I. Introduction
For most of the current era of European integration, the Court of Justice (CJEU) has not admitted preparatory work as evidence of the framers’ intent when interpreting Treaties. Even the teleology of ‘ever closer union’ is firmly rooted in the text of the Treaties in article 1 of the Treaty on European Union. This contrasts with a long trend of reviewing drafting histories of secondary legislation. Earlier literature has suggested that the reluctance to refer to Treaty preparatory work was linked to the secrecy surrounding early negotiations, and to poor public access to the documents. Some at the Court have considered that these concerns are no longer relevant especially to the more recent preparatory work produced by the constitutional conventions which were intended for public consumption. In several recent cases, the Court has reversed its initial opposition to relying on preparatory works to the Treaties. Recent literature has also been more open to the use of the travaux and accepts that such references can legitimise the Court’s reasoning.
We examine the extent to which the Court and the Advocates General have relied on preparatory work to interpret the EU Treaties and determine the ‘intent’ of their framers. Our focus is on the Treaty on European Union, the Treaty on the Functioning of the European Union, their predecessors and the Charter of Fundamental Rights. We then categorise the references and consider how they are used to develop key arguments in those cases. The references, whilst introducing a new source for legal interpretation, do not result in dynamic constitution-building. Rather, they reinforce static interpretations of existing provisions. The increased use of historical interpretation may in future even act as a counterweight to teleological interpretation. Nevertheless, extensive reference to travaux raises questions of transparency, democracy, and the overall manageability of the Union’s primary law that are difficult to answer satisfactorily.
This paper focuses on the constitutional dimension of the use of travaux préparatoires rather than the legislative dimension. Although the CJEU and the Advocates General have often relied on the travaux in their reasoning concerning the interpretation of the EU secondary legislation, this method is a novelty as regards the EU Treaties. The CJEU had not used the travaux préparatoires as a source of law to interpret the Treaties before 2005, when the EURATOM judgment was delivered. All of the cases in which travaux are used by the Court raise issues of conferral and are thus constitutional significance. First, we briefly consider how preparatory work is used as a source of law in international and EU law in order to illustrate the distinctive approach of EU law on this point. Next, we describe the cases in which the CJEU has relied on the travaux of the Treaties as a source of law. Then we build an analytical framework which draws on constitutional legal theory and international law. With this framework we analyse the possible outcomes to which the travaux could contribute and which types of outcome the Court’s judgments represent. We then extend this analysis to the larger data set which includes the Opinions of the Advocates General and compare those Opinions with the cases in which the CJEU has utilised the travaux. In our conclusions we also raise a number of issues which arise from the Court’s introduction of travaux as a source of law and raise several questions for further research.
II. The Legal Context of Preparatory Work in International and EU Law
Sources beyond the text of a legal instrument are controversial in many contexts. This is so even for preparatory work which is published. In the UK, reference to Parliamentary debates was taboo until 1993.Footnote 1 In other European countries, such as in Sweden and Finland, travaux are used routinely even when interpreting the constitution. In international law, preparatory works are routinely used to interpret Treaties, and are recognised as a legitimate source for this purpose.Footnote 2 However, travaux are often incomplete and misleading, and therefore less authentic than other elements. This is one reason why they are supplementary rather than principal means of interpretation under the Vienna Convention on the Law of Treaties (VCLT).Footnote 3 For similar reasons, The CJEU has traditionally not approved of references to the EU Treaties’ preparatory work.Footnote 4
A. International law permits reference to travaux
Reference to the preparatory works of Treaties is expressly foreseen in international law. The VCLT,Footnote 5 and particularly articles 31 and 32, govern the interpretation of international treaties. According to Article 31 VCLT,Footnote 6 the terms of the treaty should be given their ordinary meaning in context and be interpreted in the light of their object and purpose. The purpose of the treaty can be found from the text, the preamble, and the annexes. Article 32 VCLTFootnote 7 provides that the preparatory work of the treaty can be used as supplementary means of interpretation to confirm or determine the meaning when an interpretation according to Article 31 VCLT either leaves the meaning ambiguous or obscure or leads to a result which is manifestly absurd or unreasonable. In the VCLT, the travaux préparatoires are given a supporting role in the interpretation of the international treaties.
