The EU in multilateral environmental compliance mechanisms: an outside view

With reference to its unique characteristics, the European Union (EU) regularly requests a special position in treaty cooperation or external judicial control mechanisms. Recurrently, these requests are successful and lead to the EU being treated differently from other treaty parties. These situations have been captured by the concept of ‘European exceptionalism’. EU requests for special treatment can also bewitnessed in the supportive and facilitative procedures of compliance mechanisms in international environmental law. In those mechanisms, however, EU requests for special treatment are subject to careful scrutiny, and are even met with strong opposition by treaty institutions and treaty partners. Taking a closer look at the EU’s participation in compliance mechanisms, the present article discusses how certain unique EU characteristics may prompt an EU request for special treatment under compliance mechanisms and explores how compliance institutions and treaty partners have treated existing requests so far. With this outside perspective of non-EU actors, it is possible to understand where such requests can be Europe and the World: A law review https://doi.org/10.14324/111.444.ewlj.2021.33 The EU in multilateral environmental compliance mechanisms: an outside view 2 successful and where they fail to be. In this way, the insights gained permit reflection upon the EU’s participation in compliance mechanisms and whether it truly constitutes a further phenomenon of ‘European exceptionalism’.

mechanisms. Equipped with these insights, it identifies EU characteristics which may warrant special treatment in the eyes of external actors and reflects on whether the treatment accorded indeed allows for understanding the EU's participation in compliance mechanisms as yet another phenomenon of 'European exceptionalism'.

Compliance mechanisms in multilateral environmental agreements
Within a rather short period of time, compliance mechanisms have become 'a sort of "must"' in multilateral environmental agreements (MEAs). 9 These mechanisms have their roots in the idea of treaty management as pioneered by Chayes and Chayes, who argued centrally that non-compliance under treaties was not a wilful act of treaty parties. 10 After all, States have negotiated the respective treaty to which they are party and would generally consider such a negotiated outcome less inequitable than the available alternatives. Consequently, there 'was no reason to suppose that States are prone ordinarily to disregard [their] obligation to obey the law'. 11 Rather than a wilful act, Chayes and Chayes maintained that non-compliance was a consequence of different factors such as capacity problems but also of treaty design. 12 In their research they highlighted, for example, that the design of environmental treaties such as the Montreal Protocol had generally challenged States. 13 While the reduction obligation under the Montreal Protocol is framed as a State obligation, it was ultimately necessary to regulate behaviour of businesses and individuals. Whether the respective domestic systems would be successful in this regard would only become apparent over time. Thus, States 'have not been able to construct such systems with the confidence that they will achieve the desired objective'. 14 It would follow that reasons such as these, which underlie non-compliance, could clearly not be resolved by sanctioning non-compliant behaviour. 15 Instead, it was much more helpful to provide treaty parties with assistance and support, and mechanisms allowing for clarifying and adapting treaty obligations. 16 In this spirit, mechanisms of treaty management would be based on a cooperative and inclusive dialogue with the treaty parties. 17 The aim of this dialogue would be to identify the reasons for non-compliance and suitable solutions. 18 The continued interaction between the treaty institutions and parties facilitated by this dialogue was meant to ultimately induce a sense of obligation to comply with treaty obligations. 19 Compliance mechanisms, as included in MEAs today, effectuate Chayes and Chayes' idea of treaty management. 20 While these treaty-specific mechanisms consist of multiple elements, 21 non-compliance procedures are a central one: these procedures are designed to allow for the vital constructive dialogue between treaty institutions and treaty parties, seeking to instigate compliance. 22 This purpose, and their design, distinguishes them from international courts and tribunals.
