EU Legal Language, Translation and Terminology: Twelve Viewpoints on EU Multilingual Law-making1

This paper aims to offer a series of snapshot viewpoints drawn from practical experience and use them as starting points to reflect on a range of topics that illustrate certain features and dimensions of EU law-making methods. Thus, EU legislative acts may be considered messages that are communicated, but also performative utterances. They are multilingual, according to EU treaties, which leads to several linguistic implications. EU legislative acts aim for singularity, are standardised as much as possible and are intended to have consequences within the national legal systems. EU legal meaning does not always align on linguistic meaning. EU language is a variant of the national language and can be analysed using “semiotic” methods. EU law is one out of several interlinked legal orders and EU legislative drafters take into account the fact that the legislative acts will be translated into EU national languages. In this respect, there are databases of documents and terminology resources that help translators carry out their task.


Introduction
The International Conference hosted by the Faculty of Law, 'Alexandru Ioan Cuza' University, Ia i, Romania on 13-14 May 2021 was devoted to the theme of legal translation and past and present challenges in Europe. Romania has been a member state of the European Union for several years now, since its accession in 2007, and Romanian is one of the official languages of the European Union by virtue of the 2005 Accession Treaty 3 , the Treaty on European Union (TEU) and the Treaty on the Functioning of the European Union (TFEU). These treaties provide for the making of rules, laws, and legislative instruments in the languages of the European Union, and since there are currently 24 official languages, this 1 Based on the presentation given at the international conference 'Legal Translations -Past and Present Challenges in Europe', held on 13-14 May 2021, organised by the Faculty of Law, 'Alexandru Ioan Cuza' University, Ia i, Romania. Websites were active on 23 September 2021. This presentation draws on the work C. D. Robertson, Multilingual Law. A Framework for Analysis andUnderstanding, Routledge, Abingdon/New York, 2016, softback 2018. 2 Former lawyer-linguist at the Council of the European Union, expert within the UN Harmony with Nature Programme, member of the UN Association in Luxembourg. requirement involves the production of multilingual texts and a heavy engagement in translation and interpretation which includes the Romanian language.
One of the features of the EU is that EU law is regarded as a legal order, or type of law, that differs from the domestic or national law of its member states, as well as from international law, and it takes precedence over the rules of national law. The consequence of this is that multilingual EU legal texts apply to Romania, as to other member states, and the Romanian language versions of EU law take precedence over Romanian domestic rules. The situation in practice tends not to be so simple however, because EU law often takes a broad aligning and harmonising view that leaves many details to be interpreted and applied by the member states by rules made under their law. This creates a hybrid zone of the merging and inter-relationship of EU and national provisions. Often there are uncertainties in an EU text that require interpretation and implementation locally. That means that there is an extensive role for interpretation of the meanings set out in the EU texts. The question thus arises as to how one can understand and make sense, but also make full use, of the methods that are provided by the EU context in order to clarify meanings and intentions from the beginning of each negotiation process. An important part of this question is linked to the methods by which EU institutions create rules. This in turn leads on to questions of translation and terminology. In many ways questions relating to terminology lie at the heart of the EU method.
How can one understand and make sense of the EU law-making process? One answer is to follow a detailed course, another is to participate actively in the process. Here it is proposed to offer a series of snapshot viewpoints drawn from practical experience and use them as starting points to reflect on a range of topics that illustrate certain features and dimensions of EU law-making methods. This does not provide a complete detailed understanding to match a full course of study, but it provides some starting points for going more deeply. It allows us to focus on some areas that are of particular significance for understanding the processes. This presentation draws on many years of practical work, and research, as lawyer and legal-linguistic expert, most notably within the Directorate for Legislative Quality of the Council of the EU.