The actual object of interpretation is nevertheless the text of the treaties. The first step of the interpretation is to intuitively assume the ‘ordinary meaning’ of the text. Supplementary means of interpretation can then be used to confirm or revise the original hypothesis. The preparatory works can influence the interpretation through four routes: if the meaning of text is left ambiguous (Article 32.a); the interpretation would leave to manifestly absurd or unreasonable result (Article 32.b); the parties have intended to give a term a special meaning (Article 31.4); or merely to confirm the meaning resulting from the application of Article 31 (Article 32). Through this last mentioned means, the confirmation of textual and contextual interpretation, the travaux préparatoires can be relied in every case of interpretation.Footnote 8
B. The sui generis nature of EU law
The VCLT is treated differently in different contexts in EU law. The CJEU has found that the VCLT binds the EU institutions and is part of EU legal order as a rule of customary international law.Footnote 9 This line of case law indicates that the EU institutions need to respect the rules of interpretation stipulated in the VCLT when they are interpreting international treaties. However, the VCLT does not apply to the interpretation of the EU Treaties. The CJEU has found that the EU law is not in this respect ordinary international law. In Opinion 1/91 the Court contrasted the agreement establishing a European Economic Area (EEA) with Union law,Footnote 10 and stated that Union law forms its own legal order, whose aims are not just to ‘achieve economic integration’, but to make ‘concrete progress towards European unity’. Thus, Union law operates, according to the Court of Justice, in a different context, because it does not only create rights and obligations between the Contracting Parties. Instead it provides ‘transfer of sovereign rights to the intergovernmental institutions which it sets up’.Footnote 11
The Court’s interpretation of EU law is based on text, context, and telos (purpose), as can be read from the Van Gend en Loos judgment.Footnote 12 The CJEU has also stated that in interpreting EU law concepts one should use ‘the generally recognised principles of interpretation, beginning with the ordinary meaning to be attributed to those terms in their context and in the light of the objectives of the Treaty’.Footnote 13 Provisions of EU law should also be interpreted in their context and in the light of the EU law as a whole.Footnote 14 These elements of interpretation seem to be equivalent to those mentioned in the Article 31(1) VCLT, as explained above.
One significant difference has, however, prevailed for most of the EU’s constitutional history. Originally, reference to the preparatory work of the Treaties was considered constitutionally problematic in EU law. Article 32 of the VCLT envisages recourse to preparatory work of a treaty as a supplementary means of confirming an interpretation under Article 31 VCLT or to remove ambiguity of absurdity. This was ‘not a method which in the past commended itself to the Court’:Footnote 15 the preparatory work did not exist, or involved only working group-level discussions, therefore rendering reference constitutionally questionable.Footnote 16 Nevertheless Anthony Arnull, writing in 2006, expected the CJEU to become more amenable to using travaux as aids to interpretation: ‘Increasing pressure for transparency and the development of the internet have now brought many travaux préparatoires concerning subsequent amendments to the Treaties themselves into the public domain’.Footnote 17
C. Preparatory work and interpretation by the Court of Justice
There is not much literature on the status of travaux préparatoires to the Treaties before the CJEU.Footnote 18 The founding Member States of the European Economic Community, decided not to publish their travaux. Footnote 19 As a consequence, the CJEU also refrained from referring to them. In a classic exposition of this point from the mid-1970s, Advocate General Mayras noted in his Opinion in Reyners:
the States, signatories to the Treaty of Rome, have themselves excluded all recourse to the preparatory work and it is very doubtful whether the reservations and declarations, inconsistent as they are, which have been relied upon can be regarded as constituting true preparatory work. Nor can they be held against the new Members of the enlarged [Union] by virtue of Accession. Above all [the Court of Justice itself has] rejected, on several occasions, recourse to such a method of interpretation by asserting the content and finality of the provisions of the Treaty’.Footnote 20
Non-publication itself is no longer an issue: since 1994, the works have been available for consultation in the historical archives of the Union, housed at the European University Institute in Florence.
As a rule, the contextual and teleological interpretation methods are the chief methods for enhancing the CJEU’s understanding of Treaty texts – and it is the text of the Treaties, but not their preparatory work, which is used.Footnote 21 Teleology is in many cases derived from the Treaty text itself: when the text refers to ‘ever closer union’, an interpretation which facilitates this is only carrying out the literally stated intention of its drafters.Footnote 22 Ambiguity also invites teleology: the strong position of the contextual and teleological interpretation is at least sometimes a consequence of multilingualism in the EU.Footnote 23 Such ambiguity may also be an intentional outcome of the negotiation process.Footnote 24 Sometimes the contextual and teleological methods can even override the natural textual meaning of the treaty provisions.Footnote 25
Despite the lack of historical interpretation, there have been some false starts. Literature has referred to two cases in which the Court is argued to have used the travaux as a source of law in its historically-oriented teleological interpretation.Footnote 26 These cases are, however, not relevant to the question at hand. One of the cases simply does not refer to the travaux of the Treaties.Footnote 27 In the other case mentioned in the literature, the reference to the travaux is not made by the Court but instead by the applicant.Footnote 28 This is typical of the body of references to travaux: either the applicant or the defendant has relied on the alleged intention of the constitutional legislator that can be constructed from the preparatory works of the primary law, but the Court has not referred to the travaux in its reasoning.Footnote 29 In some further cases the Court used the travaux of the Acts of Accession, but not the founding Treaties.Footnote 30 Thus, none of those are in fact references where the Court uses preparatory work to interpret the TFEU, TEU, or their predecessors.
