2.1EU legal order

EU terminology falls to be seen and understood within a picture that is larger than just the EU context. Doing so helps one understand the pressures that operate on EU terminology. EU law gives the appearance of being like a legal system, with borders defined by the borders of its Member States, with an inside and an outside and a scope of application. Yet there is a hidden question here: is EU law really a system like other systems of law? Is it like the national (domestic) law of a country, or is it something different? The CJEU referred to it as a legal order in its judgment of 5 February 1963 in Case 26–62 (van Gend & Loos). What is the difference? The question and implications of the difference between a legal order and a legal system can be debated at length, but briefly the meaning intended here is that: “Legal orders are collections of norms, be it the law of nation-states, supranational entities or international law” (Magen 2015, 24). The EU is thus one order among three, with the others here being the law of nation-states and international law. This is important for meanings of terms as each context is different. On the other hand, the concept of legal system is generally applied to the laws of nation-states, and this simple definition is sufficient for the immediate purpose.11.For a detailed analysis of legal systems, see Raz (1980, 2011). The idea of a legal system is that it is a set of rules within an internal geographical space which has boundaries and frontiers. These frontiers are legal, because different laws apply on each side, as when one crosses a border from say Luxembourg to Belgium. So, by the same analogy we have an internal EU space where EU laws apply, an external space where other laws apply, and the EU frontier is the border between the Member States and non-Member States.22.For a definition of EU external border and terminology equivalents in all EU languages, see for example the EU Commission definition at https://​ec​.europa​.eu​/home​-affairs​/what​-we​-do​/networks​/european​_migration​_network​/glossary​_search​/external​-eu​-border​_en. However, unlike with a legal system, we cannot simply say that EU law does not apply outside the EU internal space, because it is possible that it might apply, albeit indirectly. An example of this situation concerns the case of former EFTA (European Free Trade Agreement) states which contracted with the EU Member States by the Agreement on the European Economic Area (EEA) of 17 March 199333.EU OJ J No L 1, 3.1.1994, p. 3; EFTA States’ official gazettes.at https://​www​.efta​.int​/media​/documents​/legal​-texts​/eea​/the​-eea​-agreement​/Main%20Text%20of%20the%20Agreement​/EEAagreement​.pdf. to incorporate a range of EU legal texts into their own internal legal systems. EU law, language and terminology have been extended to non-EU Member States, and to additional languages, notably Icelandic and Norwegian. So, although not formally part of the EU these languages nonetheless have had to develop EU terminology.

Interestingly, we can obtain a simplified view into core functions of EU law by looking at the EEA Agreement. Article 1(1) refers to promoting: “a continuous and balanced strengthening of trade and economic relations……with equal conditions of competition, and the respect of the same rules, with a view to creating a homogeneous European Economic Area.” These objectives entail: the free movement of goods, the free movement of persons, the free movement of services, the free movement of capital, the setting up of a system to ensure that competition is not distorted and that the rules are equally respected, as well as closer cooperation in other fields, such as research and development, the environment, education, and social policy. (Article 1(2)). EU law has been extended to embrace non-EU Member States, and the implication of this is that it does not fit neatly into the concept of system; instead, the concept of order, as used by the CJEU seems more appropriate.

2.2EU agreements

We can take this a stage further, insofar as each time the EU makes an agreement with a non-EU state it carries with it elements of EU law, language and terminology. Each agreement is written in the EU languages,44.There are derogations for Irish and Maltese. as well as the language of the other parties, and that implies the establishment of terminology for all the language versions of the text of the agreement. However, here we arrive at a situation of hybridity between EU conceptualizations and those of the other contracting party/ies. With EU trade agreements, for example, each agreement contains the elements negotiated and agreed between the parties, inevitably involving compromises, including on language and terminology if the contract is to be successful. These arise, in particular with respect to languages shared between EU and non-EU contracting parties, for example regarding South American Spanish or American English. A special category of agreement with a third country is a treaty for EU accession. Here, not only is the Treaty written in the accession-country language, if it is a new language, but the existing body of EU legal texts, the EU acquis, must be translated in advance to be ready for the first day of accession. The role for terminologists is enormous.55.On the accession process, see European Commission: EU Accession process Step by Step, online at https://​ec​.europa​.eu​/neighbourhood​-enlargement​/sites​/near​/files​/eu​-accession​-process​_en​.pdf.

What one sees here is that EU terminology is extended to non-EU Member States through contractual relations in the form of agreements. These agreements are made under international law, and it is important to draw distinctions between supranational law (EU law), international law, and domestic national legal systems such as Belgian, French or German law. Each order or system represents a legal, and linguistic, context for meanings. Information is constantly crossing between them. Concepts, words, and terms arise in one context and transfer to the others, in all directions and through a diversity of languages. There is interlinking, and the concept of a network or web comes to mind, similar to the world wide web. Terminologists are at the heart of this networking. To which we can add that it is multilingual, as all languages are potentially involved. International agreements can be made in whichever language is selected by negotiating parties; they just need to have access to the appropriate terminology in the languages chosen, in a written form. So, it is here that we perhaps come to the front line of terminology work: the constant development of new texts within the EU legal order and new agreements with third countries. Each time there is a need to settle the terminology in the drafting language and the other languages involved. This is the background for EU legal terminology work.

