By Rioghnach Theakston
With 24 official languages, integrated into 27 Member State legal systems, the EU legal order is both linguistically and substantively unique. Past legal and political science research has tended to treat the Court of Justice of the European Union (CJEU) as a single actor, focusing on the judgements produced by the court, as opposed to examining what happens behind the scenes in the production of the CJEU’s case law. Of course, the substance of the case law produced by this institution is significant in its own right, but this approach risks glossing over the linguistic diversity of the CJEU’s judges, Advocates General (AGs), lawyer-linguists and référendaires (legal assistants) – a potentially critical omission, given the nature of law as a linguistic construct.
When a case is brought before the European Court of Justice (ECJ), it undergoes a complex, multi-layered, multilingual process of translations before a final judgement is given. The ECJ has a single internal working language: French. Every action brought before the Court has a language of procedure, which can be any one of the 24 official languages. As French is the working language, applications that come into the Court in other official languages are immediately translated into French. The application is then allocated to a reporting judge, who is responsible for drafting a preliminary report. The Court also decides whether there will be an oral hearing and an AG’s opinion. If an AG’s opinion is included, the judge rapporteur and their référendaire(s) must wait for that opinion before beginning to draft the substance of the judgement in French. The AG’s opinion must also be translated into French. After the first version of the judgement has been drafted, secret deliberations among the judges take place, leading up to the final judgement (also in French). The final judgement is then translated into the language of procedure, and ultimately all other official languages, so that the case law can be disseminated across 27 Member States. 
This has important implications. Firstly, the final version of the judgement, as deliberated on by judges in their secret deliberations, is generally not the authentic version of the judgement. That is to say, the version in the language of procedure is usually a translation of the version signed by the judges. More generally, ECJ judgements are multi-authored collegiate documents that undergo several permutations of translation, each of which adds a new layer of complexity – increasing the risk of legal inconsistency with every step. All in all, this unique method of case law production can have implications for the rule of law within the EU, in particular regarding the uniform application of that law in 24 languages, across 27 different legal orders.
On the individual level, although the members of staff drafting the ECJ’s case law are working in French, French is rarely their mother tongue.  This has important implications for the language that develops within the institution, as well as its legal output. In the case of référendaires, whilst they must work wholly in French, they are not required to have a perfect command of that language; in most cases, it is not their mother tongue. In addition to being reined in by the legal reasoning embedded in the language they are working in, référendaires are also constrained by the CJEU’s variety of rules and conventions around language use, which foster a formulaic, repetitive style of writing. Whilst the formulaic language used in CJEU judgements can act as a safety net for référendaires, this environment also fosters repetition and ‘cutting and pasting’, such that the style of writing judgements reproduces itself. This in turn leads to the construction of self-perpetuating, de facto linguistic precedent in ECJ case law.  Overall, these factors have encouraged the emergence of a “Court French” within the institution, which differs significantly from “normal” French in grammar and syntax. “Court French”, combined with complex layers of translation and authorship, necessarily shapes the case law produced by the ECJ. 
Lawyer-linguists (those responsible for translating documents into the working language of the Court) can also face extensive problems given that they have to balance a dual professional identity of lawyer and linguist. Consequently, they are an embodiment of the necessary compromise inherent in reconciling the notions of law and translation reflected in the process of disseminating the CJEU’s case law to the wider EU.  The lawyer-linguist’s role is complicated further by the fact that translations are approximate by nature. Yet, in a legal order founded on the principle of uniformity of law throughout 27 member states, lawyer-linguists must ensure that the Court’s case law remains consistent across 24 language versions. Thus, they are responsible for legal issues that may arise because of linguistic ambiguities in texts – a task made more difficult with each additional layer of translation. This opens up the risk of case law being legally inconsistent in a minority of cases. Indeed, in previous case studies, lawyer-linguists have agreed that the process of producing case law at the CJEU could lead to problems of a legal nature, including problems at the municipal level due to mistranslation. 
The challenges of working in a language that is not your mother tongue are also reflected in the work of AGs. AGs deliver independent opinions on certain cases that are brought to the ECJ. Whilst their opinions are not legally binding, AGs provide a ‘dissenting judgement’ that encourages the growth of EU law. AGs can be speculative in a way that the ECJ cannot be in its collegiate judgments, so their opinions infuse creativity into EU jurisprudence. Their role is an important one in terms of ‘droit prospectif’, that is, indicating the likely directions for the development of EU law. 