D. The ‘explanations’ to the Charter of Fundamental Rights as a special case
Some drafting documents have already been admitted into the Union’s sources of law doctrine, but as express references in the Treaties. The Charter of Fundamental Rights expressly refers to the ‘explanations’ as a source with which the Charter must be interpreted. Article 52(7) of the Charter states that ‘the explanations drawn up as a way of providing guidance in the interpretation of this Charter shall be given due regard by the courts of the Union and of the Member States’. Article 6(1) TEU stipulates that the Charter is to be interpreted ‘with due regard to the explanations referred to in the Charter’. The explanations therefore have a special status as a EU legal source amongst other travaux préparatoires in that primary law recognises their legal value, unlike the travaux préparatoires of the other Treaties.
The story of how the explanations came to be legally binding- and which explanations are legally binding- illustrates why drafting intent is so problematic. The Charter was of course based on an international instrument which already had its own set of explanations. Conventions of the Council of Europe, including the European Convention on Human Rights, always have an explanatory report.Footnote 31 Thus, it could be thought that explanations to the Convention could have served as a starting point for those provisions which were modelled on the Charter.
In the proceedings leading to the 2000 Charter, the explanations did not have legal status, and were ‘no more than a simple explanatory report approved by the Convention presidium’Footnote 32 since, just like typical working group texts, the text was not approved in the proceedings. However, during the Convention which led to the Draft Constitution, a phrase was inserted into the preamble of the CharterFootnote 33 and the text of the Constitution itselfFootnote 34 which recognised the 2000 explanations and their later version under the 2002–3 Convention. A separate declaration to the draft TreatyFootnote 35 was, according to Jean Paul Jacqué, aimed ‘to prevent the Explanations being attributed to the authors of the Treaty’.Footnote 36 It was, however, not reproduced in the Lisbon Treaty. Instead, Article II-112(7) of the Convention became Article 52(7) of the Charter, reaffirming the use of the Praesidium-drafted ‘explanations’ as an aid to interpreting the Charter. The later version of the explanationsFootnote 37 states that they had been updated to take account of redrafting during the Constitutional convention and developments in the other sources which the Charter rights reflect.Footnote 38
The ‘explanations’ are not, as might be expected, mere explanations. The explanations may in fact not only refer to the sources of rights, but define the right itself.Footnote 39 They provide a list of articles which have the same meaning and/or scope as equivalents in the ECHR. They can even be used to interpret the Convention in a way that is difficult to reconcile with its wording.Footnote 40 Explanatory works to the Charter of Fundamental Rights now frequently appear in the judgments of the CJEU,Footnote 41 and in the Opinions of the Advocates General.
III. Preparatory Work to the Treaties and the Changing Position of the Court of Justice
Although travaux have not traditionally featured in either leading cases or legal literature on the EU Treaties, this position is changing. The Court now refers to the travaux of the Treaties. The cases we examine reflect an entirely new line in the Court’s jurisprudence. The Court uses preparatory works to buttress arguments that are central to its reasoning.
A. Military uses of nuclear energy in EURATOM?
The first of this line of judgments was given in 2005 and concerned whether the military uses of nuclear energy could fall within the scope of the EURATOM Treaty.Footnote 42 The Court first noted that articles 1 EA and 2 EA indicate that the Treaty objective is essentially civil and commercial, but that since the treaty provisions did not expressly exclude the military purposes it is essential to interpret the issue taking into account other factors than the text itself.Footnote 43 Then the Court examines the historical background of the Treaty. The travaux préparatoires indicated clearly that the member states had differing views on the matter, and that therefore the Treaty could not be seen to be intended to cover military uses of nuclear energy.Footnote 44 In the absence of a reference to military use, the difference in view led to the conclusion that no such reference could be implied: ‘the Treaty is not applicable to uses of nuclear energy for military purposes’.Footnote 45 The reference to the travaux was not decisive factor in the argumentation; instead the traditional CJEU paradigm of contextual and teleological (purposive) interpretation gained more importance. The use of travaux was conservative and supplementary: it simply precluded implying an outcome that had clearly been the subject of disagreement during drafting: the travaux supported an exhaustive reading of the competence list.