2.3EU legal order as context for terms

The EU legal order is founded on a series of agreements in the form of treaties between a number of European States. The history of these is well known, as are the reasons which led to their negotiation, signature and ratification.66.See History of the EU at: https://​europa​.eu​/european​-union​/about​-eu​/history​_en. The texts are products of the language, conceptualizations, methods and terminology of international law prevailing at the time of each treaty. International law provides a constant reference point and source of ideas, inspiration, concepts, and terms which pass over and into the EU legal order by a process of osmosis. The EU treaties can be seen as applications of international law, adapted to particular circumstances and objectives.77.On the relationship between international law and EU law, see Ziegler 2013. The method involves creating an internal space between signatory states where particular rules and methods apply between and among themselves. The EU treaties create this internal space and set out the framework for its organization and management. The process has been dynamic, starting from coal and steel with the European Coal and Steel Community (ECSC), expanding to atomic energy with the European Atomic Energy Community (EAEC, Euratom) and to other economic areas with the European Economic Community (EEC). With each treaty the terminology range increased as a consequence. The current foundation treaties, the Treaty on European Union (TEU) and the Treaty on the Functioning of the European Union (TFEU) consolidate, replace and extend the previous treaty texts in relation to the EU.88.For a list of treaties and information on them, see Learn Europe: http://​www​.learneurope​.eu​/index​.php​?cID​=310. These texts are currently the starting point for accessing EU legal language, together with ancillary treaties, in particular the individual accession treaties.99.For details of each accession treaty, see EUR-Lex at: https://​eur​-lex​.europa​.eu​/collection​/eu​-law​/treaties​/treaties​-accession​.html.

The structure of the EU legal order is important for terminology as law and legal texts are hierarchical. Terms and meanings are organized though interlinking texts to create a web of rules and relationships between texts. Law-making aims to embrace all aspects of life in society, to answer the basic question: what is the law, or rule, on such and such a matter? EU law is no exception; it is part of this web, and it addresses the matters that are set out in the treaties (primary law) as implemented and extended by delegated, or secondary-level, acts and instruments (secondary law), as well as other non-binding acts (soft law), all as mediated and interpreted by the courts of law, in particular the CJEU. This has implications for meanings and for terminology. For example, higher-level texts contain terms that are operative for lower-level texts. We enter into legal theory, but the main point is that lower-level texts are generally not free to change meanings set in higher-level texts. The terminology of the treaty texts takes precedence over that of secondary-law texts. Further, where a secondary-level act has a further implementing or delegating act, then that lower-level act may not depart from the meanings set at the higher level.1010.On delegated acts, see for example: https://​eur​-lex​.europa​.eu​/summary​/glossary​/delegated​_acts​.html. It is a matter of power relations (Forst 2018), and this is evident from the use of verbs of obligation such as must, shall, may and may not in the articles (enacting provisions), mirrored in the recitals by words such as should to mark a difference in role between these two segments of a text. To which we can add the various forms of sanction or penalty if the obligations are not complied with.

The problem, however, is that higher-level texts are generally broader and more generic in nature and expression, aiming at outlines and overarching structures. One can see this in the TEU and TFEU which are largely framework treaties. For example, Article 11 TFEU states that: “Environmental protection requirements must be integrated into the definition and implementation of the Union’s policies and activities, in particular with a view to promoting sustainable development.”.

The details are worked out in implementing texts which are technical and detailed, as can be seen from a perusal of EU Regulations and Directives in EUR-Lex.1111. https://​eur​-lex​.europa​.eu. On Article 11, see for example: Nowag (2018). The art in drafting is to embrace the higher-level-text terminology in ways that do not seek to change its meaning overtly; it is more a case of developing and expanding on the meanings in particular applications and contexts, and this in all EU official languages. This is the dynamic side of EU legal terminology as each text is a tool for doing things, and through time and experience ideas change and develop. The broad framework approach in the treaties allows this to take place most of the time through amending the secondary-level texts.

In practice, when thinking about the EU legal order, one is addressing legal texts and their relationships, and that brings one to the concept of intertextuality (Robertson 2012a). It requires skill on the part of legislative drafters, and also legal interpreters, as they seek to create a mental image of the conceptual structuring of a particular area and to think about how the pieces fit together as a jigsaw within the wider overall picture of EU law, EU policies and the teleological idea of moving in directions set out in the foundation treaties, in particular in the recitals. Thus, in the TFEU treaty we find the recital: “AFFIRMING as the essential objective of their efforts the constant improvements of the living and working conditions of their peoples”. Part of this dynamic process involves identifying new situations, finding new solutions and encapsulating everything through language and terminology, multilingually.

