berica 13 IBÉRICA 16 [2008]: 129-146 Abstract Contrastive studies of statutory legislation are very scarce world-wide. Research in legal language has mainly concentrated on adjectival law leading to linguistic insights regarding powerful versus powerless speech, fragmented versus narrative testimony, the effect on jurors of simultaneous and overlapping speech, the use of leading, suggestive or improper questions in the courtroom, etc. Language of the substantive law has so far received considerably less attention from linguists, although there is a general tendency in academic endeavours towards interdisciplinary studies. Linguistic analyses of substantive law have elucidated issues such as how to make existing or future statutes clearer, without loss of content (i.e. document design) or how law students can be taught to process legislation. The present article analyses the specific functional, linguistic and communicative characteristics of the legal genre from an applied linguist’s perspective in the context of European legal texts, as representing a unique set of features and conditions. It looks at the linguistic situation in Europe and the language policy in the EU with special emphasis on the translation regime of EU institutions. The participants in the communication and the special role of the translator in the law making process in the EU are discussed. Key words: statutory language, legal translation, European legislation, supranational law. Resumen Texto y contexto de las Directivas de la Uni�n Europea: implicaciones para los traductores Los estudios contrastivos sobre legislación parlamentaria a nivel transnacional son muy escasos. Hasta ahora, el interés investigador se ha centrado en cuestiones de derecho adjetivo, como el estudio del discurso de poder, las deposiciones testimoniales fragmentadas y las narrativas, la influencia sobre el The text and context of EU directives: implications for translators1 Diana Yankova New Bulgarian University (Bulgaria) dyankova@nbu.bg 129 07 IBERICA 16.qxp 4/10/08 18:24 Página 129 jurado del discurso simultáneo o solapado fonéticamente, el uso de preguntas incorrectas o sesgadas en los tribunales, y así sucesivamente. El lenguaje del derecho sustantivo, por el contrario, ha recibido mucha menos atención por parte de los lingüistas, aunque comience a aparecer una corriente académica favorable a los estudios interdisciplinares. Los análisis lingüísticos del derecho sustantivo se han encaminado hacia la clarificación y simplificación del lenguaje parlamentario, sin que éste pierda contenido, o alternativamente a la didáctica de la interpretación legislativa a los juristas. Este artículo pretende analizar las características funcionales, lingüísticas y comunicativas únicas y peculiares a los géneros legales desde el punto de vista de la Lingüística Aplicada al contexto de la legislación europea, al mismo tiempo que se centra en la situación europea en lo relativo a política lingüística, con un énfasis especial en el sistema traductológico de las instituciones europeas. Asimismo, se debate sobre los participantes del proceso de elaboración de las leyes en la Unión, y el papel del traductor en dicho proceso. Palabras clave: lenguaje legislativo, traducción jurídica, legislación europea, derecho supranacional. Introduction There have been few or no substantial studies of the important linguistic elements in EU law and their implications for the understanding and application of the law; language and communication issues have not been informed by scholarly discussion (Tosi, 2005; McAuliffe, 2006). The idiosyncratic communicative situation within which EU legal texts are created has an immediate bearing on the texts produced. Within this context, the concepts of text type, original text, translation, text producer, text recipient, etc., acquire new meaning and merit new interpretation. One motivation for analyzing EU directives is the role these instruments play in the process of approximation of legislation and the purpose they fulfill. Directives occupy a special place among statutory texts since they are binding with regard to the results to be achieved, but not to the exact methods of achieving these results. It is left to the discretion of each Member State to choose the best designed and most effective procedures and practices with respect to its specific context. Member States are encouraged to transpose directives correctly and on time, and are also monitored and penalised for incorrect or late transposition, but the EU Commission can only offer recommendations as to the forms and means that Member States adopt in implementing a particular directive. DIANA YANKOVA IBÉRICA 16 [2008]: 129-146130 07 IBERICA 16.qxp 4/10/08 18:24 Página 130 The translation of EU-produced legal texts already presents multifarious problems for translators from Member States, which are voiced in private, as well as at seminars of the Translation Service and in the on-going “fight the fog”2 campaign (Wagner & Martin, 1998; Wagner, 2000a). Member States and countries seeking accession to the European Union have to harmonize their existing institutions, create new ones and find the language to communicate adequately within unified Europe which implies the standardization of legal terminology –a laborious and demanding task. This article is part of a larger study of the contextual and textual characteristics of EU legislation focusing on the broader context of the production and consumption of legal texts, with special emphasis on translators who are an essential part of harmonization of legislation owing to the specific supranational and