B. ‘A sound budgetary policy’
In the Pringle case, the CJEU referred to the travaux of the Maastricht Treaty, when it evaluated whether the ESM Treaty was in breach of the ‘no bail-out clause’ in Article 125 TFEU.Footnote 46 Using a textual method to interpret the wording used in Article 125 TFEU, the Court found that the Article is not intended to prohibit all financial assistance to another Member State. Next the Court used a contextual method. The economic policy context of Article 125, in particular articles 122 TFEU and 123 TFEU, led the Court to the same conclusion. Only then did the Court examine the objective of Article 125 TFEU. This was found in the preparatory works of the Maastricht Treaty: ‘it is apparent … that the aim of Article 125 TFEU is to ensure that the Member States follow a sound budgetary policy’. Given this objective, it found the proposed European Stability Mechanism could be compatible with that article if the conditions ‘to such assistance are such as to prompt that Member State to implement a sound budgetary policy.’Footnote 47 Since ‘the ESM and the Member States who participate in it are not liable for the commitments of a Member State which receives stability support and nor do they assume those commitments, within the meaning of Article 125 TFEU’ it followed ‘that Article 125 TFEU does not preclude either the conclusion by the Member States whose currency is the euro of an agreement such as the ESM Treaty or their ratification of it’.Footnote 48 In this case the travaux préparatoires had a decisive role in determining the question at hand, since the preparatory work was used to justify the legality of the ESM mechanism despite the no-bailout provision. Similar reasoning appears in the OMT reference from the Bundesverfassungsgericht, where the Court refers to Pringle in its definition of monetary policy and the travaux to the Maastricht Treaty for the aims of the Article 123(1) TFEU prohibition on direct credit facilities and direct purchases of Member State debt.Footnote 49 The aim discovered in the preparatory work was, as in Pringle, ‘sound budgetary policy’.Footnote 50
C. Judicial review of ‘regulatory acts’
The travaux to the present Treaties have been examined in the Inuit Tapiriit Kanatami appeal.Footnote 51 Here, they were cited by the Court of Justice in its review of what was intended by the revision of Article 263 TFEU as regards ‘regulatory acts’ subject to review without individual concern. In Inuit, the CJEU makes a statement of principle on their use: the interpretation of EU law requires not only taking into account the wording and the objectives of EU law provisions, but also their context within EU law as a whole. It expressly approves reference to preparatory works: the origins of a provision ‘may also provide information relevant to its interpretation’.Footnote 52 The Court states that the authors of the Lisbon Treaty did not have intention to alter ‘the scope of the conditions of admissibility already laid down in the fourth paragraph of Article 230 EC … it is clear from the travaux préparatoires relating to Article III-365(4) of the proposed treaty establishing a Constitution for Europe that the scope of those conditions was not to be altered’.Footnote 53 Thus, regardless of the traditional view that travaux to EU Treaties are not relevant to their interpretation,Footnote 54 it is clear they are relevant and should be reviewed. As Advocate General Kokott explained in her Opinion in Inuit, ‘the practice of using conventions to prepare Treaty amendments’ and ‘of publishing the mandates of intergovernmental conferences’ entitle the Court to make greater use of the Treaties’ preparatory works: greater access to the proceedings encourages use of the travaux ’as a supplementary means of interpretation if, as in the present case, the meaning of a provision is still unclear having regard to its wording, the regulatory context and the objectives pursued’.Footnote 55
IV. An Analytical Framework: Features of Constitutional Reasoning with Travaux
Accepting a new source of law for constitutional interpretation begs the question of what kind of impact such a change might have. Constitutional interpretation can have different outcomes depending on whether the constitution (here EU primary law) is seen as a finished product that can be changed only through amendments (as in skyscraper originalism) or as a framework for governance that can be complemented to fit to the existing circumstances over time where judiciaries can take part in the constitutional construction (framework originalism).Footnote 56 This relates to the kind of constraints that there are in constitutional interpretation, and what kind of role the text, the original meaning and the intent should have as such constraints.Footnote 57 This, in turn, relates to the question of whether the constitutional interpretation simply ascertains in a static manner that certain line of interpretation corresponds what has been intended, or whether the constitutional interpretation results in the dynamic construction of the constitution by interpretation – filling in gaps and amending problematic provisions through interpretation.Footnote 58
Constitutional interpretation can have static or dynamic effects. Static references to the travaux focus on the historical process of the constitution creation and the original purpose that can be found from the travaux. Static references to the travaux uphold the original constitutional idea behind the constitutional provisions. These references can either simply ascertain an interpretation that can be formed by means of textual interpretation (static/ascertaining), or they can also be used to determine a meaning when the text alone leads to an ambiguous interpretation as to the original purpose of the provision is (static/extra value). Dynamic references to the travaux can fill gaps or develop the constitution in other ways in order to justify constitutional change. Dynamic references could be used to justify a certain interpretation that pertains to new circumstances that cannot have been seen in advance. Table 1 illustrates the functions that the travaux can have in a constitutional interpretative framework.
Table 1 Functions of travaux in a constitutional interpretative framework
In the CJEU judgments examined here, the Court’s references support a static interpretation: the constitutional framework is reinforced, but not changed. Thus, the Court’s references to travaux do not engage in teleology beyond that already found in the relevant treaty provisions. However, in each of these cases, the references bring extra value to the textual interpretation. The references are not made simply to ascertain a fact that could be read from the constitutional text itself. This means that the references the Court makes belong to group 2 (static/extra value; Table 1).
The next section examines cases in which the Advocates General have relied on the travaux to the Treaties. After describing the Opinions we apply the above framework to them in order to explain when the Court might choose not to rely on the travaux in its argumentation.
