Corps de l’article

Let me start with a sporting analogy. Every two years Ireland and Australia compete in a football match using a set of compromise rules agreed by the Gaelic Athletic Association and the Australian Football League. The resulting sport is a hybrid game containing elements of both national sports (the Australians for example must come to terms with a round ball and the Irish have to adapt to the more robust tackle that is a feature of the Australian game). The making of EU law is also all about compromise. Legal texts are kicked around from one institution to another and are subject to various linguistic inputs which make it impossible to determine what the original text is. Bengoetxea describes the process thus:

According to the Community method for elaborating law, the Commission proposes and the Council and Parliament dispose. It might be the case that the preparatory materials leading to and arising from negotiations carried out by the legislators at the Commission, the Committees of national experts, the Council (and the Committee of Permanent Representatives – COREPER) and the Parliament committees and Plenum, and even the input of the Committee of Regions or the European Economic and Social Committee are themselves the result of the combination of translations going in different directions. By contrast with many policy and consultation documents (for example, Green and White Papers) available normally in English or French, perhaps also occasionally in German, Spanish or Dutch, a draft regulation, directive and a decision of general application initiated in a German or French language draft might receive feedback from the Polish or Portuguese delegations and is translated back with new language additions. The proposal itself might be based on English language drafts elaborated outside the Commission. Which, would we then ask, has been the language of the legislative initiative?

Bengoetxea 2011: 100

Given this peculiarity of the genesis of EU legal texts it is perhaps not surprising that national languages undergo a transformation in the translation process, a bridge between the seemly contradictory goals of ever closer legal integration and the preservation of multilingualism (Baaij 2016). In her groundbreaking study, Biel, drawing on the concept of textual fit,[1] seeks to determine the extent of this transformation in the case of translations of EU law into Polish. Textual fit is a linguistic distance between translations and non-translations of a comparable genre. To measure this distance the author uses the methodologies of corpus linguistics. A number of corpora are analyzed to facilitate this task (JRC EN acquis, JRC PL acquis, pre-accession and post-accession corpora of national Polish law and the National Corpus of Polish (NKJP)). The comparison leads her to conclude:

[…] translated EU law markedly departs from the generic conventions of Polish law, invading its integrity and colonising the genre. As a result, translated law creates a distinct, foreignising, more European variety of Polish (a eurolect), which in addition to being a product of efforts to ensure a uniform interpretation and application of multilingual law, seems be a by-product of the unequal interaction between a majority and minority culture

p. 16

This concept of ‘eurolect’ (coined by analogy with the term ‘sociolect’) has been 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). This aspect of EU legal languages has recently become the subject of increased academic attention. The UNINT project contains Eurolect Observatory for the Interlingual and intralingual analysis of EU legal varieties.  The languages used in the EU sub-corpus are compared with matching national transposition measures.[2]

The divergences between EU legal Polish and the national variety are listed (pp. 289-292). These include a higher mean sentence length, an increased depersonalization through the passive voice and impersonal patterns, a greater variation and instability of terminological equivalents and an overrepresentation of obligation modals (musi, powinien, należy). As a modal, należy combines with an infinitive to form an impersonal agent-defocusing structure, in which an entity on which the obligation is imposed is not mentioned in the subject position or at all (pp. 158-166). This shifts the focus to a verb and its object. Należy is three times more common in regulations and five times more common in directives than in Polish law. Its use differs in and across translations where it functions as a modal accompanied by an infinitive, imposing an obligation, in particular in annexes to directives:

EN Acquis: The tests shall be performed according to the protocols specified by the Member States designated as rapporteur.

PL Acquis: Badania należy przeprowadzić [(one) should perform the tests] zgodnie z protokołami określonymi przez Państwa Członkowskie wyznaczone jako sprawozdawcy

Biel 2014: 344-345

Lawyer-linguists at the EU institutions are instructed to reserve ‘shall’ for the enacting provisions. A variety of other means are called upon to express obligation in the recitals and other parts of EU legal acts (should, is/are to, must, etc.). It has raised interpretation issues in many a jurisdiction. Gémar (1995) discusses its translation into French in legal texts. Should it be viewed through a grammatical lens as an auxiliary which at first glance appears to express the future or alternatively should it be examined in the context of legal discourse as an element of syntax, as part of a sentence?[3]

The divergence between the two varieties is attributed in part to interference of the source language (p. 295). It should be noted here that EU legal English is itself a eurolect. Robertson (2012) suggests that it is a hybrid genre, built on civil law foundations, but displaying some Common Law features including the creation of precedents by the Court of Justice. The lack of experience of often ill-equipped translators is also suggested as a possible cause (pp. 73-75). In this context it is worth noting the debate in Canada over whether legal translation can effectively be carried out by translators who lack legal training.[4] I remember revising an early GA draft of the Common Sales Law which contained a reference to ‘specific performance.’ The translator had failed to recognize that he or she was dealing with a legal term (‘The rendering, as nearly as practicable, of a promised performance through a judgment or decree; specifically, a court-ordered remedy that requires a precise fulfillment of a legal or contractual obligation when monetary damages are inappropriate or inadequate, as when the sale of real estate or a rare article is involved’ (Black’s Law Dictionary)). This resulted in the literal translation of each element of the term (comhlíonadh (performance) sonrach (specific) instead of the correct technical equivalent (‘sainchomhlíonadh’ – which is found in the GA translation of the domestic act - Land And Conveyancing Law Reform Act 2009- http://acts.ie/ga.act.2009.0027.13.html#sec52). I would argue that it is more difficult to make a translator out of a lawyer than a lawyer out of a translator. Biel proposes that translators (experienced and novice) should be made aware of translation universals and that tools for the quality assessment of translations should take account of the parameters of textual fit (p. 308). I would also suggest that the results of this study have implications for EU terminological resources. Meyer (1985, 1997) has shown that translating into a sub-language of one’s maternal language is akin to producing a text in a foreign language and translators in both cases need the same type of collocational and lexical information from their lexicographical resources.

This study is cogently argued and extremely well-written. Typographical errors are scarce (one rare exception is the use of ‘resolution’ for ‘regulation’ on p. 126 and p. 294). Another minor quibble involves the reference to different editions of the EU drafter’s bible, the Joint Practical Guide (the most recent version (2015) is available on-line at http://eur-lex.europa.eu/content/techleg/EN-legislative-drafting-guide.pdf). I would also have liked more data on whether any minor lexicogrammatical changes in the post-accession corpus of Polish law are confined to transposing acts, where the dependence on EU law at a lexical level is expressed by the standard wording: ‘A word or expression that is used in these Regulations and is also used in the Directive has, unless the contrary intention appears, the same meaning in these Regulations as it has in the Directive.’ The great merit of this work is that it extends the methodology of corpus linguistics to the field of legal translation. At a time when the linguistic regime of the EU is under even closer scrutiny with the imminent departure of the UK from from the Union, this makes this fine contribution to the discipline of jurilinguistics even more welcome.