Twelve Viewpoints on EU Multilingual Law-making
EU law, language and terminology constitute a vast domain, and we merely need to think about the many thousands of texts that have been produced over the years to appreciate this. There are many dimensions to the EU, for example its history, economics, politics, legal methods, and policies. A quick tour of the TEU and TFEU give an idea of the range of domains covered, but it is only the tip of the iceberg. To these policy areas we can add language, for EU law is explicitly multilingual. Thus, we can think on the one hand about the linguistic policy of the EU in terms of its rules on language use, its linguistic regime (the law of language), but also the ways in which language is used to create the texts on which EU law and practice are founded (the language of law). Here we can think of the texts that have been produced, but we also need to address the processes by which the texts are arrived at, as well as their subsequent life as practical tools intended to be used throughout the member states, and beyond, to bring about changes and effects.
Everyday EU life is based ultimately on a foundation of written texts in the EU official languages. Our focus of attention here in particular is on the process, or procedural side of EU law-making. It includes drafting, translation, negotiation, and interpretation, but also terminology and methods by which terms are developed for the language versions. The nature of EU language is that it is heavily dependent on legal forms and methods, given that the foundations are in legal treaties and its function is to regulate activities between member states. Thus, the approach needs to be combined not only with an awareness of the economic and international contexts, but also with an awareness of law and legal methodology. That in turn leads to a composite vision that embraces both law and language, a legal linguistic viewpoint.
EU law and language are founded on treaties, and these in turn are contracts between states for the purpose of achieving a range of shared objectives. Treaties are made under international law, using international law concepts and terminology, and the EU treaties are not an exception. Most treaties, however, do not seek to regulate such a wide range of economic activity. Nor do they generally organise such detailed and sophisticated law-making arrangements. Nor do they necessarily assert priority over national domestic law. Nor do they have so many official languages. All these features mark out the EU treaties as different, and more ambitious. They create a framework structure at a primary level (primary law) that provides for delegated acts to convert the framework into a myriad of details (secondary law) via instruments bearing names such as 'regulation', 'directive', 'decision', and 'opinion', each with a particular purpose and set of characteristics defined by law. Primary law and secondary law texts create a system, or context, for the use of specialist legal language with specialist terminology.
We can study EU texts as lawyers looking for rules and their meanings in relation to rights and obligations to be acted on. We can also study the texts as linguists looking at the languages, the terms and concepts, formal structures, discourses, grammar, syntax, variety of alphabets, and by extension translation, revision and terminology. A legal-linguistic viewpoint combines these dimensions and embraces both legal and linguistic expertise; it is the underlying viewpoint presented here. It is the viewpoint of the EU lawyer-linguist, whose aim is to ensure the production of quality texts in all EU languages fit for purpose within the EU and in the legal system of each member state.
We will explore this legal-linguistic viewpoint on EU law-making by adopting 12 viewpoints on EU legislative acts. They are anchored in the idea of each act constituting the communication of a message, and we progressively explore some of the implications in the EU context. This is the first viewpoint offered. The others look at acts as performative utterances and compare legal and linguistic ways of looking at them. There follows a translation viewpoint which embraces the terminology viewpoint, and the usefulness of standardisation in language is touched on. The viewpoint of semiotics can help drafters and translators when analysing terms and concepts. This is the case with interpretation across languages, as EU acts aim for a single legal message, but sometimes the language versions contain variations. Furthermore, drafters and translators write in an EU context, but the texts are often intended for implementation in national systems and require 'transposition' and rewriting in the national language forms and terms. EU language is a different variety form from national language. Against that, one must recall the dimension of international law; EU law is one of several legal orders, each having its own set of contexts. Lastly, there is the viewpoint of machines and technology, and here the attention is restricted to the databases for documents and terminology. One could add more viewpoints, but space and time are lacking. We will take each Viewpoint in turn for a brief visit.

Viewpoint 1: EU legislative acts as communication of a message
Communication implies: an 'utterer', 'receiver', 'message', and 'medium'. What does that mean in an EU context? We can ask of EU texts: who utters? who receives? what message(s), which medium(s)? We can think of EU legal texts produced in the language versions as linguistic products, and we can place them within a dynamic setting that views them within an ongoing procedure, or process, of creation, interpretation, use and application. We can explore each phase in a process and ask wh? questions (in English): who, what, which, when, where, why, and how?) For example: drafting texts: who drafts, how? why? which type of text? Or translation: who translates? why? how? what type of translation? We can ask similar questions about the negotiation of EU texts, as well as their amendment, the multilingual alignment of language versions, the finalisation of the multilingual text in all languages, the adoption of the texts, their signature, publication, and entry into force. All these aspects form part of the process of creating the 'multilingual message'.
We can go further. Once the act has been adopted, signed, and published in the Official Journal of the EU and comes into force, the question of its use arises. This involves interpretation, understanding and application. The drafters have spent many hours trying to anticipate possible ways in which the text may be read, or misread, by users. They can seek to point out and explain in recitals in order to influence readers, but each reader comes to the text with their personal baggage of culture and understanding, as well as particular interests to be furthered. The process of reading involves understanding, and that in turn demands a certain level of knowledge. Most EU acts are technical, and care is taken to consult future users to ensure the desired results actually materialise. The official readers and interpreters are courts, both national and the Court of Justice of the EU, but it is the latter whose rulings and decisions are determining. The EU Court works with all languages and its interpretation methods are therefore multilingual. What if there is a mistake or different nuance in some language versions?
In a few words we have evoked the EU law-making process. The EU legal drafter must pay attention to the various aspects, but so must the translator. The subtle variations of legal acts are crucial, as people's livelihoods depend on words and their meanings. The translator worries about linguistic elegance and accuracy, but the main requirement is for accuracy with as much elegance as can be attained. Yet there is the problem that oftentimes negotiators settle on ambiguity and uncertainty as the only means to achieve a compromise and that can pose problems for languages that structure their concepts, grammar, and syntax in different ways. The only solution is to draw attention to these issues while the text is still under creation and see how other languages are handling the difficulties.