2.4EU legal texts as sites of engagement

Each text is a site of engagement for language, and that includes terminology and, in a multilingual context, translation. So, it seems useful to explore these relationships as part of the web of connections. Which EU texts are legal texts? How are they connected? Yet also how are they made and how are they interpreted and applied? This takes us into the process, or procedural, side of EU law, in particular how the multilingual legal texts are made. It is precisely in this dimension of EU law that terminologists play a vital role. Not only do they assist EU translators with difficult and complex technical texts, but they analyze and generalize knowledge painfully acquired term by term through a range of documents and materials that are made available to everyone working with EU law and texts. So, one needs to reflect on the text-producing methods, translation into all the EU languages and the roles and sources of information created by terminologists. We are generally interested in specific terms and in the relationships between them, but also in how they are created. It brings us into the realm of the formation of terms: the creation of a concept, its representation by a term in a particular language (primary term formation), and its representation by terms in other languages (secondary term formation). We encounter conceptual, structural, and cultural divergences between languages, and we note how the cultural legal identity of EU law plays a role.

EU culture thinks of unity in diversity. This is inherent in the wording and structure of the TEU and TFEU (“common action to eliminate the barriers which divide Europe” (TFEU recital)). EU law aims towards singularity while respecting diversity. This desire for singularity extends to terminology. An EU text is a single multilingual text, indivisible, with each language version expressing the same meaning; at least that is the idea and the ambition. Meanings are to be singular and language versions are to be equivalent. That places a pressure for terms across languages with respect to a given concept to be equivalent. This is quality. A reader in any EU country reading any EU language version must have the same understanding as others in a different country reading a different language version as regards any given EU text. This is equivalence. It is a heavy burden for drafters and translators. It is a core role for EU terminologists to help achieve this ambition. It is essential.

Which are the legal texts of the European Union that interest us? If we take a look at the internet portal to the European Union https://​europa​.eu and select our language, here English, we find a plethora of texts of all sorts telling us about the EU. We can access official documents, publications, statistics, data, and yet more resources. For the present purpose our attention is placed on legal texts, namely those which are created for the primary purpose of communicating EU law. Here, from a general legal perspective we can think about law-making texts, law-interpreting texts, and texts which discuss and analyze EU law for the benefit of professionals but also for assisting, guiding, and informing non-specialist general readers. Our choices have terminological implications as the purpose of each text and readership will vary and that has implications for rhetorical strategy, formal structure, content, and the use of language more generally in each particular text. To which we can add the dimension of texts intended for an external readership as opposed to those intended for internal communications among colleagues. Present attention is placed on legislative-style texts as provided for in the EU treaties as these are the primary expressions of EU law.

If we think of law-making as an act of communication, between utterer and receiver, we can think of the legislator as utterer and this circumstance conditions the structure and content of the texts, as well as the particular terminology. To be complete, we should study the receiver dimension, which would take us to the users of the texts in all languages and all Member States, but also to court cases on the interpretation and application of the legislative texts. Here, the CJEU provides the authoritative reference point through its decisions which can all be accessed via its web portal.1212. https://​curia​.europa​.eu​/jcms​/jcms​/Jo1​_6308/. Note that this site also mentions the need for freelance-legal translators on a permanent basis and explains the procedures for applying, thereby opening the door to practical experience: https://​curia​.europa​.eu​/jcms​/jcms​/j​_6​/en. Through its decisions interpreting EU legal texts, the CJEU influences the meanings of texts and terms for all language versions, and these interpretations must be read alongside the texts. In cases of divergences between language versions the CJEU has developed methods of multilingual interpretation (Robertson 2009; Baaij 2012, 2015; Derlen 2009; Schübel-Pfiste 2004; Paunio 2013). It is important to note that while there may be divergence between linguistic meanings of individual language versions, law requires a single meaning as it is linked to command and action. This implies that legal meaning is not the same as linguistic meaning, which can be a problem for a reader of one language version unaware of other language versions and EU Court interpretations, for they must all be read together. Another implication is that the meanings of terms in texts are coloured by being embedded in a legal text and interpreted according to legal methods. Thus, terms about trade will be viewed through law-tinted spectacles. Lastly, we must mention the preliminary negotiation texts, the travaux préparatoires as well as academic literature analysing and discussing legislative texts and judicial interpretations and decisions. These can have impacts on the meanings given to terms. The CJEU has a distinctive terminology of its own, but here attention is placed on the legislative side as it constitutes the essential foundations of the EU legal order.