multilingual nature of law creation in the EU. The law making process in the EU Apart from cases where one single national jurisdiction is expressed in one language, throughout the world there exist several other basic legal contexts; namely, different legal systems are expressed in one language (e.g., the English and Scottish system); one and the same legal system is expressed in different languages (e.g., the Swiss system and partly the Canadian system); the third option is the bilingual or monolingual system of public international law. Unlike other international organizations whose resolutions are directed to governments only, the EU is the only international body that passes directly binding laws to the citizens in its Member States and this is precisely what frames EU’s language policy, embodying the aspiration that all citizens should have equal treatment regardless of their mother tongue. It is necessary for such texts to be available in the national languages of the countries concerned. The legislative initiative within the EU lies with the European Commission which, in the case of regulations and directives, drafts proposals that are adopted by the two decision-making institutions, the Parliament and the Council, before the law becomes implemented in national law. This is known as co-decision and is the main legislative procedure, requiring that all legislative proposals with the necessary translations get approved several times in the three institutions: the Commission, Council and Parliament. THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 131 07 IBERICA 16.qxp 4/10/08 18:24 Página 131 The Acquis Communautaire refers to the collection of EU legislation and encompasses primary legislation (the Treaties), secondary legislation (deriving from the Treaties) and case law of the Court of Justice. There are five types of secondary EU instruments: regulations, directives, decisions, recommendations, and opinions, of which the first two are the most important and most recurrent. It is essential for the translator to be aware of the parts comprising each instrument, the functions they perform, and whether and to what extent the instrument is binding. For instance, regulations are totally binding, while directives are binding in regard to the results to be achieved, but not to the exact methods of attaining these results, which is left to the discretion of each Member State; decisions are binding on those to whom they are addressed, while recommendations and opinions have no binding force. EU instruments, like any other statutory texts, are formulaic. The arrangement is the following: they begin with the title, the name of the enacting authority (the Commission, the Council, the Parliament), followed by the citations which set out the legal basis, then by the recitals laying down policy considerations, and finally, the Articles, which provide the substantive provisions. Since most of the information is clustered together in the introductory section and is presented as a long list of cases and conditions to which the following articles apply, it presents difficulties when adapted into the very different structure of national statutes. This formal complication, added to the existing conceptual non-correspondence in certain legal institutions between national legislations and the Acquis Communautaire, renders the whole process of harmonization somewhat intricate and problematic. More often than not, directives are lexically abstract because of their specific position as legislation to be adopted by the diverse legal systems of the member states. Consequently, the respective national drafters are reluctant to alter the wording when implementing the directives, lest a misunderstanding might arise of the original legislative intent. In so doing, the national drafter runs the risk of making the meaning of the text hard to comprehend by the public at large, since at times it diverges from the respective national language of the law, and does not abide by its generic conventions. The process of drafting EU legislation has been criticized by a number of academics and practitioners, most notably by Braselman (1992), who questioned the possibility of all language versions being equally authentic on the grounds that there can be, in some cases, only partial equivalence, since different languages conceptualize the same situation differently, and this DIANA YANKOVA IBÉRICA 16 [2008]: 129-146132 07 IBERICA 16.qxp 4/10/08 18:24 Página 132 leads to interpretation problems when no original version exists. Along the same lines, Solan (2006) contends that the Babel of Europe facilitates communication. According to him, a supranational legal order should meet three goals: show respect for individual members, be faithful to the intent of the drafters, and be efficient. The absence of a single authoritative text and the usage of many language versions increase the likelihood for judges to pay closer attention to the intent of the legislature and to provide information which is not found in monolingual translation. Mezzarese (2000: 162) offers a similar standpoint: a close inspection of some of the main sorts of problems met by legal translation (i.e. by the inter-lingual translation of legal language from one natural language to another) can enlighten, and lead to a better comprehension of some of the classical topics debated in dealing with legal interpretation (i.e., with the sort of intra-lingual translation of legal language in the scope of one and the same natural language). It remains to be seen if the unprecedented multitude of languages proves a hindrance or enhancement in