Viewpoint 2: EU legislative acts as performative utterances
A performative utterance is a text or oral statement which changes (legal) relationships. Legal texts specialise in regulating relationships. They operate within an abstract conceptual 'world of law' but serve to influence human behaviour, and thereby have effects in the wider world of things. From this perspective, EU law has many types of 'utterances'. Mainly these are written, but the oral is also present. Thus, we can think of legislative, judicial, and administrative acts. Each is perceived as a written text, but each has also been negotiated orally through discourse in multilingual environments, with the aid of translators and interpreters.
That said, perhaps we can add a category of 'quasi-performative' or ancillary acts whose purpose is to help to implement and explain the thinking, purpose and methods of a legal act, and generally help it to achieve its intended purpose; for example, explanatory notes and guidance. However, we might also broadly include EU opinions, recommendations and guidelines which are not seen as 'binding' and obligatory in the same way as a legal act, but which point the way towards desired actions and results. EU discourse draws a distinction between 'hard law' and 'soft law' in this respect.
This second viewpoint opens the door to thinking about types (genres) of act, how they are made, the language styles they use, who the authors are, the domains they cover, and more. We can extend the list by asking wh? questions: what, who, where, when, which, why and how. With these questions we come within the specialist domain of EU legal drafters.
EU legislative acts are constructed in a hierarchy: level 1 (primary law) comprising the EU treaties, and level 2 (secondary law) comprising delegated acts such as regulations, directives, decisions. Sometimes there are further delegated acts, as where a regulation provides for the EU Commission to adjust technical annexes to it by a further act. Secondary law acts are made under powers in primary acts set out in articles. Each time there is a dimension of authority and power relationships, but there is also a linguistic and terminology dimension that impacts on translation. For the language of the higher-level act applies equally to the lower-level act which is not free to change or alter it. Thus, for terminology, terms in higher level-acts apply to, and prevail over, those in lower levels made under them.

Viewpoint 3: EU legal acts are multilingual: legal basis
If we think about multilingualism in EU legal texts, we can ask a range of questions, for example: 1) Which language(s) is a text written in? 2) Is there an article somewhere that says which language versions are 'authentic', official and determining? An authentic text is one that speaks directly and does not need to be 'proved' by other forms of evidence. 3) What is the 'linguistic regime' that applies to the text?
These questions reveal a relationship between the language of law and the law of language. The lawyer uses language as a tool for action and obtaining effects. The linguist studies language and reproduces a message through translation. With EU law, the fact of multilingualism binds lawyers and linguists more closely. We can consider that there is a 'legal' side and a 'linguistic' side: (a) the law of language, the linguistic regime, rules on language; (b) the language(s) of law, legal language, the language of legal texts and discourse. The law of language sets rules on the language code(s) to be used, when and how they are to be used, but there are also the ways in which language is used in specific texts, which may be laid down by law. We can think of terminology, for example terms set by law in a text, perhaps through definitions.
What is the legal basis for EU multilingualism? Let's return to our questions:

1) Which language(s) is a text written in?
EU Treaties are adopted in (all) the EU languages. One can see this from publications in the Official Journal or by consulting the EUR-Lex website. EU texts are created in stages, generally with drafting in one language, then immediate translation. It is the final adopted version that counts. When a new language accedes to the EU, that language version is added through the mechanism of an accession treaty.

2) Is there an article that says which language versions are 'authentic'?
In EU treaties we expect an article on this. We find it in the TEU, in Article 55. In the TFEU we find it in Article 358, which simply applies Art. 55 TEU, which is in the following terms: 1. This Treaty, drawn up in a single original in the Bulgarian, 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.

This Treaty may also be translated into any other languages as determined by
Member States among those which, in accordance with their constitutional order, enjoy official status in all or part of their territory. A certified copy of such translations shall be provided by the Member States concerned to be deposited in the archives of the Council.
There are some points to note. We have a single message expressed in many languages. Each is equally authentic and so each is a source for direct interpretation. This is a starting point for multilingual interpretation of the treaty text. But are all the sentences and terms in each language the same or equivalent as regards meanings? They express the same message, and they insert information in the same place in the texts across languages. These are aids to interpretation. Now, let us turn to the third question, on the EU linguistic regime.