Non-compliance procedures can be initiated in different ways ('triggering'). Central is the possibility for a treaty party to initiate a procedure with regard to its own situation ('self-triggering'). In this way, a treaty party can submit to the supportive procedure to discuss any difficulties it experiences with achieving compliance. 23 So-called party-to-party triggering is more controversial. 24 These possibilities permit a treaty party to request the opening of the procedure in respect of another treaty party. Yet, even though these possibilities exist to allow for protecting the common interest of treaty parties, 25 they are very rarely used. Despite the supportive nature of the process, treaty parties appear to consider it inappropriate to put another party under scrutiny. 26 To ensure non-compliance procedures are nevertheless used as regularly as possible, additional possibilities to initiate these procedures exist. At times, treaty institutions have the possibility to trigger the procedure where they become aware of any difficulties of a party. 27 In some settings, even members of the public, mainly environmental non-governmental organisations (eNGOs), are accorded triggering rights. 28 However, State actors appear reluctant to provide such rights more generally. 29 This tendency is indeed regrettable as an opening of the mechanism in this way could enhance its effectiveness, and even its legitimacy. 30 Once initiated, the non-compliance procedure is generally structured as a two-stage process. Increasingly, the first stage is governed by treaty institutions which are specifically mandated to deal with compliance matters (further 'compliance bodies'). 31 These compliance bodies have different compositions. The compliance body under the Convention on Long-Range Transboundary Air Pollution (LRTAP), 32 for example, comprises a limited number of State representatives. 33 Conversely, the compliance body under the Water Convention 34 is staffed with experts serving in their individual capacity. 35 Irrespective of their formal composition though, compliance bodies are mandated to deal with compliance matters by assessing a situation of a treaty party in view of treaty obligations. 36 The assessment of compliance bodies is included in a report or the like, usually directed to the treaty's plenary political body. 37 Widely, compliance bodies are competent to include recommendations for measures to support a treaty party found non-compliant in rectifying the situation. 38 Examples of such measures include access to technical assistance or the offering of expert advice. 39 Taking into account the role and competences of compliance bodies in the two-stage process, their assessment cannot be considered binding. 40 Nevertheless, their assessment clearly has immense value for the treaty party to whose situation they relate as it provides clarification for necessary behaviour under a treaty. In this sense, the assessment can also be an instructive resource to other treaty parties for evaluating their own situation.
The second stage of the non-compliance procedure is governed by the plenary political body of a treaty, usually the Conference or Meeting of the Parties (CoP or MoP). The plenary political body builds on the results of the procedure's first stage and reviews the compliance assessment. 41 Ultimately, the purpose of this review is to find acceptance among treaty parties for that assessment, its findings and any proposed measures. 42 This acceptance is ultimately reflected in the adoption of a so-called non-compliance decision, usually by consensus. 43 By confirming a situation of non-compliance, a non-compliance decision highlights that a party does not yet fulfil obligations it owes its treaty partners. 44 Recommendations included in the decision are meant to provide that party with options for possible action to rectify the situation, 45 while not binding the party to a specific action. 46 In this light, non-compliance decisions do not establish a new obligation to take action. Rather, this obligation already stems from the respective existing treaty obligation and the principle of pacta sunt servanda. 47 Nevertheless, non-compliance decisions clearly play a crucial role in ensuring parties meet treaty standards: these decisions effectively underline and, together with the included assessment, clarify a party's obligations and inform the taking of action. 48 This role is further emphasised when measures such as reporting requirements demand that treaty parties account for the action they have taken, 49 or follow-up procedures exist to evaluate the progress in rectifying a situation of non-compliance. 50 While not obliging a party to act, such soft pressure can, in practice, be equally powerful in inducing a change in behaviour. 51

EU participation in compliance mechanisms: a closer look
The EU is party to numerous MEAs, 52 of which a large number have also established compliance mechanisms. 53 While MEAs regularly accommodate, to some extent, the EU's unique features, far less is known about any such accommodating in the context of compliance mechanisms. An initial idea of which characteristics may require accommodating, and how this could be done to create a special position for the EU, can be developed by considering past situations in similar contexts. This initial idea serves as a foundation for then examining case studies from compliance practice to explore whether compliance institutions and treaty partners indeed grant the EU the special treatment it desires.

The EU as a party to MEAs and to compliance mechanisms
At the international level, the EU's unique characteristics are frequently accommodated in treaty relations, also in MEAs. As compliance mechanisms are based in MEAs, the special treatment granted to the EU under treaties is likely to be relevant to its participation in these mechanisms. The related considerations can, however, also more broadly inspire ideas for how EU special treatment under compliance mechanisms may present itself. This is also true for experiences gained in the context of certain external control mechanisms where EU characteristics have challenged their design. After all, compliance mechanisms are also control mechanisms, even if of a supportive nature. These situations are thus equally valuable for developing an idea of which EU characteristics may be accommodated, and how, under compliance mechanisms.