the application of EU law. It is however striking, that so far there have been remarkably few law suits involving disputes over different language versions of official EU documents. Institutional EU English Institutional EU English is considered something different from “standard English” and this is determined by several factors. Sometimes legislation in English is drafted by non-native speakers and communication within the European institutions mirrors the situation worldwide. As Flesch (1998: 3) states “there are more non-native English speakers in the European institutions that communicate through English than British and Irish officials combined”. Being supranational, EU legislation applies regulations and directives not existent in national legal systems. The process of drafting also involves considerations regarding the exact wording of the documents, which have to be translated into all the 23 official languages. Source texts presented to translators have been considered by some as being of a “sub- standard quality” (Martin, 2000: 2). Wagner expresses the view that “written communication in the Commission is too much, too unclear and too out of touch” (Wagner, 2000a: 6). Horst (2000: 2) has even gone so far as to say that: THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 133 07 IBERICA 16.qxp 4/10/08 18:24 Página 133 as long as the European public has the impression that the language used for the process of European unification is an administrative language, foreign to everyday language, a sense of belonging will not emerge and the idea of the European unity will be rejected. Eurojargon has few archaisms, borrowings or unnecessary technical vocabulary, and lacks any of the typical convoluted characteristics of legalese. What hinders communication is the extreme textual uniformity. In Tosi’s (2005: 385) view lexis and syntax are clear enough, and yet lexical vagueness and weak logical connections spread a sense of mechanistic virtuality that makes the voice of Europe sound awkward, abstract and completely distant from any language spoken in everyday life. The result is that non-standard syntactic structures become acceptable forms of language use. Concerning vocabulary, there are some peculiar lexical combinations, for example “Member State” for a country that has joined the EU, or the metonymic use of proper nouns, such as Maastricht and Schengen. Another important point to consider is the standardization of texts produced in the Union. All the different versions have to be uniform not only regarding the content, but also regarding the organization of the text. The layout, articles, paragraphs, sentences have to match completely in order to facilitate reference to the document in any of the official languages. The full stop rule requires “an equal number of full stops in source text and translations” (Trosborg, 1997a: 152). All language versions of drafted documents have to be identical, paragraph by paragraph and, where possible, sentence by sentence. This requirement is tied to the legislative process itself within this supranational environment. Hence, regulations and directives can be easily modified. The equal value of all texts is a legal fiction3 that is at the core of EU multilingual context; all language versions are functionally equivalent and are binding on citizens in the respective Member State. At present, there is an atmosphere of rethinking and reconceptualizing of EU practices both in terms of institutional organization and legal drafting, which would undoubtedly have an impact on translation. The Chairman of the European Union’s constitutional convention, Valery Giscard d’Estaing is in favour of plowing through the hundreds of pages of treaties and agreements that constitute the foundation for the EU, bringing that paperwork to some “30 or 35 pages in all” –something that’s at once DIANA YANKOVA IBÉRICA 16 [2008]: 129-146134 07 IBERICA 16.qxp 4/10/08 18:24 Página 134 comprehensive and comprehensible, not to mention digestible, just like “the other great constitutions” of the world (Dickey & Meyer, 2002: 14). Likewise, the Translation Service has for years appealed for a higher standard of drafting and clearer texts, including application of what is known as the KISS rule (“Keep it Short and Simple”) and the “Fight the Fog” campaign. The translation regime of EU institutions Each EU institution has its own translation service which is responsible for translation of all the texts issued by such an institution: the Commission, the Council, the Parliament, the Court of Justice, the Court of Auditors, the European Economic and Social Committee and the Committee of the Regions. By far, the largest translation service is the Commission’s Directorate-General for Translation (DGT) comprising about 2,300 people translating all of the written communication. It is located both in Brussels and Luxembourg. Other EU institutions such as the Council, Parliament, Court of Justice, Court of Auditors, etc., have their own translation departments. The ratio of translators to interpreters at the DGT is 3:1. It is the largest public language translation bureau in the world. There is one Language Department for each official EU language and translators work in single-language units specializing in different subjects. EU translations encompass a wide range of documents on a large variety of topics: Council regulations and directives to be published in the Official Journal, minutes of Council meetings, committee reports, opinions of the Council legal service, press releases, letters and speeches, amounting to more than 3 million pages per year. In the