3) What is the 'linguistic regime' that applies to the text?
The TFEU in Article 342 says: Article 342 delegates the task of setting the linguistic regime to a secondary level act. It provides the power for the act, and it specifies the type of act. The article is implemented by Regulation No. 1 determining the languages to be used by the European Economic Community. Article 1 of it, as amended, states: The official languages and the working languages of the institutions of the Union shall be 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.' We can ask what this article means in practice. To provide an illustration, we can select one short article and reproduce it in the language versions. We can bear in mind that every article, as well as every recital, in every legal act, as well as other types of EU text, follow a similar pattern, however long. Each time there must be matching language versions. The article selected is Article 7 of Regulation No. 1. It is very short, which means we can reproduce it here. It touches on the Court of Justice. However, the main interest for the immediate purpose lies in the way the concept of 'régime linguistique' in French is rendered across languages. In English we have 'languages to be used'; other languages have different equivalent formulations. We can ask whether these differences in approach matter, and while we can answer negatively here as the EU Court has control over its rules of procedure, we can pose analogous questions in relation to other texts on different matters, and, in that way, we are led to reflect on the processes by which one might go about interpreting multilingual texts. Here is Article 7: Bulgarian: . Croatian: Jezici koji se koriste u postupcima Suda odre eni su u njegovom poslovniku.
English: The languages to be used in the proceedings of the Court of Justice shall be laid down in its rules of procedure.
French: Le régime linguistique de la procédure de la Cour de Justice est déterminé dans le règlement de procédure.
Irish: (no version currently available as secondary legislation was not produced in this language) Italian: Il regime linguistico della procedura della Corte di Giustizia è determinato nel Regolamento di procedura della medesima.

Viewpoint 4: EU legal acts are multilingual: linguistic implications
We can return to the concept of communicating multilingually and think about some other aspects. For example, with drafting we are interested, among others, in methods of drafting, styles, templates, guides, and terminology. In the case of translation, we want to know about methods of translating (accuracy versus style), equivalence in every nuance of message, and many other matters, which take us into revision and legal-linguistic revision. We are led to speculate on EU language, and we see how each EU language has origins in the national language but has been adapted to the EU context and become an EU variety. For each EU language, we can study features and influences, such as non-native drafting for French or English, and the occurrence of calques and unusual turns of phrase and formulations. We can note that French was an original EU drafting language, with concepts from different systems, but then English became a major drafting language. This contains traps for the unwary as English was originally a translation language, with translator variations in words in different texts as compared with the four original founding languages which may have kept the same terminology. So, one must check them for term consistency too.
EU languages have a sometimes-difficult relationship with the national form. To fully understand EU language, one needs to be aware of the EU economic context, its use of legal language and forms, and the specialised and adapted terminology and methods worked out over the years in relation to each domain. One finds oneself within a deep field for linguistic study. It includes not just the various ways in which language is used in EU texts, but also a comparison of forms between EU and national law language in different member states, as well as comparisons between EU and international law language, terminology, and forms. One question is how far the language used in a text may render it inaccessible to EU citizens and whether that should be a source for concern. These issues lie within the field of applied linguistics, LSP (language for specialised/specialist purposes), and by extension to each language, such as ESP in the case of English.

Viewpoint 5: EU legislative drafters draft for translation
EU legal texts are mainly drafted by non-native speakers, and this has probably been reinforced as a result of the UK exit. The texts must be translated into the EU languages. They must be understood by people in all EU countries, but also beyond, as the EU trades with the world. They must conform to EU methods and practices. The texts contain EU concepts and terminology. Different types of acts serve different purposes. EU directives are drafted to be 'transposed' into national law and national language in each Member State. Transposition is a 'legal' activity that involves creating a national legal text. Yet it is also a 'linguistic' activity that involves converting EU language into national language. We see here a process of 'intra-lingual' translation, involving translation from one variety to another of the same language. Terminology lies at the heart of practical questions, centring around the meanings of words. Difficulties arise if the same words are used in EU and national contexts and if they have different meanings attached, as there is the risk of confusion and misunderstanding. Similarly, where a nationallaw trained lawyer reads an EU text without knowledge of EU law and its specialist-language context, they may apply their national-law understanding to the EU text, resulting in misunderstanding and erroneous decision making. In this context, we can mention that EU drafters follow EU guidance, notably the Council Manual of Precedents 4 and the Joint Practical Guide, 5 but also the Interinstitutional Style Guide. 6 These guides exist in all the EU languages, and they facilitate uniformity and standardisation.