The primary texts of the EU legal order are the treaties, in particular, the TEU and the TFEU. We can summarize their outline: title, participating states, preamble / recitals that set out background considerations, enacting or operative provisions which set out obligations, signatures, followed by protocols, annexes, and appendices under various names that deal with technical matters. A similar approach is reflected in secondary-level acts, except that the states do not need to sign them themselves as they have delegated powers to the EU institutions to make acts and sign them directly, albeit under close control through participation in negotiating each act and with a comitology procedure for political control included in each text.1313.On comitology, see EUR-Lex, Glossary of summaries at: https://​eur​-lex​.europa​.eu​/summary​/glossary​/comitology​.html. EU law-making is a collaborative process between the EU institutions and all the Member States in all the languages.1414.On the EU legislative process see: https://​ec​.europa​.eu​/info​/law​/law​-making​-process​_en; https://​www​.civitas​.org​.uk​/content​/files​/OS​.2​​.Law​making​​-and​​-Legislative​​-Process​.pdf; https://​www​.cons​ilium​​.europa​​.eu/​en/​council​​-eu​/decision​​-making/​ordinary​​-legi​slative​​-procedure/; https://​ec​​.europa​​.eu/​info/​law/​law​​-making​​-process/​types​​-eu​​-law​_en. Article 288, and following, of the TFEU sets out the foundation provisions for EU legal acts. Each act serves a specific function: a regulation has general application and is binding in its entirety and directly applicable in all Member States; a directive is binding, as to the result to be achieved, upon each Member State to which it is addressed but leaves to the national authorities the choice of form and methods; a decision is binding in its entirety and is binding only on those to whom it is addressed. Lastly, recommendations and opinions have no binding force. To find out what this means in practice, we need to examine individual texts. Each one is adapted in its contents to the specific task that it is called on to perform; in principle every clause and word is inserted for a specific reason, and that includes ambiguous wording.

There is not space to analyze individual texts here, but we can draw attention to guidance materials for drafting. Within a general context of Better Regulation (European Commission n.d.) the main guidance available includes the Joint Practical Guide of the European Parliament, the Council and the Commission for persons involved in the drafting of European Union legislation (European Union 2015) and the Manual of precedents for acts established within the Council of the European Union (General Secretariat 2002). These set out the ideals even if they are not always strictly adhered to. The Joint Practical Guide sets out general drafting guidelines, including on terminology, notably to avoid terms specific to particular systems (Guideline 5). The Manual of Precedents proposes styles, or models, for different types of EU acts, as well as information and guidance on terminology and standardized expressions, phrases, and paragraphs to be included. The guides have been carefully prepared in all EU languages so that the contents are equivalent. Standardized language, with set meanings across languages, facilitates legal drafting, but also translation, legal-linguistic revision, and the final adoption of texts. It is a step towards automation. General guidance on drafting in the EU languages is provided by the Interinstitutional Style Guide (European Union 2022).

3.1EU drafting

Legislative drafting manuals have a particular interest for terminology. This becomes apparent when one analyzes individual legal texts from a terminology perspective. At the level of broad domains, we see a general domain of text structure and terminology connected to the core construction of the texts. This is the foundation structure or what is informally known as the cuisine interne of the EU lawyers and lawyer-linguists. Essentially it is a domain of terminology that has been decided in advance and the main task for each draft text is to ensure that the relevant elements are included in the appropriate format. It brings us to the conceptualization of the legal text as a vehicle for action and to bring about some change in the legal order and the way of doing things, as a performative utterance. This in turn touches on the legislative intent, and it takes place in the policy field addressed by the text. It is the reason for the text, and the terminology here is connected to the language of the policy field, whether it be agriculture, competition, environment, or free movement. The knowledge of the experts in the field is essential in order to advise and give guidance on the terms and expressions that are in use.

It is possible to analyze a text through the policy domain linked to each term in it. Starting with the core cuisine interne structural terms used for every text in the particular genre (regulation, directive, etc.), one can identify terms for the policy fields addressed (agriculture, competition, environment, etc.). However, other domains are present, for example texts often contain penalties or sanctions, which may be civil or criminal (see e.g. Ruggieri 2014) which form part of the domain of civil or criminal law. Normally penalties are a matter for national authorities, so, by implication the reference carries over from EU to national law. The text may include links to international law, and it may make references to national bodies and authorities. Each one is different, according to its purpose.

The mention of penalties and to Member-State national law, draws attention to different roles for different types of EU legal instrument. If the act is a regulation its scope is to address the internal EU arrangements, but if the act is a directive its purpose is to operate on the domestic law of each Member state with the aim of aligning their legal systems towards objectives set out in the directive. This generates the thought that one can imagine two dimensions from a terminology perspective. On the one hand there is the dimension that is internal to EU legal language where the same act applies directly throughout all the EU Member States in the same way and the meanings are largely under the control of the EU institutions and bodies, in particular the CJEU. On the other hand, there is an outward dimension where the EU act specifically addresses the legal systems of the Member States and sets out provisions that are to be implemented in their domestic law in the form of domestic legal acts, or other methods, as appropriate. Here it is the national form of the terminology that often functions as the immediate source of obligations for individuals and businesses through the national legislation. The EU drafter seeks to select terminology in the EU act that will facilitate the process of appropriate transposition and drafting of the national laws. The pragmatic risk to avoid is of a reader of any language version of an EU text attributing national-law oriented meanings to EU texts without taking into account that these may not accurately reflect the intended EU meanings designed for all EU languages.