case of international bodies (Schäffner, 2002) a distinction is made between texts for internal use –i.e., texts that are used by politicians and negotiators among themselves, and those for external use –i.e., texts that are directed to the public at large (see Koskinen (2004) for discussion of EU culture). The texts that are translated into all the official languages fall into the following groups: • documents that are essential in the final stages of the decision- making process; • all texts that are for adoption by the Council; • documents that are of general interest for the citizens of the Member States. THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 135 07 IBERICA 16.qxp 4/10/08 18:24 Página 135 In all other instances, mainly at the intra-institutional level, functional and pragmatic considerations are operative and this means less effort without loss of transparency or efficiency. This includes the daily administrative work of the institutions and the initial stages of legislative drafting which are drafted by in-house officials in one or two working languages: mostly English and French. The same principle applies in the case of interpreting, and multilingualism is ensured where it is required: sittings of the European Parliament and its committees, as well as Council of Ministers sessions are provided with interpretation in all the official languages, while at administrative meetings or discussions between specialists in SIGs (Special Interest Groups) do not require full interpretation and working languages are resorted to. Interpreting usually faces more problems, since the number of actually used language combinations is higher. At a meeting, for instance, any delegate can make a speech which has to be rendered in all the languages. When documents are drafted in English, they are more likely to be understood by everybody and will not necessitate a translation. The outgoing documents drafted mainly in English or French within the Commission are then translated by native speakers in the other official languages. Translators translate out of several languages but as a rule into their mother tongue. Incoming documents used to be translated exclusively by French and English translators but with the current number of official languages, it is very difficult for the French and English Language Departments to deal with all the translations as a whole. Consequently, there is a tendency now to use return or two-way translation by translators whose native language is other than English or French but who have an excellent command of those languages. Another recent development is the introduction of editing services for French and English to assure the linguistic quality and the translatability of those documents which are mostly drafted by non-native speakers.4 A country entering the EU has to translate the Acquis Communuataire comprising about 90,000 pages of primary legislation and all secondary legislation passed up to the day the country joins. This responsibility lies with the national government, while EU institutions finalize and publish the translated texts in the Official Journal of the European Union. To further illustrate the matter, by the end of April 2006, DGT had recruited 473 full- time translators.5 There are also additional translators recruited as seconded national experts for fixed periods and a projected support staff of 126. DIANA YANKOVA IBÉRICA 16 [2008]: 129-146136 07 IBERICA 16.qxp 4/10/08 18:24 Página 136 DGT also outsources work to freelance translators. Approximately every eighth official working in the EU institutions is a translator (Toscani, 2002). The total translation output of the DGT in 2005 was 1,324,231 pages in all (then) 20 official languages. In preparation for the biggest enlargement of the EU, which took place in 2004 when 10 countries joined with 9 new languages, the Commission introduced a demand-management strategy to improve communication with the citizens whereby departments were required to produce documents not longer than 15 pages (compared to the pre-accession average of 37 pages). With a view to the smooth integration of the Bulgarian and Romanian languages in 2007, the DGT set up a Task Force for “Enlargement 2007”, opening Field Offices in the Commission Delegations in Bucharest and Sofia in 2005. The first 16 Bulgarian and 20 Romanian translators were recruited in January 2006, with the final figure to be 60 translators per language. The written tests of the translators were held in July and were marked in September, the author being one of the markers. To date, about 70% of the documents forming the Acquis has been finalized in Bulgarian and Romanian. Another language that was granted full official status on January 1, 2007 is Irish. However, due to a derogation6 such as the one requested by the Maltese government, only the regulations adopted jointly by the European Parliament and the Council, together with the correspondence between EU institutions and the Irish citizens will be translated. In December 2005, the Commission signed an agreement with Spain on the use of Basque, Catalan and Galician, based on decisions adopted by the Council in June 2005 authorizing the use in the EU institutions of regional languages with no official status in the respective Member State. They now have an intermediary status, whereby speakers of those languages have the right to communicate with European institutions in these languages, but the translation costs are covered by the Spanish government. The cost of translation in absolute values is shown in Table 1. With the inclusion of Bulgarian, Irish and Romanian, the