Viewpoint 6: EU legislative acts are standardised as much as possible
Standardisation is an important feature of EU language as it helps in drafting, translation, and interpretation as meanings become established and settled across languages. Linked to the EU terminology databases, notably IATE, 7 and computerassisted translation, standardisation also helps to speed up processes as past solutions can be reused without detailed enquiries each time. The approach extends to the formulation of legal texts. Each part of an EU legislative act has a standardised format in terms of layout and style. Numbering is standardised. Recitals do not include verbs of obligation such as 'shall' but try instead to use conditional verbs such as 'should', because recitals are not intended to be 'enacting and binding' like articles. There are also standardised formulae. For example, there are special agreed wordings in relation to 'subsidiarity', 'comitology' (committee procedure), 'transposition', and more. Standard clauses are worked on with special care across all languages and then become equivalent.
For each EU language we can consider that there is a 'horizontal' and a 'vertical' viewpoint. The concept helps us separate two sets of dynamics for equivalence or variation. The horizontal viewpoint looks at all language versions in a row, and we can imagine lining up all the terms for the same EU concept across languages and comparing them, like we proposed earlier in relation to Article 7 of Regulation No 1. The vertical view on the other hand looks at the internal side of an individual language, its past, present, and possible future in relation to terms used and other expressions and concepts. This view addresses consistency within the language across texts and drafter and translator variations. We can summarise these viewpoints by suggesting that the horizontal viewpoint asks whether all 4

Manual of precedents for acts established within the Council of the European Union,
Online at https://op.europa.eu/en/publication-detail/-/publication/431ccffd-00c2-491a-b423-ce709af0d6c3, accessed on 24/09/2021. 5 Joint practical guide of the European Parliament, the Council and the Commission for persons involved in the drafting of European Union legislation, Online at https://op.europa.eu/en/publication-detail/-/publication/3879747d-7a3c-411b-a3a0-55c14e2ba732, accessed on 24/09/2021. 6 Interinstitutional Style Guide, Online at https://publications.europa.eu/code/en/en-000100.htm, accessed on 24/09/2021. languages carry the same message, and there is equivalence across language versions. The vertical viewpoint asks about the history and internal patterns, past, present, and possible future within each language, and about internal consistency in terminology between texts in given domains. The task of the drafter and the translator is to try to be consistent both 'vertically' and 'horizontally'.

Viewpoint 7: EU language can be analysed using 'semiotic' methods
'Semiotics' is the study of signs. Language comprises signs, as words are 'signs' or 'tokens' for ideas. A sign 'stands for' or 'represents' something, such as a physical object or an idea. Here we can note that law is essentially abstract and consists in ideas represented by words which lead to behaviour, actions, and effects in the world of phenomena. Legal language uses signs (language) to represent ideas and concepts. So does EU legal language. EU concepts are expressed as words, and here we think of terms, and terminology. EU concepts are expressed in equivalent terms in the different EU languages, and this is the 'horizontal' aspect of EU languages, just discussed. The problem is to be able to know whether the terms and texts in each language are actually equivalent? For a translator, the problem is also to decide which of several possible terms may be the most apt for the text in hand, or whether to coin a new term. It is suggested that one can make a semiotic analysis of terms across languages and compare findings. It is what translators do automatically, but there is a difference between a two-language context and a 24language context, just as there is a difference between translating a legal text for information as to its contents and translating to produce an equal authentic legal text with the same value as the 'source' text for effects and interpretation. In such contexts some explicit semiotic analysis can be helpful.
There are different views and theories in semiotics, but this is not the place to delve into them, just as we cannot delve into legal or linguistic theory. Nonetheless, we can draw on a few ideas that may prove useful in practice. The first remark to make is that this article is grounded in semiotic theory. Ferdinand de Saussure taught that the viewpoint defines the object. Here, we have selected 12 viewpoints and used them to help us understand EU legal language. He also proposed that the concept of the sign could be seen as comprising two elements, the signified (thing), and the signifier (word). This is helpful when working with two languages, but what about when working with 24 languages, where one worries about variations between them? Here the insights of Charles Sanders Peirce (pronounced: Purse) come to our aid. He proposed the sign as comprising three elements: object (thing), representamen (word), and interpretant. It is the concept of the 'interpretant' which is added that is particularly useful, though difficult to pin down, as it involves thinking about the precise relationship between the other two.
Thinking about the interpretant leads us to ask about what exactly the 'object' is and whether the object in one language is actually the same as in another language. It sets us off in a direction that thinks about cultural attitudes and variations, especially in national legal contexts, when negotiators come from different national and legal contexts to the EU context with the need to create EUwide texts that can work for everyone. The Peircean 'interpretant' helps to analyse subtle variations in context and culture, and to make comparisons. It does not solve the problem, but like taking a machine apart, it helps to understand where some of the difficulties and confusions lie and to start working towards a shared approach with agreed terminology, which can be useful. The translator is empowered to ask questions to understand the intentions of a source text more clearly. That also is part of the work of EU legal-linguistic revisers when preparing a legal text for adoption and signature.
Another dimension for semiotic analysis lies with transposition of EU directives into national law and language. Apart from the structural methods adopted to make national legal texts, where different approaches exist, there is the question of transposing EU language and meanings into the national context. Does one keep the EU meanings, or does one convert into domestic conceptualisations? Will the EU Commission come along afterwards and find fault and raise court proceedings for not implementing the EU directive correctly? One must analyse and compare EU and national terminology and their contexts. EU drafters, translators, interpreters, and transposers into national law can all be assisted through analysis using semiotic methods.