An EU directive is intended to be, first, interpreted and, second, rewritten (transposed), as necessary, in accordance with national law and language methods in each Member State, which implies different linguistic and legal contexts. The intended readers are, at a minimum, experts of EU law and language, experts in each Member State working in their official languages, experts anywhere in the world interested in the text, and a broad general readership, since law is supposed to be understandable by the citizens. The process of selecting terminology needs to be seen against that background. Each text and its tasks are, more or less, complicated, and we can mention here an even more complex example of a framework directive, such as Directive 2007/46/EC,1515.Directive 2007/46/EC of the European Parliament and of the Council of 5 September 2007 establishing a framework for the approval of motor vehicles and their trailers, and of systems, components and separate technical units intended for such vehicles (Framework Directive) (OJ L 263, 9.10.2007, p. 1). with multiple regulations under it. Interestingly, that directive was later converted into a regulation: Regulation 2018/858/EU.1616.Regulation (EU) 2018/858 of the European Parliament and of the Council of 30 May 2018 on the approval and market surveillance of motor vehicles and their trailers, and of systems, components and separate technical units intended for such vehicles, amending Regulations (EC) No 715/2007 and (EC) No 595/2009 and repealing Directive 2007/46/EC (OJ L 151, 14.6.2018, p. 1). Hybridity and fuzziness form part of the EU legal order. In practice, each term has to be studied individually and the parameters surrounding it must be thought through with care. One needs to be aware that the transfer of information from an EU text into a national context is a political and legal exercise, but also a linguistic activity that implies translation within a language, intra-lingual translation, from EU variety to national varieties (Robertson 2011, 2013). It takes place for each EU language, in parallel.1717.For an account of Irish experiences, see Ó Lúing (1976).

3.2Multilingualism

The EU legal order is multilingual and each time a new Member State has acceded and brought a new official language, that language has been added as an EU language (Gallas and Guggeis 2005; Ost 2009; Robertson 2017). The process has been managed via the accession treaties which amend the EU texts to include the new language, by translating EU texts into the new language, and by drafting the accession treaty in the old and new EU languages (see e.g. Šarčević 2001). The process of preparation and accession for accession is complex and lengthy. It implies the creation of terminology in the accession language to express the full range of EU concepts in use. This comes within a context of translation in preparation for joining the EU. The role of translators and of terminologists is of vital significance. The EU Commission manages and assists with the process.1818. https://​ec​.europa​.eu​/info​/policies​/eu​-enlargement​_en

The foundations of EU multilingualism are at two levels: primary law and secondary law. At the primary law level, there are two aspects. On the one hand, the treaty texts are written in all EU languages, and contain a language clause at the end that states which languages are authentic. Thus, we find Article 55(1) TEU which states that:

This Treaty, drawn up in a single original in the Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hungarian, Irish, Italian, Latvian, Lithuanian, Maltese, Polish, Portuguese, Romanian, Slovak, Slovenian, Spanish and Swedish languages, the texts in each of these languages being equally authentic, shall be deposited in the archives of the Government of the Italian Republic, which will transmit a certified copy to each of the governments of the other signatory States.

The article is applied to the TFEU by an intertextual cross reference in Article 358 TFEU which states that: “The provisions of Article 55 of the Treaty on European Union shall apply to this Treaty.”

On the other hand, the TFEU establishes the EU linguistic régime for internal activities. Here we find Article 342 which states:

The rules governing the languages of the institutions of the Union shall, without prejudice to the provisions contained in the Statute of the Court of Justice of the European Union, be determined by the Council, acting unanimously by means of regulations.

The relevant Regulation is Council Regulation No 1 determining the languages to be used by the European Economic Community.1919. OJ 17, 6.10.1958, p. 385–386. These provisions taken together provide a foundation for translation and terminology work in respect of all the EU languages. Regulation No 1 does not mention translation, and each language version of the EU treaties has the status of original source, as that is the signification of Article 55(1) quoted above, but we can reflect on how all EU language versions could be produced if not through having recourse to translation and terminology work. The provisions in question also provide a foundation for multilingual interpretation by the CJEU as it is required to draw the meaning of an EU legal act from all the language versions. Here it is relevant to note that the EU multilingual act is conceived as a single text expressed in, say, 24 languages, and its purpose is to aim at a single message expressed in parallel by all the language versions. To facilitate this, the layout of EU texts is aligned so that the same information appears at the same location for each language. This requires juggling in page lengths for secretaries as languages differ in the amount of space required, but it facilitates interpretation, translation and terminology work, as one can use a tool such as EUR-Lex, the EU data-base for legal acts, to compare different language versions for style, approach and content.

Each EU language has its own history and linguistic features. From a perspective of EU multilingualism in texts, one can mention two dimensions: a vertical dimension and a horizontal dimension. The former looks at the particular language and how it has been used in past texts, in other texts, and in new texts under preparation. This viewpoint seeks alignment and coherence within the particular language across all the EU texts and their domains. It is rooted in the national language, and it endeavours to keep as close as possible to it. However, each language is just one out of many. That leads to the thought that there is a horizontal viewpoint that is text-oriented, lays all the versions in a row and compares them: text by text, paragraph by paragraph, sentence by sentence, word by word. This viewpoint asks: do all the language versions carry the same meaning? There is pressure towards singularity, and that extends to term creation: can one form of term be adapted to all languages? The classic example of this is with the euro. When the courts come to analyze EU acts, this is part of what they do. Each language has its own features and forms of expression. Translators have personal styles. Divergences arise, but the question all the time is: do they matter? Do they affect the end result? Each time it is a question of individual analysis. The EU Interinstitutional Style Guide provides detailed guidance on EU multilingual language. However, it is proposed to move on now to address one particular branch of EU law: environmental law. It serves to illustrate some of the issues that have been canvassed.

