No. 18896/91, at paragraph 34. See also U. Erdal, n. 40 above.

  • 66

    See J. Kokott, n. 38 above, at 1 (emphasis added). See also INTERIGHTS, n. 52 above.

  • 67

    Burden of Proof Directive, n. 50 above, Recital (17).

  • 68

    Racial Equality Directive, n. 50 above, Recital (21). See also COM (1999)566 final, n. 54 above, at 9.

  • 69

    Pulp Mills, n. 13 above, at paragraph 101.

  • 70

    McIntyre labelled this request a ‘radical contention’. O. McIntyre, ‘The Proceduralisation and Growing Maturity of International Water Law’, 22:3 Journal of Environmental Law (2010), 475, at 493.

  • 71

    Pulp Mills, n. 13 above, at paragraph 162.

  • 72

    Ibid., Joint Dissenting Opinion of Judges Al-Khasawneh and Simma, at paragraph 22. See also C. Tinker, n. 21 above, at 54; and A. Trouwborst, ‘Prevention, Precaution, Logic and Law: The Relationship between the Precautionary Principle and the Preventative Principle in International Law and Associated Questions’, 2:2 Erasmus Law Review (2009), 107.

  • 73

    D. Kazhdan, ‘Precautionary Pulp: Pulp Mills and the Evolving Dispute between International Tribunals over the Reach of the Precautionary Principle’, 38:2 Ecology Law Quarterly (2011), 527, at 528, 545–546.

  • 74

    In general it has been argued that State responsibility ‘best protects the rights of innocent victims of environmental harm by shifting the burden of proof of causation’. C. Tinker, n. 21 above, at 68–69. See also J. Ellis and A. FitzGerald, n. 25 above, at 783.

  • 75

    See also C. Foster, ‘International Adjudication – Standard of Review and Burden of Proof’, n. 11 above, at 86.

  • 76

    ICJ 22 September 1995, Nuclear Tests (New Zealand v. France), [1995] ICJ Rep. 1995, Dissenting Opinion of Judge Weeramantry, at 342 (‘Nuclear Tests’).

  • 77

    A discussion on what a prima facie case of not being precautious means requires a separate study, and is beyond the reach of this article.

  • 78

    The required extent to which the lack of such risk (and thus scientific (un)certainty) should be proven depends on the substantive understanding of the PP at stake.

  • 79

    See O. McIntyre, n. 70 above, at 493.

  • 80

    Gabčikovo-Nagymaros, n. 14 above, at paragraphs 49–54. In the words of the Court: ‘would have permitted Hungary, without incurring international responsibility, to suspend and abandon works’. Ibid., at paragraph 49.

  • 81

    See also D. Kazhdan, n. 73 above, at 540–541.

  • 82

    Gabčikovo-Nagymaros, n. 14 above, at paragraph 54.

  • 83

    Ibid. (emphasis added).

  • 84

    On the (non-)application of the PP in this case, see E. Hey, n. 20 above.

  • 85

    EC-Hormones, n. 15 above, at section VI.

  • 86

    D. Unterhalter, ‘Allocating the Burden of Proof in WTO Dispute Settlement Proceedings’, 42:2 Cornell International Law Journal (2009), at 209; H. Horn and P.C. Mavroidis, ‘The Burden of Proof in Environmental Disputes in the WTO: Legal Aspects’, 18:2 European Energy and Environmental Law Review (2009), 112, at 116.

  • 87

    EC-Hormones, Panel, n. 15 above, at section VIII.D.6.

  • 88

    EC-Hormones, n. 15 above, at section IV (emphasis added).

  • 89

    Ibid. See also J.J. Barceló III, ‘Burden of Proof, Prima Facie Case and Presumption in WTO Dispute Settlement’, 42:1 Cornell International Law Journal (2009), 24.

  • 90

    Again, depending on the underlying concept of the PP, the nature and extent of scientific uncertainty that is accepted might vary. In other words, such a BoP does not mean that the party bearing the BoP needs to prove the full absence of a risk, as indicated in general by the opponents of the special allocation of the BoP. See J. Ellis and A. FitzGerald, n. 25 above, at 797.

  • 91

    See also D. Freestone and E. Hey, n. 17 above, at 13. In their words, the precautionary approach would entail that ‘legal, administrative and technical procedures that facilitate the implementation of this approach must be applied and, where not available, developed’.

  • 92

    See E. Hey, n. 20 above.

  • 93

    See COM (2000)1 final, n. 16 above.

  • 94

    Ibid., at 20. See also D. Freestone and E. Hey, n. 17 above, at 6.

  • 95

    See COM(2000)1 final, n. 16 above, at 20.

  • 96

    That is, there is a particular level of certainty that no harm will occur.

  • 97

    See A. Trouwborst, n. 7 above, at 192–193.

  • 98

    Hey and Freestone argued that ‘the threshold of significant risk has become easier to cross, which from a legal perspective would imply that once a prima facie case is made that a risk exists, then scientific uncertainty works against the potential polluter rather than, as in the past, in his/her favor’. D. Freestone and E. Hey, n. 17 above, at 13.

  • 99

    Pulp Mills, n. 13 above, at paragraph 22.

  • 100

    See T. Franck, n. 9 above, at 477.

Ancillary