The EU's nature as an international organisation
The EU's status as an international organisation of sovereign States with its own international personality has prompted accommodating in international cooperation in different ways. 54 On the one hand, with the EU not being a State actor, treaties need to authorise its participation in a treaty. For this purpose, treaties typically include specific provisions in relation to a regional (economic) integration organisation (REIO), 55 enabling the EU to join a treaty. 56 On the other hand, as the EU has only limited treaty-making powers, treaties regularly accommodate the internal competence divide between the EU and its Member States (MS) by allowing for their parallel participation. 57 Since internally the EU shares the environmental competence with its MS, parallel participation is the regular case in MEAs. 58 In the case of parallel participation, the EU and the MS participate as parties in their own right so that they each owe their treaty partners fulfilment of the entire treaty. 59 Nevertheless, treaty clauses often require a clarification on the competence divide between the EU and its MS. 60 The idea of a declaration of competence is to provide more clarity on the respective responsibilities of the EU and its MS under a treaty. 61 This clarity is specifically relevant in case of disputes and alleged failures to fulfil these responsibilities: legally, both the EU and its MS can be held accountable for fulfilment of the entire treaty. Yet, factually, the competence divide at EU level indicates who -the EU or the MS -has the actual ability to resolve the issue under dispute. The declaration of competence thus seeks to assist treaty partners in identifying against whom a legal action may be more promising. 62 The question of who can be held accountable poses itself similarly in compliance mechanisms in the event of parallel participation when treaty institutions or other treaty partners trigger proceedings. A declaration of competences required under the treaty may serve as a basis to judge this question. However, given the practical difficulties with these declarations, a pragmatic alternative could be to allow for the joint participation of the EU and its MS in compliance procedures. 63 54  The case of parallel participation also requires special rules for the EU and its MS when it comes to their party rights. With regard to voting rights, treaties or other (constituting) legal instruments tend to include specific rules which seek to ensure that the EU and its MS do not exercise their rights simultaneously. 64 Instead, where the MS exercise their rights, the EU must abstain and vice versa. 65 How the EU and its MS arrange this exercise of rights is generally an internal matter. 66 However, treaties may condition the participation of international organisations by the participation of their constituting member States, thereby not providing the former with a fully autonomous status. 67 In this case, international law indirectly addresses the internal matter of the competence divide between the EU and its MS, and forces them to act jointly. 68 Any existing specific voting rules relating to treaty-based institutions are also relevant to compliance mechanisms, in any case for the plenary political body at the last stage of the procedure.

The EU's autonomy
The principle of autonomy has been pointedly described as 'part of the orthodoxy of EU law'. 69 Indeed, from an EU law perspective, the principle and its constituting elements are essential for arguing why the EU legal order is distinct from that of international law. 70 In this sense, its autonomy can be understood as one of the EU's unique characteristics.
In the context of the development or submission to external (quasi-)judicial control mechanisms, the EU has indeed recurrently requested to have this autonomy accommodated. Evidence of this can be found in relation to the EEA Agreement, 71 which was renegotiated after the EU's Court of Justice (CJEU) judged it incompatible with the principle of autonomy. 72 More recently, the Court found also the accession agreement to the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR) to be in breach of the EU's autonomy, 73 requiring the EU to renegotiate this agreement in order to deliver on its constitutional promise. 74 The principle of autonomy can thus prompt the EU to make requests for special treatment, 75 thereby making this principle relevant to external actors, potentially, also to compliance institutions and treaty partners in the context of compliance mechanisms.
It was pointed out earlier that compliance mechanisms differ from external (quasi-)judicial control mechanisms in their purpose. 76 A further difference also lies in the effects of their respective decisions: while (quasi-)judicial decisions are binding on a party, non-compliance decisions are not; rather they factually assist and softly put pressure on a party to act accordingly. Nevertheless, the Court's recent assessment of CETA, the Comprehensive Economic and Trade Agreement between the EU and Canada, may suggest that autonomy concerns are not exclusively confined to cases of binding (quasi-)judicial decision-making. 77 It is thus worth having a brief look at this case.