V. Travaux References in Advocate General Opinions: Identifying Emerging Trends?
Before the Court of Justice itself recognised travaux, the preparatory documents of the Convention and subsequent IGCs had been cited in a dozen or so documents. In the immediate aftermath of the Convention, Advocates General attempted to introduce travaux as material for constitutional interpretation.Footnote 59 After a false start hobbled by the failure of the Constitutional Treaty, references begin anew after the entry into force of the Lisbon Treaty. Here, as above, the majority of references tend to support a line of reasoning that does not require the documents. There are other arguments in favour of the outcome, or the interpretation which is used. Travaux are not decisive.
In one category of cases, the references are entirely trivial. In another, they explain changes to the Treaties, and the purpose of those Treaty changes, but still remain static in nature. But arguably some of these references may employ a more dynamic approach. In one case, the Advocate General added a category of instrument to an international competence of the Union that is recast in the Treaty. In this case an argument could perhaps be made that the legal rule is derived not primarily from a textual interpretation of the Treaties but comes in large part as a consequence of a creative and even teleological interpretation of the new text – a dynamic constitutional construction found exclusively through a reference to the travaux.
A. Some references are trivial
First, the mundane. One class of Opinions refers to Working Group documents, but in arguably only the most trivial and superficial ways which do not genuinely contribute to the legal outcome in the Opinion. In the Biocides case,Footnote 60 Advocate General Villalon referred to a Convention Working Group documentFootnote 61 and claimed that the drafters had in mind comparisons of national administrative legal systems when discussing what ‘delegated’ legislation, now in Article 291 TFEU, should mean.Footnote 62 They surely did so, but the document to which he refers makes no mention of this comparison.
In his Opinion on Enhanced Cooperation for the Unitary Patent,Footnote 63 Advocate General Bot referred to a Praesidium note (so not a final working group report).Footnote 64 He used this to support an intention to clarify competences in the Treaty revision: ‘At the European Councils of Nice, in 2000, and Laeken, in 2001, the Member States clearly expressed their desire that the sharing of competence between the Union and the Member States be clarified’.Footnote 65 That the Convention was framed by the Laeken declaration objectives is hardly a legal innovation, and does not contribute to the line of reasoning involving the constitutionality of the proposed system.
In Commission v Germany,Footnote 66 it had been claimed that the Commission had not pursued its case against the Member State efficiently, and that this precluded an action under Article 260 TFEU. In accepting the established position that infringement proceedings do not have to be brought within a particular time, but involve a large amount of discretion, Advocate General Wahl mulled evidence to the contrary. Travaux were referred to in this context – as evidence considered, but not sufficient to persuade the Advocate General: AG Wahl acknowledged that the purpose of reforming the infringement systems was inspired also by a need for speedy action also in infringement proceedings brought by the Commission.Footnote 67 He did not consider this sufficient to reverse the established discretion of the Commission in bringing proceedings.
B. The majority support, but do not found, legal propositions
A large bulk of the references support legal propositions that are not wholly founded on the preparatory works but which are clearly augmented by them. This is the case for all of our highlighted CJEU judgments at earlier stages, whether at first instance (Inuit) Footnote 68 or the Opinions of the Advocates General before the CJEU (EURATO M,Footnote 69 Inuit Footnote 70 Pringle Footnote 71 and Gauweiler Footnote 72 ). The Advocates General tend to draw conservative conclusions, for example to note that since the preparatory works did not show an intent to develop EU competence or the Court’s jurisdiction, an extension could not be implied.
In Telefonica,Footnote 73 Advocate General Kokott attempted to interpret what the Treaty annulment grounds refer to when Article 263 TFEU refers to an act which does not entail implementing measures.Footnote 74 Her references to the legislative history of that provision, namely the Court of Justice discussion circle in the Convention,Footnote 75 are used to support two points. First, she observed that a German version of the Treaty provision translates as similar to the current French text.Footnote 76
A second reference to the same document is then used to determine the ‘intention’ behind the words ‘implementing measures’ in that same revision: why must regulatory measures that can be challenged without individual concern, as ordinary legislative acts, be ones which do not entail implementing measures? The text of the preparatory document provides an express claim: ‘The addition of ‘implementing measures’ was intended to ensure that the extension of the right to institute proceedings was restricted to cases where an individual “must first infringe the law before he can have access to a court”.’Footnote 77
In the ESMA Short Selling Rules Opinion,Footnote 78 Advocate General Jääskinen also referred to a Convention Final Working Group document.Footnote 79 In this case, the reference simply refers to the legislative history and identifies that the distinction came from the Working Group. It does not seek to claim a specific interpretation based on that document. Subsequent references to literature reflect on points which are also covered in the Working group document, but which is not referred to for additional support.