cost of translation is projected to rise by approximately 30 million Euros for all institutions. Year Population Cost per year Cost to each citizen THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 137 DIANA YANKOVA such as the one requested by the Maltese government, only the regulations adopted jointly by the European Parliament and the Council, together with the correspondence between EU institutions and the Irish citizens will be translated. In December 2005, the Commission signed an agreement with Spain on the use of Basque, Catalan and Galician, based on decisions adopted by the Council in June 2005 authorizing the use in the EU institutions of regional languages with no official status in the respective Member State. They now have an intermediary status, whereby speakers of those languages have the right to communicate with European institutions in these languages, but the translation costs are covered by the Spanish government. The cost of the translation in absolute values is shown in Table 1. With the inclusion of Bulgarian, Irish and Romanian, the cost of translation is projected to rise by approximately 30 million Euros for all institutions. Year Population Cost per year Cost to each citizen 2003 379 million EUR 541 million (of which DGT 230 million) EUR 1.45 (of which DGT 0.60) 2006 453 million EUR 800 million (of which DGT 300 million) EUR 1.76 (of which DGT 0.66) Table 1. Cost of translation in absolute values. EU translation tools The translation tools that DGT has at the disposal of the translators are of three main types: terminology tools, translation memory technology, and machine translation. An example of a terminology tool is EURODICAUTOM –the central terminological bank at the European Commission which is also the largest terminology database in the world. An abbreviation for Europe dictionnaire 07 IBERICA 16.qxp 4/10/08 18:24 Página 137 EU translation tools The translation tools that DGT has at the disposal of translators are of three main types: terminology tools, translation memory technology, and machine translation. An example of a terminology tool is EURODICAUTOM –the central terminological bank at the European Commission which is also the largest terminology database in the world. An abbreviation for Europe dictionnaire automatisé, the bank is open to the general public as well, containing more than 6,500,000 terms and 300,000 abbreviations operating in all the EU’s official languages and Latin. It is soon to be replaced by IATE, an interinstitutional terminology database. TRADOS Translator’s Workbench (TWB) is an example of the second type of tool: translation memory technology. It is an integrated translation support tool with multilingual capability, a text recognition and replacement programme enabling translators to incorporate phrases, sentences, paragraphs translated before, checking for matching units and particularly useful for texts in accession negotiations. When an original text is fed, similar or identical fragments from previously translated texts appear as translation suggestions. An instance of the third type of tool (providing machine translation) is EC Systran, short for “System Translation”. It can translate up to 2,000 pages per hour, thus offering fast access to information in languages which requesters do not know, the main use of the tool being the standard translation of a document. In the latter case, the rough machine output must be edited and the amount of effort will depend on the text type: letter, minutes, manual, etc. In 2004, the total sum of pages translated by EC Systran amounted to 696,347, with the Commission being the main requester with 556,138 pages, a third of which (181,060 pages) was requested by the DGT. At present, the tool provides translation for 18 language pairs: from English into Dutch, French, German, Greek, Italian, Portuguese and Spanish; from French into Dutch, English, German, Italian, Portuguese and Spanish; from German into English and French; from Spanish into English and French; and, finally, from Greek into French. There are also administration and documentation tools that help the process of translation. Vista is the DGT’s electronic archiving system, containing all incoming (mainly original) and outgoing (mainly translated) documents in the Commission since 1 January 1994, providing the rapid retrieval of any document. Eur-Lex contains the primary and secondary legislation and preparatory acts in all official EU languages, as well as national implementing measures and case law of the European Court of Justice. It is open to the DIANA YANKOVA IBÉRICA 16 [2008]: 129-146138 07 IBERICA 16.qxp 4/10/08 18:24 Página 138 general public and allows the possibility to consult the Official Journal of the European Union. All these tools have been developed on the basis of the specific needs of the translator within the EU environment. In their daily work, translators need the appropriate terminology in the form of glossaries, databases, or dictionaries; they need reference documents such as electronic archives, and the possibility to resort to previously translated texts. The participants in the communication and the role of the translator The addresser of a piece of legislation is not an individual, but a team of specialists who write the text together in different stages before a final draft is approved in order to become law (for a detailed description of the process, see Yankova & Vassileva, 2002). The exceptional communicative situation in the creation and consumption of texts within the European Union finds expression in the character of the participants in the process. In