3.3Environmental law

The fragility of the environment became a major concern for the international community in the second half of the twentieth century. The conceptual and normative foundations of legal provision in this area were laid down in 1972, the year of the Stockholm conference. Since that date, we have seen a legal blossoming in this domain (Bettati 2012, 21) And yet the term environment itself is not defined in the major treaties, codes of practice or guidelines (Birnie et al. 2009, 4–6). No definition of the term is to be found in the EU Treaties themselves. So what does the term mean?

An analysis of Article 11 TEU and of Article 191 TFUE provides some hints. An analysis of the list of objectives in the environmental domain set out in Article 191 and in Article 192 TFUE would suggest that the term covers humans, natural resources, land-use, town and country planning, waste and water. The inclusion of town and country planning suggests that the term is not confined to natural resources but also covers the cultural landscape (Krӓmer 2012, 1–2). Animal welfare, although it is not a general principle of community law would appear also to come within the scope of Article 191 TFEU (Case C-189/01; see also Krämer (2012) and Jans and Vedder (2012, 32–33)).

The lack of a definition is both a hindrance and an aid. It allows the institutions to adapt to developments and new needs for environmental protection but it also makes it impossible to pin-point the Union’s policy in what has been termed an almost impossible domain (Jans and Vedder 2012, 33). In addition, the definition provided by the Commission is also somewhat vague: “all of the aspects whose complex interrelationship provides the framework, the setting and human living conditions such as they are in reality or such as they are experienced.”2020.Communication de la Commission au Conseil pour un programme des Communautés européennes en matière d’environnement, JOCE n°C 112 de 1973, p. 1. Original reads: “l’ensemble des éléments qui dans la complexité de leurs relations constitue le cadre, le milieu et les conditions de vie de l’homme tels qu’ils sont ou tels qu’ils sont ressentis” (the translation is ours). We also know that the word environment only appeared for the first time in the Single European Act in 1987 although it had already appeared in the case-law of the CJEU. The CJEU has held that environmental concerns could be relied on to restrict fundamental rights and that environmental instruments could be based on Article 1002121.Cf. for example Case 21/79. The word environment appears for the first time in that judgment (Franck et al. 2002, 31–34). of the EC Treaty2222.For equivalences between articles of the EC Treaty and the TEU and TFEU see the Table of Equivalences at: https://​eur​-lex​.europa​.eu​/LexUriServ​/Lex​Uri​Serv​.do​​?uri​​=OJ:C:2012​:326​:0363​:0390​:EN:PDF. (which aims at the harmonization of Member State law for the functioning of the common market) and in later cases on Article 235. In Case 240/83, Judgment of the Court of 7 February 1985, Procureur de la République v Association de défense des brûleurs d’huiles usagées (ADBHU), the Court held:

that the principle of freedom of trade is not to be viewed in absolute terms but is subject to certain limits justified by the objectives of general interest pursued by the Community provided that the rights in question are not substantively impaired. There is no reason to conclude that the directive has exceeded those limits. The directive must be seen in the perspective of environmental protection, which is one of the Community’s essential objectives.

So we can see that an objective, that the EU had long espoused, took many decades to find legislative expression. We turn now to the way in which EU terminology is formed in this domain and in others.

4.1Terminology and the CJEU

A coherent and consistent terminology is vital to ensure legal certainty. Guideline 6.2 of the Joint Practical Guide provides:

Consistency of terminology means that the same terms are to be used to express the same concepts and that identical terms must not be used to express different concepts. The aim is to leave no ambiguities, contradictions or doubts as to the meaning of a term. Any given term is therefore to be used in a uniform manner to refer to the same thing, and another term must be chosen to express a different concept.2828.This guide was created by the legal services of the Commission, the Council and the Parliament to instruct “persons involved in the drafting of European Union legislation”.

Union acts are to be clear, easy to understand and unambiguous (Guideline 1.1). In order to achieve this goal, the Commission often has recourse to definitions, a technique borrowed from the common law.2929.On the hybrid nature of EU law see Robertson (2012). This technique has a controversial place within the law in general. Championed by many (see, for example, Goltzberg 2016), legal definitions are also subject to criticism arising from wider logophobic tendencies (see Hutton 2009, 68). Mac Aodha (2018) analyzes definitions in the field of environmental law. These are characterized for the most part by their simplicity and transparency.3030.This is not always the case, however. The term disposal for example is defined in Article 3(19) of Directive 2008/98/EC as “any operation which is not recovery even where the operation has as a secondary consequence the reclamation of substances or energy. Annex I sets out a non-exhaustive list of disposal operations”. As this definition is built on other definitions, there is a lack of transparency. Ultimately, the technical precision is provided from the embedding of definitions in directives and their operationalization by the CJEU in the case-law. In this respect the CJEU has a key terminological role to play as Berteloot (2000, 7–8) points out:

Indeed, the Court is a vehicle for the transmission of legal terminology in that it transfers concepts of national law, in particular principles and abstract ideas, from a national system to the Community system, from whence they may even make their way to another national system. Examples include the principle of proportionality in administrative law and the duty to have regard for the welfare of officials, both of which concepts were brought into Community law under the influence of German law.3131.The original reads: “La cour est en effet un vecteur de terminologie juridique dans la mesure où elle fait passer certains notions de droit national, notamment des principes et des notions floues, d’un système national au système communautaire, d’où elles sont mêmes susceptibles de migrer vers d’autres systèmes nationaux. C’est le cas du principe de proportionnalité en droit administratif ou du devoir de sollicitude à l’égard du fonctionnaire introduits en droit communautaire sous l’influence du droit allemand.” (the translation is ours).