In the Broadcasting Rights Convention Case, Footnote 80 Advocate General Sharpston reviewed both a Working Group document as well as a mandate for an intergovernmental conference that eventually led to the Lisbon Treaty in order to examine whether the new exclusivity clause in Article 3(2) TFEU intended to depart from the ERTA test for exclusivity from which its language is borrowed.Footnote 81 ‘If the negotiating history of Article 3(2) TFEU shows anything, it is that there was no intention to depart from the ERTA principle.’Footnote 82 Likewise, in his View on Opinion 1/13 (on the Hague Convention on Civil Aspects of International Child Abduction), Advocate General Jääskinen noted that Article 3(2) TFEU was intended to reflect the case law of the Court of Justice concerning when an external competence was also exclusive.Footnote 83
In the Anton Las Opinion,Footnote 84 Advocate General Jääskinen referred to the origins of the current ‘national identity’ provision in Article 4(2) TEU. A European Convention working group had recommended that the provisions now in TEU on national identity should have been clarified to expressly state that the EU is obliged to respect ‘essential elements of the national identity [which] include… their choices as to language’.Footnote 85 Jääskinen’s point is simply that, since the clarification was mooted, ‘[t]he concept of ‘national identity’ therefore concerns the choices made as to the languages used at national or regional level.’ This clarification was not made – language is not currently mentioned in Article 4(2) TEU – but the Court recognised it as part of the concept in its Runevic-Vardun and Wardyn judgment.Footnote 86 However, had the Court followed the Advocate General, it would have not led to an outcome that favoured the language choices of the state: Jääskinen instead compared his enhanced concept of national identity with the concept of linguistic diversity, with the effect that Jääskinen would have denied the possibility of derogating from fundamental freedoms on these (national identity) grounds.
Finally, In her View to the Court’s Opinion 2/13 (Draft ECHR Accession Agreement), Advocate General Kokott examined the intent of the Constitutional Convention in framing what eventually became the exclusion of the Court’s Common Foreign and Security Policy jurisdiction in Article 24(1) TEU and 275 TFEU.Footnote 87 The Convention had considered extending the Court’s jurisdiction, but declined to do so.Footnote 88 For Advocate General Kokott, this meant the draft proposal’s jurisdiction provisions extended the Court’s jurisdiction contrary to the intent of the framers.
C. Could travaux introduce new propositions of law?
So far, Advocates General have not used travaux in a particularly controversial way. They are trivial or support propositions of law but are not solely responsible for founding them. They tend to support conservative readings of pre-existing provisions. Many show that a change was mooted, but not adopted – and that therefore, this change could not be implied by the Court’s interpretation. Thus, if anything, travaux references can limit the scope for dynamic interpretation. However, under one reading, the recent Venezuelan Fisheries case could be considered quite unusual – and one where the travaux would have clearly augmented the Treaty text.
In Venezuelan Fisheries, Advocate General Sharpston examined inter alia the effects of the EU’s new legal personality on the legal rules which governed its international action.Footnote 89 It was suggested that the Union could not make unilateral declarations, but could only enter into bilateral ‘international agreements’ under Article 218 TFEU. According to Sharpston’s reading of the Convention Working Group final report on legal personality,Footnote 90 the reasons for attributing legal personality entitled the Union to also issue unilateral declarations which bound it under international law. By making the Union a subject, it would:
… be able to avail itself of all means of international action (right to conclude treaties, right of legation, right to submit claims or to act before an international court or judge, right to become a member of an international organisation or become a party to international conventions, eg the ECHR, right to enjoy immunities), as well as to bind the Union internationally.Footnote 91
Thus, although the Treaties did not expressly envisage unilateral declarations, here Advocate General Sharpston would have implied the power in part through a construction based on the Working Group final report’s intention to grant the Union full capacity as an international actor. This could be seen as a dynamic reference, since the immediate outcome is to augment the categories of legal instruments with declarations in international law.
VI. Applying the Framework to Advocate General Opinions
The Opinions can be also be located in the same analytical grid in order to analyse in which circumstances the CJEU has also referred to the travaux in its reasoning. References to the travaux in the Advocate General Opinions can be categorised in following way:
We can see that the Court refers to the travaux only in cases in which such references have brought some added value to the Advocate General’s argumentation. The references help determine the meaning of the Treaty provisions in question where other interpretative tools have not provided sufficient clarity. The travaux references are used to determine the objective and purpose of the Treaty provisions in question. Viewed from this angle, it is no surprise that the Court’s judgments which utilise the travaux have not led to dynamic interpretations at least in the sense that the interpretation would change the textual meaning of the Treaty provisions. However, over time, even textual meanings of different concepts can change to respond the new social circumstances.Footnote 92
VII. On the Use and Abuse of Travaux
Even though the older preparatory works of the Treaties might not have been widely reported or published, the travaux préparatoires of the more recent Treaty amendments, such as the Constitutional Treaty and the IGC leading to the Lisbon Treaty, are rather well documented. Nevertheless, use of the travaux to Treaties often raises at least three major questions of principle.