Viewpoint 8: EU legislative acts aim for singularity and EU legal meaning does not always align on linguistic meaning
EU law aims to be a coherent self-consistent 'system' of law. This ambition underpins its methods, and for each legal situation the aim is to have one answer on 'What is the law?' That implies that EU law aims towards unity (in diversity) and equality, but also 'singularity' in the sense that each legal act should convey the same message with the same rules and effects across all languages (horizontal viewpoint). But with 24 language versions, maybe some texts vary (revealed through semiotic analysis). We took a simple example with Article 7 of Regulation No 1 where we can see that the French expression 'le régime linguistique' is rendered with different equivalences across languages and, moreover, there are patterns of similarity. The issue is not important for that text, but for other texts differences in interpretation may be a source of difficulty.
The EU Court of Justice interprets and analyses the EU language versions. We can imagine the sort of questions that are posed. For example, what do the texts say linguistically? What is the meaning to be drawn from each version (bearing in mind a semiotic analysis)? We can refer to this as the 'linguistic meaning'. But, if there is divergence in the linguistic meanings between texts, whose meaning prevails? A legal system requires a single rule applicable to everyone, a single legal meaning for all the texts. There seems to be an irreconcilable conflict. However, being law, the solutions are found through legal means, and, in this case, they concern the rules and methods of interpretation followed by the Court. The EU Court has its methods and approaches for multilingual interpretation. It is not the place here to delve into them, except to suggest that the Court seeks to find out the true intention behind the text, which can involve looking at the negotiation documents, and the Court can be expected to think how alternative possible solutions sit with the rest of EU law and with the wider legal culture of European legal systems, international law, human rights, and so on. Judicial interpretation can be a complicated process at times.
From a linguistic and terminological viewpoint, what matters is the solution: words on pages. Yet there is an implication, and that is that some language versions of a text are held to express the legal meaning, while others do not exactly. For the latter, a gap opens between linguistic meaning and legal meaning. The consequence is that EU legal acts must be read with court decisions on their interpretation. Ideally there should be an amendment and alignment of the 'deviant' texts, but that is often not feasible. The consequence is that drafters must be aware of EU Court interpretations when amending an act, and translators must think whether the message is accurately translated and equivalent to other languages and bear in mind that there may have been a Court judgment on interpretation that is relevant. This reality emphasises how EU translators come into a position of being like legal drafters for their language. They must pay attention to EU drafting guidance and methods. That said, it is the task of the EU legal-linguistic experts (lawyer-linguists) to check all language versions of legal texts before they are adopted and signed, and part of this role is to check such issues. So, some of the burden is taken off the shoulders of translators, but the revisers must often cope with large volumes of text and so it is preferable if the translators are aware. That said, EU Terminologists assist translators with terminology, and this can include making research into possible court interpretations of terms for inclusion in the IATE database.