The multilingual nature of EU law and need for a uniform application of that law inevitably complicate the CJEU’s interpretative role.3232.It is possible that the linguistic complexity of EU law is sometimes overstated. Interpretation in the legal-linguistic context of India, where there are three legal systems and a multitude of languages, is infinitely more daunting. Lerat (2014, 91), in proposing his model for a multi-lingual EU dictionary is also sanguine about the scale of the task: “There are jurisprudential differences but we should not overstate the reality: on both sides of the Channel and (of the Atlantic) a good is still a good, an acquisition is still an acquisition and a purchase is no more than a purchase, etc.”. The original text reads: “Il existe aussi des différences doctrinales, mais il ne faut pas en survaloriser la réalité: de part et d’autre de la Manche (et de l’Atlantique), un bien reste un bien, une acquisition une acquisition, un achat un achat, etc.” (the translation is ours). Where a single language is unclear, the CJEU will often consider the semantic similarity of the other versions as significant (Paunio 2007, 390). Paunio contends that the aim or purpose of the legislation is primary and that semantic issues are therefore often considered as of less importance in practice (Paunio 2007, 392). However, Baaij’s (2015, 142) analysis of the CJEU judgments over 50 years points to at least a similarly dominant role for the literal or textual approach. The same author identified 170 judgments between 1960–2010 in which linguistic divergences between Union acts were examined by the CJEU. Faced with this linguistic maze, judges often have to play the role of terminologists or lexicographers. A search for the term dictionary on the Curia3333. https://​curia​.europa​.eu​/jcms​/jcms​/j​_6​/en/ database yields some 243 results. It is in the Opinions of Advocates General that such references are mainly found. General-language lexicographical sources, such as the Shorter Oxford English Dictionary, the Cambridge Dictionary and the French Larousse, are cited in addition to more technical works, such as D. Crane’s Dictionary of Aeronautical Terms. Legal dictionaries, such as Black’s Law Dictionary, are also consulted. Courts cite general dictionaries precisely because they are not contaminated by the law (Hutton 2009), but the motivation for citing legal dictionaries is less clear. Devinat (2014) identifies the somewhat surprising motive of a quest for the ordinary meaning of terms. Given the principle of equal authenticity mentioned above, one might ask which ordinary meaning is being sought by the CJEU? One approach adopted by the Supreme Court of Canada, where both the English and French versions of statutes have equal authority, is to simultaneously identify one meaning. For example, in Monsanto Canada Inc. v. Schmeiser, the meaning of use/exploiter is discussed:

The starting point is the plain meaning of the word, in this case ‘use’ or ‘exploiter’. The Concise Oxford Dictionary defines ‘use’ as ‘cause to act or serve for a purpose; bring into service; avail oneself of’: The Concise Oxford Dictionary of Current English (9th ed. 1995), at p. 1545. This denotes utilization for a purpose. The French word ‘exploiter’ is even clearer. It denotes utilisation with a view to production or advantage: ‘tirer parti de (une chose), en vue d’une production ou dans un but lucratif. […] Utiliser d’une manière avantageuse’: Le Nouveau Petit Robert (2003), at p. 1004.6 (emphasis added) 3434.Monsanto Canada Inc. v. Schmeiser, [2004] 1 S.C.R. 902, 2004 SCC 34.

Such parity of treatment of lexicographic sources could never be achieved in the context of a multilingual EU law. Rare are the cases where lexicographic sources from more than one language are consulted (a notable exception being Case C-90/16, where the Oxford Advanced Learner’s Dictionary, the French Larousse, the German Duden and the Polish Słownik języka polskiego were all consulted for a definition of the term sport). It is perhaps no coincidence that references to lexicographical sources are largely confined to Opinions of the Advocates General. The treatment of terminology by the CJEU in this context leads one to evaluate the merits of a corpus-based approach to terminology.