In its CETA Opinion, 78 the Court first recalled certain competences that must not be accorded to the respective tribunal in order to observe the EU's autonomy. Unsurprisingly, it reaffirmed that the decision whether a measure under scrutiny falls within the competence of the EU or the MS must remain with the Court, as this was ultimately a question of EU law. 79 In a similar vein, the Court restated that the external body must not have the competence to interpret any type of EU law. 80 Compliance institutions are certainly not accorded the competence to provide an interpretation of EU law, 81 binding or otherwise. Yet, the issue of the competence divide may occur once a non-compliance procedure is initiated in respect of either the EU or an MS and a domestic measure. While a decision by a compliance institution on the correct respondent in the case would not have binding effect, it is unclear whether such decision-making may still be considered problematic. A compliance institution could potentially avoid deciding the competence question, though, if it always addressed the EU and its MS. 82 In respect of the CETA, the Court was also prepared to make some concessions on the need to protect the autonomy of the EU legal order. It found that the competences of the respective tribunal were sufficiently limited in order not to impinge on the competences of the EU institutions: the tribunal was competent neither to annul an EU measure nor to request that the EU render the measure compatible with CETA. 83 The competences of compliance institutions are limited in a similar way inasmuch as their competence to find on a situation of non-compliance does not include the power to directly eliminate the source of non-compliance. Instead, compliance institutions are empowered to make recommendations 84 and take predominantly supportive measures which allow a party to come back into compliance. 85 With their decisions, compliance institutions can therefore not prescribe a certain policy path for the EU should it be found non-compliant. 86 Notably, in its Opinion, the Court then suggested that indeed factual pressures are also capable of raising autonomy concerns. Specifically, it indicated that a competence to issue a penalty in respect of the party in breach of treaty obligations would be problematic. Yet, also the award of damages could exercise external pressure on the EU institutions, leaving them with no other option than revising their policy choice. 87 The issue of factual pressure may indeed be relevant in the context of compliance mechanisms. After all, non-compliance decisions are public and thus generally accessible. 88 The inclusion of measures in response to a situation of non-compliance may thus subject the EU to external (political) pressure to implement them. 89 Within the treaty system, such pressure could result from follow-up procedures seeking to verify whether parties actually rectify a situation of non-compliance. 90 However, even if such pressure could be considered high, it would not necessarily leave the EU with no other option than to revise a policy or legislative act. How the EU can rectify a situation of non-compliance 77 In this vein, Ankersmit (n 2) 338f. 78  depends on the reason identified for non-compliance: it may be a failure to submit a report; it may, however, also be an institutional practice lying at odds with treaty obligations. On a general level, it is thus not likely that the pressures building around non-compliance decisions raise the same autonomy concerns as indicated in the CETA Opinion. 91 Yet, the fact that the Court even considered factual pressure to be possibly problematic demonstrates the ongoing evolution of this principle in the case law and should spark some wariness.

EU requests for special treatment in compliance mechanisms
Having shown how certain EU characteristics may prompt an EU request for special treatment in the context of compliance mechanisms, it is appropriate to now turn to the practice. The following discusses three case studies of non-compliance procedures in which the EU made requests for special treatment by reference to its unique characteristics. While in some cases the EU special treatment was granted, in others it was opposed. By taking the perspective of compliance institutions and treaty partners, the focus in these case studies is to shed more light on whether and how the EU was treated differently from other treaty parties following its requests.