Viewpoint 9: EU language is a variant of the national language
All EU languages have their roots in the national language, and each language tries to keep close to the official national form. So, spelling, grammar, alphabet, syntax, general terminology, and suchlike aim to be as close as possible. But there is, and there must be, adaptation to the EU context. The biggest issue is probably terminology, and it arises most acutely when a state and new language apply to join the EU. Typically, the new language has few or no terms for the EU institutional structures and the vast range of economic and technical terminology in use every day by the institutions and member states across all the policy sectors of the treaties. The terms must thus be created. It is a specialist task. It requires specialist knowledge of the different fields. It requires collaboration between subject experts and linguists, and the results need to be checked and validated, which is done through the process of making and adopting an accession treaty and translating the EU 'acquis' into the new language. Term creation is a specialist role for terminologists who have their methods.
Pre-accession terminology work mainly involves 'secondary term creation' from existing EU concepts and terminology. The situation is more nuanced in ongoing EU work. Often drafters and negotiators decide to create new terms, or adopt terms already circulating in the member states for the particular domain. When that happens, the terms acquire EU meanings. Terminology must be found or created in all languages for the concept. Being an EU official language obliges each language to develop a very wide vocabulary range. It also requires significant resources to maintain it, which is why in the case of lesser-used languages such as Irish and Maltese there have been some temporary derogations that have been agreed over time.
EU terms exist for technical specialist purposes to express EU concepts. We can take standard examples such as 'regulation', 'règlement', 'Verordnung'… Terms across languages are 'equivalent labels' for a same concept ('horizontal viewpoint'). Terms are 'signs': EU terms aim to have the same 'object' and the same 'interpretant', but a different 'representamen'. There is a conceptual unity across the language versions, and we can think of this as part of EU legal culture. This culture impacts on creating new terms (cf. sheepmeat, gender mainstreaming). The Euro is an example, as basically the same word is used for all languages (cf. Greek ). Moreover, legally the term is invariable grammatically in singular and plural. There are other examples of EU legal-linguistic culture. For example, EU legal drafters should search for concept and terminology solutions that are not too specific to a particular national legal system (Guideline 5, Joint Practical Guide).
Transposing an EU directive into national law implies 'intra-lingual' translation, and semiotic analysis helps in analysing the similarities and differences. Each term in a legal text takes its context from both the text and the system, but perhaps there are similar parallel texts which have a bearing too? Ultimately, it is the courts that determine meanings, and there is a further implication for meaning here as not only is the context highly legalistic, but each court follows the methods applicable to it to arrive at its interpretations. Each court is part of a particular legal system and follows the rules on evidence and interpretation laid down within it. The practitioner must navigate through all these considerations.

Viewpoint 10: EU legislative acts are intended to have consequences within national legal systems
The EU exists for the benefit of the Member States: that is its purpose. There is an 'internal' context of EU law and language, and here one thinks of the role of the regulation to apply directly without transposition. The treaties create the EU internal context, and it applies to all. We think of the example of the 'internal' (single) market. People and businesses live in national territories (as do EU officials), and EU regulations are intended to apply everywhere 'in the same way'. This approach enables people, goods, and capital to move widely without obstacles. EU directives are different in that they apply throughout the EU but are mediated in each Member State by national law, through the process of transposition. Directives can be imagined as being framed more broadly than regulations, but some can be very technical and specific. The process of national implementation is monitored by the EU Commission. It brings actions before the EU Court of Justice for non-compliance with treaty obligations. Individual parties may raise proceedings in domestic national courts and seek a ruling from the EU Court on the interpretation and application of an EU act.
EU acts regulate certain areas directly, such as customs duties and the external frontier. In other areas they seek to 'harmonise' and coordinate between Member States to bring about levels of equivalence. One can think of directives on equal pay and treatment for women and men, and directives relating to the environment, the basis being linked to economics and competition. Another approach consists in following 'international private law' methods intended to regulate indirectly through indicating which legal system is to apply. Here EU law does not directly provide the 'solutions', but sets rules that can be used to determine whose national law applies in particular situations to provide the solutions. One example concerns court jurisdiction and choice of court for litigation. Another concerns social security regimes and calculating benefit entitlements. Once again, these are technical areas. On the other hand, many areas are excluded from EU action. For example, rules about family and personal status are matters of national law, as are the rules about property rights. Taxation is also excluded, but one has EU rules on value added tax (VAT). The EU concept of 'subsidiarity' implies that matters best handled at national level should be undertaken at that level. The concept has been transformed into a standard 'subsidiarity' clause that is included in the recitals of relevant acts in all languages.
EU law and language thus have their limits. The core role is to address crossborder cooperation and fair competition within a shared customs union and an internal market. Through the agency of the EU institutions, member states have conferred rights, but also created obligations, for persons and businesses. The EU viewpoint is primarily economic, and this perspective colours all EU legal language and concepts. That said, each EU legal act is adopted for a purpose and seeks to explain its aims and its limits in its recitals. Each act is a product of negotiation. The paymasters and ultimate decision-makers are the member states, and the role of the institutions and organs is to enable everything to function for their collective benefit and on their behalf.