4.2Corpora and legal terminology in the EU context

In the last quarter-century or so, terminology studies have witnessed a shift from the “principles of the Vienna school towards socio-cognitive and corpus-based descriptions of terminology” (Goźdź-Roszkowski and Witczak-Plisiecka 2011, 5; see also Temmerman (2000) and Pearson (1998)). Authors have increasingly turned to corpus linguistics, in particular, “to explain phenoma relevant to terminology and have borrowed criteria or methods from that field to feed terminological analysis” (L’Homme 2020, 3). The effects of this corpus revolution are gradually3535.Their utility in the field of translation is long-established; see for example, Baker (1995, 1996, 1999) and also in the domain of legal translation – Biel (2010, 2014). beginning to spread from cognate areas to the domains of jurilinguistics and jurilexicography. The use of corpora in contract interpretation (Mouritsen 2019), in judicial interpretation (Mouritsen 2017), as a paedagogic tool for trainee translators (Biel 2017), and as an interpretative tool in analysing terminology and other features of legal language in an institutional context (Prieto Ramos 2019) all point to a methodology in ascendance. Biel (2010) identifies the following types of legal variation that are identified and studied by means of corpora:

1. External variation: how does legal language differ from general language and other languages for special purposes?

2. Internal variation: how do legal genres differ from each other?

3. Temporal variation: how does the current legal language differ from a historic one?

4. Cross-linguistic variation: how does it differ across languages?

5. Idiosyncratic variation: corpus analysis, for evidentiary purposes, of witness statements and other legal texts.

Corpora have also been exploited in the area of environmental law. The LexALP3636. http://​lexalp​.free​.fr​/Codes/ project was designed to create a terminological database (French, German, Italian and Slovene) based on the Alpine Convention and designed to harmonize legal terminology relating to the management and preservation of the Alpine space (see Lavault-Olléon and Grossmann 2008). Those authors point out that legal terminology particularly in this domain should not be confined to attested legal terms “but must include the legal implications of scientific, technical and even general terms.” In fields like environmental law we are dealing with a mixed domain of truly technical terms, biological or medical terms, and then a whole host of semi-technical-cum-popular scientific3737.Further research should be carried out on the relationship between scientific and legal definitions. Hutton (private correspondence of 26 January 2017) argues that when we are dealing with “statutory definitions or similar stipulated definitions, there is probably no requirement that there be a consistent position taken, since there are overriding policy matters involved […]. However, for technical scientific terms with little or no circulation in ordinary language the onus is on the legislator not to transgress against the meaning that a scientist would give the term – except of course that scientists often disagree”. He also points out that there is a kind of circularity in that general lexicographers often include scientific definitions as part of ordinary meaning, often giving the scientific definition as the first meaning. concepts.

Corpora have also been used to analyze the multilingual language of the EU including its legislation3838. Mac Aodha (2018), for example, looks at divergences between the environmental terminology used at EU level and that used at national level by creating and analysing a corpus of EU legislation and case-law for the Irish language. and the CJEU case-law.3939. Peruzzo (2019) draws on a corpus of 16 texts of the European Court of Human rights to “analyse the techniques used to convey system-bound concepts in the target language”. One particular focus of corpora studies in this domain has been the concept of eurolect (coined by analogy with the term sociolect), a term used to denote the language used within the institutions themselves and in the texts produced by those same institutions for the citizens of the EU (Goffin 1994). The UNINT project contains the Eurolect Observatory for the interlingual and intralingual analysis of EU legal varieties. The eleven languages used in the EU sub-corpus are compared with matching national transposition measures (Mori 2018).

It should be pointed out, however, that the use of corpora in the legal domain is controversial. Vijay Bhatia et al. are sceptical about the utility of corpora (particularly large corpora) for the study of legal language, as the title of their 2004 article suggests: “Legal discourse: Opportunities and threats for corpus linguistics.”4040.It should be recalled, however, that this criticism is that of a discourse analyst who works in a field characterized by a qualitative methodology – see for example Maingueneau (1996); Hardt-Mautner (1995). According to the authors, the formalism of the legislative genre (“a frozen genre”) obviates the need to have recourse to corpora (in particular large-scale corpora). They are not convinced also that corpora can take account of the interdiscursivity of legal discourse (Bhatia et al. 2004). We would agree, however, with Biel (2014, 94) who points out that that feature of legal discourse is less characteristic of the legislative genre. In their concluding remarks, Bhatia et al. (2004, 224) argue “one needs to appreciate that qualitative analysis begins where corpus linguistics ends”. This is to ignore the nexus between quantitative analysis and qualitative analysis (cf. Charaudeau (2009), Degand and Bestgen (2004)).

Finally, although the utility of corpora for the study of EU terminology has been established, the warning note struck by Prieto Ramos (2019, 8) should be heeded: “The broader the area of investigation and the aspirations for generalization, the more complex (and the riskier) the definition of corpus sampling criteria that will ultimately underpin the acceptability of the research findings”. As Williams (2003) has pointed out ultimately the domain in question will determine to methodology to be used. An example of a corpus-based approach is provided by IATE.

4.3IATE4141. https://​iate​.europa​.eu/

IATE4242.Our focus here is on the main database but it is important to note the presence of two spinoffs of IATE: the popular public website termcoord.eu managed by the Terminology Coordination Unit of the European Parliament and EurTerm, the interinstitutional terminology portal used by EU staff. (Inter-Active Terminology for Europe) is the terminological database of the European Union. It contains over 8 million terms in the 24 official languages of the Union.4343.Latin is also included. In this section, the history of this linguistic tool is first traced, its strengths and weaknesses are explored, and finally we examine the impact of the new version of the database: IATE 2.