The first is whether it is appropriate to refer to preparatory work which has not been published as evidence of drafting intent. Taken to extremes, this raises the prospect of preparatory work which significantly alters the literal meaning of texts being revealed when it suits the institutions. The explanations to the Charter arguably sidestep this issue since the text was published and the Treaties expressly refer to their use. Pre-Convention era preparatory work is more problematic: at the far end of the spectrum, international agreements are often negotiated under more secretive diplomatic arrangements. Regulation 1049/2001 on access to documents, for example, refers to the protection of the Union’s public interest in international relations and has been interpreted as protecting various preparatory documents leading to international agreements.Footnote 93 Even if public interest has led to the publication of such documents in some recent cases, this is not yet the norm for international agreements to which the Union is party.Footnote 94
Use of the travaux is also linked the question of democratic participation in such rulemaking. To what extent can it be argued that preparatory documents have the same democratic credentials as the final text of the Treaty? The case of the explanations to the Charter illustrate this difficulty well: a text that had not been discussed and the ramifications of which were at best unclear became an interpretative guide with significant legal consequences as illustrated by of its extensive use by the Court. This objection will typically apply to early preparatory work but will also apply to international negotiations where the preparatory work is protected from public scrutiny by the Union’s ‘access to documents’ policy.
Third, if the aim is to clarify or simplify Treaty provisions, the full release of extensive preparatory work may be a cure worse than the disease. It would mean that anyone seeking to understand a Treaty provision must become acquainted with not only its text and the general canons of interpretation but also the full extent of the preparatory documentation. The proper publication of the travaux gives the CJEU a better mandate to use them as a source of law, as the Pringle and Inuit cases demonstrate, but may make the cognitive task of understanding them difficult bordering on the impossible.
A. Current practice is conservative despite admitting travaux
The cases above show that preparatory works can now be used, but whilst Pringle in particular has raised vocal dissent, this specific aspect – that is, the use of preparatory works to interpret the meaning of a Treaty provision – is less controversial in each. The EURATOM interpretation case shows one example of how travaux could be used to support cautious or conservative readings of implied powers. There, discussion in the preparatory works showed that military use was not omitted by accident: the possibility of including this in the EURATOM Treaty was rejected. That rejection was then confirmed when the Court was pressed to rule on this point. But it is interesting to note that whilst the outcome is conservative, the use of these particular travaux overrides many of the historical objections to references to preparatory work: for example, these date from the time period when Advocate General Mayras considered it the intention of the drafters not to refer to them. They also preclude a route which, had the drafters commenced their work today rather than in the distant past of European integration, they may have wished to take. Thus, the reference locks in a restriction and prevents the dynamic future interpretation of the Union’s constitutional documents, providing greater pressure for the express revision of the Treaties.
In Pringle, it is hardly contested that a key purpose of the no bail-out clause was indeed to ensure prudent macroeconomic governance. Whether that provision should therefore also preclude the ESM is not decisively determined by the travaux, but by the court’s appreciation of the effects of the ESM on this prudence. In Inuit, the travaux reference is also limited to confirming the relatively unopposed interpretation of Treaty change: the preparatory works in question clearly illustrate the origins of the distinction between regulatory and legislative acts – one also mooted by the Court’s own discussion group at the time of the Constitutional Convention. This is then confirmed by the Court’s interpretation of the provision.Footnote 95
B. Reference to the intentions of drafters is open to abuse
Travaux can be used in a conservative way, but their acceptance may equally lead to a dynamic and perhaps even unpredictable approach to interpretation. Koen Lenaerts, the president of the Court which handed down the appeal judgment in Inuit, suggested the travaux will be increasingly influential as the current Treaty provisions are interpreted.Footnote 96 However, some doubts might be raised about the utility of travaux references. Many of the criticisms that have been levelled against the use of preparatory work to secondary legislation apply also to the Treaties’ preparatory work, and in some cases even more strongly.
Soren Schønberg and Karin Frick’s 2003 review of travaux in secondary legislationFootnote 97 noted that the conflicting intentions and statements of co-legislators made it difficult to identify a single coherent intent;Footnote 98 that a historical intention could ossify interpretation (consider: what could the founders possibly have said about interstate electronic commerce?); and finally, that the drafting of the documents themselves could be so ambiguous as to preclude drawing conclusions as to what was intended.
All of these concerns also hold true with respect to travaux. The question of whose intent matters is especially acute in the context of the Treaties. The input ranging from the Convention working groups, its plenaries, those involved in the legal review of the draft treaties, the intergovernmental conferences leading up to the reconfiguration of the texts in the Lisbon Treaty all offer reasons for questioning claims as to intention. Negotiators involved in the proceedings may, if interviewed, give diametrically opposed views of what was intended by particular Treaty provisions. National parliamentarians approving changes may hold particular views on what is intended. Significant differences of opinion are evident at various stages of the legislative processes.Footnote 99 And what weight should be given to the way in which Treaty changes are marketed in the national referenda? Can we – and should we – also measure what is intended by those readers of the drafts, without whom many changes cannot be ratified? The European Union, like all legal systems, has its share of legal fictions. But is it too bold to claim that an intention can be identified in preparatory works? And do only written, contemporaneous documents of particular bodies matter, or should the entirety of the evidence be considered?