Historically, AGs drafted their opinions in their mother tongues, so issues of multilingualism and translation rarely constrained the content or style of their opinions. However, this changed as a result of the EU’s 2004 mega-enlargement. This wave of EU expansion resulted in the introduction of 10 new official languages, increasing to 13 by 2013 and bringing the total from 11 to 24. This would have dramatically increased the workload for the translation service. Even though the Court’s translation service is its largest department – making up almost half of its total staff – the resulting 552 language combinations would have been simply untenable. Hence, from May 2004 onwards, a mixed translation system was put in place to reduce pressure on the translation service. Although direct translation is used whenever possible, given the workload, translation through a ‘pivot language’ (French, English, German, Spanish, Italian) is the norm.  For AGs, this has meant that, since 2004, some have been writing their opinions in languages other than their mother tongue.
Given the importance of AG opinions for the development of EU law, through critical assessment and dialogue with the Court, whether these linguistic changes have affected AGs is an important question. Indeed, research has found that the opinions of AGs have been affected – stylistically and possibly substantively.  One emerging tendency is that AG opinions are becoming more similar linguistically to ECJ judgements, which are written in a hybrid, formulaic legal language and built up like “Lego building blocks”. Opinions drafted after 2004 by AGs in a language other than their mother tongue include more markers for translation variables and are stylistically simpler and less fluent than the ones drafted by AGs in their mother tongues prior to 2004. Interestingly, the changes have also made AG opinions drafted after 2004 in mother tongue languages simpler and less fluent. This suggests that the CJEU’s institutional language is influencing its jurisprudence, regardless of drafting language. 
How we evaluate the practical impact of these changes depends on where we believe the purpose and value of AG opinions lie. On the one hand, if their main purpose is to persuade the ECJ to re-evaluate aspects of EU law, AG opinions becoming linguistically more like judgements could be beneficial; it results in both parties speaking the same institutional language. On the other hand, if they exist to provide additional information for interested outsiders, retaining their discursive, academic style may be important. Having an accessible resource for interested parties may be particularly relevant in this context, as it can compensate for the lack of transparency in producing ECJ judgements.  All in all, there is a consensus that the value of AGs’ opinions lies in their persuasiveness.  Language is, of course, a fundamental part of the ability to persuade. Hence, this raises the question as to what extent the language reform in the ECJ may have had an impact on the substance, or persuasiveness, of AG opinions. Since AG opinions have a formative influence on the ECJ’s case law, it is worth considering the implications for the development, formation and application of EU law, if the creativity of these opinions is limited by linguistic constraints.
All in all, questions about the purpose of AG opinions must be addressed, if we are to understand how changes in the ECJ’s internal linguistic regime affect the development of EU law. Although AGs represent only a part of the institution of the ECJ, their opinions feed into the multi-layered, multilingual process of producing the ECJ’s case law. In fact, AG opinions reflect that process in their own construction. Hence, the impact of new linguistic conventions on AG opinions highlights the importance of developing a coherent account of the wider, behind-the-scenes process of producing the ECJ’s multilingual jurisprudence, if we are to develop a more nuanced understanding of EU law and the inconsistencies that inevitably arise in this evolving, multilingual system.
 Behind the Scenes at the European Court of Justice: Drafting EU Law Stories, Karen McAuliffe, 1 Jun 2017, EU Law Stories: Contextual and Critical Histories of European Jurisprudence. Nicola, F. & Davies, B. (eds.). Cambridge University Press, p. 35-58 23 p. 3. (Law in Context).
 Albertina Albors-Llorens, Securing trust in the Court of Justice of the EU: the influence of the advocate general, 14 Cambridge Y.B. Eur. Legal Stud. 515 (2011-2012); Rosa Greaves, A Commentary on Selected Opinions of Advocate General Jacobs, 29 Fordham Int’l L.J. 715 (2005); Alicia Hinarejos, Social Legitimacy and the Court of Justice of the EU: Some Reflections on the Role of the Advocate General, 14 Cambridge Y.B. Eur. Legal Stud. 631, (2012)