Parallel participation of the EU and its MS: the case of an EU emissions ceiling
The first case study relates to an EU request for special treatment under the compliance mechanism of the Gothenburg Protocol, 92 which effectuates the LRTAP. The EU had acceded to the Protocol in 2003, 93 joining its then 15 MS. Parties to the Protocol are required to limit their emissions of certain air pollutants to specific levels set out in the Protocol's Annex II. 94 The EU is no exception in this regard: the Annex includes an emissions ceiling also for the EU, which is the sum of its MS' ceilings. In 2012 the EU had reported on the 2010 and 2011 emissions of NOx from its 15 MS. These emission data reports revealed that emissions of NOx in the EU were well above the dedicated EU emissions ceiling. Based on that information, the treaty's Secretariat concluded that the EU was in non-compliance with its obligations. 95 The Secretariat therefore referred the EU to the treaty-based compliance body, the Implementation Committee. 96 In the proceedings before the Committee, the EU admitted that indeed, its report showed emissions above its ceiling. Yet, it argued that the emission inventories of some of its MS would still be subject to further adjustments. The exceedance of the emissions ceiling would thus probably be further reduced. 97 Appropriately, the non-compliance procedure should be suspended until the EU MS had finalised their activities. The Committee rejected that EU request. It underlined that, while the EU's explanations are understandable, they did not justify emissions in excess of its ceiling. By becoming a party to the Protocol, the EU had taken up the obligation to reduce emissions below a fixed ceiling. This EU ceiling was absolute and not dependent upon the individual ceilings of its MS. 98 Nevertheless, although not explicitly raised by the EU, the Committee considered whether the EU's nature as an international organisation required a different view on this point. However, the Committee highlighted that the EU did not declare or argue it lacked competence to fulfil its obligation. It could thus be presumed that it had the necessary ability to actually meet that obligation. 99 The Committee agreed with the EU, though, on how to understand the EU emissions ceiling. It accepted that an emissions ceiling typically relates to a specific territory. At the time of the case, the EU's territory indeed comprised that of 28 individual States. However, when the EU acceded to the Protocol, only 15 States were MS of the EU. Thus, the EU emissions ceiling had to be understood to relate to the territory of its first 15 MS. 100 In view of these considerations, the Committee adopted a draft decision on the compliance matter. In that decision it found that the EU was in non-compliance with its emissions reduction obligations. 101 The Executive Body as the treaty's plenary political body, without further discussion, 102 adopted this decision. 103 In this case, the compliance institutions followed up only one of the EU requests, the one in relation to the interpretation of the EU's emissions ceiling. The compliance body accepted that an obligation relating to the territory of a treaty party requires distinct considerations in respect of an international organisation. Specifically, it would require taking into account the development of the EU since its accession, and the type of obligation it undertook with its accession. While the compliance body refers in these considerations to the special nature of the EU as an international organisation, it is difficult to see how a State party would be treated differently. Accepting that an emissions ceiling relates to a territory, as the Implementation Committee clearly argued, a change in a State party's territory since its treaty accession would also have to be taken into account. While for a State party this situation may arguably be less likely to occur than for an international organisation, the considerations still lead to the same conclusion. The fact that the Implementation Committee followed up the EU's request as regards the interpretation of its emissions ceiling can thus not be understood as special treatment.
In respect of the EU's first request, relating to its treaty performance, the compliance body did not come to understand that the EU's nature as an international organisation warrants special treatment: the EU's reference to its factual dependence on the MS for fulfilling its reduction obligation was not accepted. Rather, the Implementation Committee highlighted how the EU is a treaty party like all others. As a party in its own right, the EU must be prepared to deliver on the obligations it undertook. However, the related considerations of the Implementation Committee indicate that it would have been prepared to take the EU-MS competence situation into account -something that international law does not demand. 104 While in this instance the EU request was not successful, inasmuch as the EU was not treated differently as a result, the special nature of the EU clearly had a bearing on the compliance body's considerations. It is questionable, though, whether, in view of the emphasis on the equality of the EU with other treaty parties, this case can actually be considered as one pointing to exceptionalism.

Alternative forms of dispute settlement: the case of party-to-party triggering
The second case study relates to an EU request for special treatment under the compliance mechanism of the Water Convention, a framework convention which requires its parties to prevent, control and reduce transboundary impact on transboundary surface waters and groundwaters. 105 The EU had become a party to the Convention in 1995. 106 To date, 25 EU MS are also parties to the Convention.
At its first meeting, the Convention's compliance body, the Implementation Committee, started discussing an EU-specific situation which had transpired elsewhere: the European Commission had submitted a legal note with regard to party-to-party submissions by EU MS to the compliance body under the Espoo Convention. 107 The records from the meeting of the Implementation Committee do not reveal the argument made in that note. 108 Yet, after liaising with the compliance body under the Espoo Convention, 109 the argument to be resolved by the Implementation Committee became clearer: 110 in its note, the European Commission had argued that EU law prevented EU MS from relying on State-to-State triggers. Thus, EU MS would be barred from initiating non-compliance procedures in respect of each other and in respect of the EU. What the EU thus effectively sought was to limit the scope of application of the compliance mechanism in respect of its MS.