The clarity of preparatory works may also leave much to be desired, even in the post-Lisbon era. In our primary field of research, a major topic of inquiry is the extent and limits of legislative powers in the field of criminal law.Footnote 100 Has the Union exhausted its competence for criminal legislation in Article 83 TFEU, or might similar provisions be founded on other legal bases?Footnote 101 This appeared to be a major discussion point during the negotiations. Nevertheless, the whole body of evidence led the UK House of Lords to conclude at the time it both required an answer so as to ensure the effectiveness of its ‘opt-in’ and that it did not know the answer.Footnote 102
The mere possibility of referring to the travaux also complicates this issue. The EU Treaties in their current form are hardly a glowing example of clarity, either in length or in style. The possibility that these ambiguities might now require clarification in the form of exponentially greater, and less clear, combinations of preparatory works should be additional cause for concern. When will they be instrumental to the outcome, and when will they be ignored? It is tempting to echo Gareth Davies’s very recent conclusion on Court’s approach to legislative change: ‘in situations where the Treaty is involved, which are almost all the ones that matter, the legislature has no capacity to force the law in a certain direction’ since the Court’s interpretative powers are so vast.Footnote 103
C. Will the use of travaux slow the integration process?
At the other end of the spectrum, it should be considered whether the preparatory work to the Treaties will reveal conservative tendencies that then become, as the declared intent behind the revisions, susceptible to change only through Treaty revision. During the last Treaty revisions, a clear intent behind the grant of IGC mandates themselves was to clarify and delimit the competences of the Union. If taken seriously, this could mean that teleological reasoning, particularly in its dynamic and forward-looking aspects, will receive a counterbalance in the form of conservative historical interpretation supported by the travaux. Historical interpretations which rely on these travaux can reinforce the constitutional nature of the Treaties. This happens in cases like EURATOM when preparatory work shows a particular outcome had been mooted but was not accepted. Whilst it is not desirable that this new line of interpretation would halt the constitutional development of the Union, the use of the travaux as an interpretative tool in the cases reviewed above can also lead to the opposite outcome. In Venezuelan Fisheries the justification and reasoning behind Treaty change is not invoked by the Court, but the reasoning is adopted and leads to a dynamic outcome as suggested by Advocate General Sharpston.Footnote 104
One interesting question in this respect is how far into the historical travaux the Court will be prepared to venture. If, as in Inuit, it refers to the ratio legis of Treaty reform in the immediate aftermath of that reform, this runs little risk of ossifying a historical but anachronistic approach. However, if the travaux to the 1950s Treaty drafts were to be reviewed decades later, as in EURATOM, the outcome may not correspond to present or future needs which, whilst controversial at that time, could be more easily justified at the time the interpretation is sought. In EURATOM, it can probably not be argued that the Union is ready to harmonise rules on the military uses of atomic energy without Treaty revision. But what, for example, should we think about the continued use of the 1950s public policy exceptions in the free movement context, recognised since the 1970s to be an incomplete list in form and substance? It can hardly be argued that the failure to reform the Treaty derogations to fundamental freedoms demonstrates an acceptance that the environment can never be protected in this way.
VIII. Conclusions
Some have argued that the travaux could contribute further to developing teleological arguments.Footnote 105 Some see that recourse to the intent of the drafters adds legitimacy to the Court’s argumentation.Footnote 106 Thus far the Court itself has used the travaux préparatoires in a static manner. It determines the original intention of the constitutional legislator in situations in which the text itself would not give satisfactory solution. This differs from the Court’s general line of teleological jurisprudence, which is highly dynamic. Although the readings which it has employed are conservative, it has seen fit to refer not only to the most recent Treaty revision, but to travaux of the Maastricht and even the EURATOM Treaties.
The Court appears reluctant to refer to travaux when they provide no added value. When the travaux confirm evidence also found elsewhere, they are omitted by the Court of Justice. The Court’s travaux references also, so far, lead to conservative outcomes. In the cases in which the Court does follow Advocates General in referring to travaux, both the Advocate General and the Court use travaux in a static way. A particular Treaty interpretation is supported – the text is clarified and this provides added value – but the reference does not propose a construction that is developed exclusively by interpretation.
We conclude that thus far the Court has used the travaux in a static manner. The Court has not exercised dynamic constitution building based on references to the travaux. This does not mean, however, that the Court’s new line of interpretation using the travaux of the Treaties could not be used in a dynamic manner in the future. Further research is required on when this might be appropriate, and when it is appropriate to refer to a historical intent when providing a contemporary interpretation of the Treaties.