Viewpoint 11: EU law is one out of several interlinked legal orders
EU law is created by treaty and the current twin foundations are the TEU and TFEU. The EU Court has called it a 'legal order'. 8 There are other legal orders and the most prominent of these are national law and international law. EU law is a derivative of international law and EU treaties are international law acts. Each 'order' and each 'legal system' provides a context for meanings (cf. semiotics). Yet, concepts and terms are created in one context and move to other contexts. They may move through adoption in drafting or translation. For example, EU law may implement a UN text, and that may in turn lead to national laws through 8 Van Gend en Loos, Case 26-62, judgment of 5 February 1962, Online at https://eurlex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A61962CJ0026, accessed on 12/08/2021. transposition, carrying the concepts and terms with them. Alternatively, an EU text may draw on national experiences to make an EU regulation or directive. It is simpler to work 'downwards', from higher, more international levels towards lower, more specific ones. This is linked to the relationships between the legal orders. The UN level is world-wide and embraces both EU member states and many more third countries. A legal text adopted at UN level binds all states, including all EU member states. Maybe that means there is no need for action at the EU level. Or perhaps there is a need for an EU act to coordinate at the EU level. Perhaps, further, that in turn may lead to implementing measures at the national and local levels. On the other hand, a UN act may deal with a matter where there is no EU competence. It is more difficult for a process in reverse, where laws at a 'lower', more local and national level, are sought to be incorporated in a 'higher' and more widespread act. Other states must agree, and they may have different ideas. Thus, for example, it can be difficult for an EU member state to seek to have its own forms of law on a matter accepted at the EU level. It is an example of the complex set of relationships that arise between member states, but also between the different orders and legal systems.
We can envisage an example from environmental law, where the EU coordinates with its member states to negotiate international texts (cf. climate change). Here we can imagine a transfer of concepts and terminology 'upwards' to the international world level, possibly with the intention that they will come back down again to EU, at national and local levels, via the implementation of international texts. Each time, and all the time, there is a transfer of ideas, knowledge, concepts, terms, and methods between the levels. Concepts and Terms are units of knowledge, and so it all resembles a giant network of relationships, like a web with 'nodal points' in the form of texts. It reminds one of the Internet and the world-wide web. Terminology work is at the heart of the networks, with databases, collaboration, advice, guidance, and much more.

Viewpoint 12: Online databases of documents and terminology
Lawyers, translators, administrators, and stakeholders all need access to information. The EU institutions provide many sources of information. For example, there is the Official website of the European Union at https://europa.eu in all EU languages. It provides a starting point for researching any EU topic. Each EU institution has its own website. EU legal texts can be found on EUR-Lex at https://eur-lex.europa.eu/homepage.html. EUR-Lex can be used to compare terms in different language versions. EU Terminology can be found on the IATE website at https://iate.europa.eu/home. IATE is used by linguists and translators to store knowledge about terms. The European Parliament has OEIL, a 'Legislative Observatory' on the EU decision-making process at https://oeil.secure. europarl.europa.eu/oeil/home/home.do. This site helps one to follow the legislative process of EU acts, in particular co-decision acts. The EU Court of Justice has CURIA, a website for court cases, procedures, guidance at https://curia.europa.eu/jcms/jcms/j_6/en/. This site helps one in connection with the court and judicial side of EU law. The Publications Office of the EU is the EU official publisher, with its website at https://op.europa.eu/en/home. These are just a few of the sites that are available for information in each language.

Conclusion
There is so much to be said, but we do not have time or space for everything. The 12 Viewpoints merely 'scratch the surface' and provide glimpses on some issues in a general way, to open doors to further and more specific enquiry. The method of adopting viewpoints opens perspectives and possibilities for different ways of looking at, and thinking about, EU law, legal language, and methods. Each viewpoint can be expanded and developed in detail. For example, one can analyse the structure of individual legal acts. One can analyse the terminology in an act for the domains connected to each term. One can explore translation strategies to represent legal information in individual texts. A particular area concerns multilingual drafting and different approaches to it. We can sketch out three attitudes in a few words: Approach 1: draft, finalise a source text, then translate (but problems for other languages?) Approach 2: draft simultaneously in several languages and merge (Canadian method; but EU?) Approach 3: 'co-draft', with translation of each draft, and legal-linguistic alignment of all languages before finalisation, adoption, and publication (EU method…). Each approach has its advantages and disadvantages. One can also reflect on the use of corpora, computer-assisted translation and drafting, and much more. With these words we bring this paper to a conclusion. Sources of additional information available online relating to the themes covered are attached below.