Interestingly, the Implementation Committee decided to consider how EU law could be understood in this context. It traced the argument back to Article 344 TFEU, 111 which it understood to prevent EU MS from submitting a dispute concerning a treaty to which the EU and its MS are parties to external judicial control. However, the Implementation Committee rejected that this EU law requirement had any relevance for the respective compliance mechanism. It argued that the Convention, in its Article 22, would indeed provide for dispute settlement. Yet, the compliance mechanism was designed as an alternative to such dispute settlement. Unlike methods of dispute settlement, the compliance mechanism would serve the purpose of avoiding disputes rather than settling them. Therefore, and due to their non-binding nature, procedures before the Implementation Committee would 'not infringe on [EU] competence'. 112 As a result, special procedures and criteria for dealing with party-to-party submissions by EU MS were not considered further.
The relevant report of the Implementation Committee was presented at the 2019 MoP. 113 The records of this meeting indicate that the EU representative in the MoP did not object to the conclusion by the Implementation Committee. 114 The issue was not discussed further in the MoP.
This case is quite interesting for two reasons. First, the EU request for special treatment is motivated not only by an issue of EU law, but also by a provision of primary law which is considered to constitute an element of the EU legal order's autonomy. 115 The request may thus be an indication of at least the existence of autonomy concerns also in respect of compliance mechanisms from an EU point of view. 116 Second, the compliance body was prepared to consider the request and the underlying EU-specific characteristic on their merits. However, once the compliance body concluded that special treatment was not a necessary requirement from the viewpoint of international law, it did not discuss the matter further -conduct reminiscent of investment tribunals, which have also tended to reject the EU law argument against their jurisdiction. 117 Nevertheless, the compliance body's approach to the special request may suggest that the body would have been prepared to accord special treatment to the EU and its MS if it was necessary as a result of EU law. Even then, it is unclear whether such treatment could indeed be seen as special, and thus a case of exceptionalism. The answer would depend on how other parties are treated when they request an exception from the scope of application of a compliance mechanism based on their internal law.

Maintaining the EU judicial review system: the case of recommendations
The last case study discusses the earlier sketched EU request for special treatment in the context of the compliance mechanism under the Aarhus Convention. In 2005 the EU acceded to the treaty, 118 which requires its parties to grant the public certain procedural environmental rights: access to environmental information, participation in environmental decision-making and access to justice in environmental matters. 119 Ireland was the last of the EU MS to ratify the treaty in 2012.
In 2008 the eNGO ClientEarth made a submission to the Convention's compliance body, the Aarhus Convention Compliance Committee (ACCC). 120 The eNGO argued that the situation regarding judicial review at the EU level would not comply with the Convention's obligations for access to justice. In short, this situation would be created through the so-called Aarhus Regulation. 121 This legal act would only allow for judicial review of selected measures by EU institutions, and only by environmental organisations. Individuals would thus be limited in their right to judicial review. In addition, the Court's jurisprudence on legal standing for individuals but also organisations would make access to justice in environmental matters virtually impossible. 122 The ACCC decided to consider the communication in two parts; 123 part II related to the issue of access for the public to the EU courts. The ACCC continued proceedings in relation to part II of the communication in 2015 with a public hearing to discuss the substance of the communication. 124 In its subsequent draft findings it found the EU to be non-compliant with Article 9(3) Aarhus Convention, 125 which incorporates a right for the public to access to justice in environmental matters. In this context, the ACCC also drafted recommendations for how the EU should ensure observance of its obligations in the future. One of these recommendations was that the CJEU should take fuller account of these obligations when it interprets and assesses EU law.
The EU was invited to comment on these draft findings before their adoption by the Committee. 126 In its comments, the EU strongly opposed the recommendations drafted in relation to the Court: these recommendations would 'ignore, with all due respect, the specific features of a regional economic integration organisation like the EU, with its special institutional framework and unique legal order'. 127 The comments went on to highlight the Court's central role, forming an essential element of the autonomy of the EU legal order. Specifically, whether the Court takes into account the obligations under the Aarhus Convention was to be decided by the Court itself. An international agreement like the Aarhus Convention could not affect the allocation of powers as fixed by the Treaties. 128 Upon accession of the EU to the Aarhus Convention, no amendment of these Treaties had been demanded by international partners. 129 The EU comments then sought to highlight how the ACCC had failed in assessing the EU situation correctly. In particular, the ACCC had disregarded the evolving nature of the Court's case law and had only taken into account case law until a certain 'cut-off' date. Yet, it had not considered pending cases in its assessment, which could have further developed the Court's Aarhus jurisprudence, even though those had been pointed out by the EU representative. 130 Of those pending cases, the ACCC had specifically dismissed from its assessment cases resulting from preliminary reference procedures. Thereby, the ACCC had failed to grasp the specific nature of the EU and its system of judicial protection, of which the preliminary reference procedure is an essential part. 131 This argument was also central to demonstrating that the EU judicial system would ensure access to justice in environmental matters 'to a sufficient degree': 132 while natural or legal persons may not always have a right to challenge EU law of general applicability at EU level, they would certainly have the possibility to plead the invalidity of such an act in domestic procedures. The possibility and also obligation of domestic courts to call on the CJEU It was again concerns about its autonomy as one of its unique characteristics that prompted the EU requests for special treatment in this case. The requests, however, are made in respect of two different sets of issues: the request concerning the different way of confirming the compliance body's findings, 'taking note' instead of 'endorsing', may be considered more of a procedural issue. However, the request as to the deletion of the recommendation in the non-compliance decision effectively concerns the competences of the compliance mechanism in relation to the EU. The discussions in the political body were much more focused on that second request, which may indicate that it was more of a concern to treaty partners. Nevertheless, in approaching this request, treaty partners, again, 147 showed a willingness to consider the substance of the EU's request and explored whether the non-compliance decision would have the effects the EU was concerned about. Yet, considering the legal effects of non-compliance decisions under international law, EU international partners did not judge special treatment for the EU to be necessary. Rather, they emphasised how the EU had the same obligations, under the treaty and the compliance mechanism, as other treaty parties and thus had the same position in the treaty system. This emphasis on equality may even be an indication that the EU's international partners have begun to consider the EU's requests as being fuelled by an effort 'to have a freer hand' than others and are seeking to counteract this. 148

EU participation in compliance mechanisms: special or exceptional?
The above case studies provide the opportunity for some tentative conclusions as regards the EU's participation under compliance mechanisms. Under these mechanisms, the EU's requests for special treatment appear to be motivated by those unique characteristics which it also seeks to have accommodated in other international settings: its nature as an international organisation and its autonomy. In all of the instances discussed, treaty institutions and treaty partners showed a willingness to consider the respective characteristic. This willingness may point to a more general perception that the EU is just 'not a normal treaty actor', 149 and its participation is likely to require accommodating to some extent; or, it is simply a reflection of the facilitative nature of the compliance mechanism, which takes into account the situation, factually and legally, of a party. 150 Specifically in view of the EU's autonomy and its comprising elements, though, it is unclear whether external actors actually consider these as unique characteristics. The case under the Aarhus Convention may be indicative of external actors being inclined to understand autonomy concerns as simple issues of internal law and not being ready to accommodate them.
In the cases discussed, treaty institutions and treaty partners predominantly refused to grant the EU the requested treatment. While they considered the argument made, and even engaged with its EU law side, they decided on the request purely based on whether EU special treatment was necessary under international law. This focus on necessity may be considered a slightly different approach than at the stage of treaty-making where the EU receives special treatment by means of REIO clauses, yet, not only where this is necessary for it to join an agreement, but also to accommodate its, often political, practice of mixity. 151 If that difference in approach were accepted, it could be explained in light of the stages of treaty relations: at the stage of treaty-making, treaty partners define how they want to enter these relations. In contrast, compliance mechanisms relate to the implementation stage of a treaty and thus unfold against agreed-upon commitments. Accommodating a special situation at this stage, specifically if it were to limit the commitment a party undertook earlier, would be legally possible; yet, politically, it is clearly difficult to square also with the cooperative nature of a compliance mechanism and may thus warrant a stricter approach to the issue.
In the cases analysed, even where the EU received the treatment it requested, it is questionable whether, as a result, the EU was actually treated differently from other treaty partners. Indeed, the compliance body based its solution as to the interpretation of an emissions ceiling on necessity in view 147 See Section 3.2.2. 148 Nolte and Aust (n 8) 414. 149 Odermatt (n 6) 293. 150 See, e.g., art 15(2) Paris Agreement, which establishes the compliance mechanism and requires its body, quite broadly, to 'pay particular attention to the respective national capabilities and circumstances of Parties'. 151 Safrin (n 5) 1337.