the context of supranational law, legislation is produced in a long process of draft- making, revisions and modifications within all the language versions. There is constant consultation and cooperation between text originators, legal experts, translators and revisers. The individual and independent voice or imprint is completely lost within this multi-authored prose: “the co-decision procedure entails at least 31 steps by 11 different services in the three main institutions and four of these involve the European Commission’s Translation Service” (Wagner, 2000b: 4). Any model of communication can also be viewed as a model for translation but the difference between a monolingual communicator and a translator is in the encoding of the message. The translator re-encodes a message into a different language, the message is approximately the same as the one received and is aimed at receivers who are not the same as the original sender. Therefore, the translator is neither the sender nor the addressee proper or the intended recipient. One of the most influential theories of text comprehension is Van Dijk and Kintsch’s (1983). It describes the comprehension process starting from recognizing words until constructing a representation of the meaning of the text. In 1988, the theory was extended with the construction-integration model (Kintsch, 1988), followed by a completely updated theory in 1998 (Kintsch, 1998). Kintsch’s (1988) model of the comprehension process encompasses the following: Step A – text- THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 139 07 IBERICA 16.qxp 4/10/08 18:24 Página 139 based analysis; Step B – knowledge-based analysis; Step C – gap-bridging; and Step D – coordination.7 When a recipient reads or hears a text, an understanding of that very text is created in his mind. The process of constructing a situation model is called the “comprehension process”. Texts are interpreted word by word, and meanings of smaller units are combined into propositions. Recipients build three different mental representations of the text: a verbatim representation of the text, a semantic representation which describes the meaning of the text, and a situational representation of the situation to which the text refers. The propositional (verbatim) representation consists initially of a list of propositions that are derived from the text. This list is transformed into a network of propositions, until the short term memory buffer is full. These bigger chunks are stored in the textual representation in episodic memory and language users activate an old or construct a new mental model of the events or situation the text is about. At each level, there is a microstructure, or propositions for the local content, and the macrostructure, or propositions representing the global structure of the text. The end product of the construction-integration process is an episodic text memory consisting of a text base and a situation model. Of the two, the situation model is harder to specify because it depends on the individual recipient’s knowledge and experience. The translator’s approach will be to construct a model with a very high degree of approximation. The level of expertise of the translator is of utmost importance in the construction of such a model –as is the text itself– and the language difficulties that the translator is faced with. Following Newmark (1986), the translator’s communicative competence consists of four areas of knowledge and skills: • grammatical competence: the knowledge and skills to understand and express the literal meaning of utterances; • sociolinguistic competence: knowledge of and ability to produce and understand utterances appropriately in context, i.e. as constrained by topic, status of participants, purpose of interaction, etc.; • discourse competence: the ability to combine form and meaning to achieve unified texts (cohesion in form and coherence in meaning); • strategic competence: communication strategies which may improve communication or compensate breakdowns. DIANA YANKOVA IBÉRICA 16 [2008]: 129-146140 07 IBERICA 16.qxp 4/10/08 18:24 Página 140 In Nida’s view (1975), the comprehension of a text depends on the level of expertise of the receiver. A properly encoded message has a specific length and a certain degree of difficulty which corresponds to the channel capacity of the intended recipient. A literal translation of the message will invariably produce a more difficult text of the same length, while the secondary recipients of the text have a narrower channel capacity. The solution he suggests is to supply the necessary redundancy in the translation, in terms of expressly stating elements which are covert in the original text, and to provide background information necessary for the proper unpacking of the text. According to Alexieva (1999: 57), the ease with which we comprehend a text depends to a large extent on the following: the text’s semantic density or the degree of condensation of the predications forming its content structure, in terms of the ratio between the explicit predications (rendered by means of finite clauses) and the implicit predications (given in a condensed form by means of nominal, participial or infinitival phrases). The type of text and its functional content is a very important parameter in its proper comprehension. The functional content of a text can be described in terms of the following features (Alexieva, 1994): dominant functions and types of meaning; density of information; type of cognitive models employed in the making of the text. The main function of statutory texts is to impose obligations and to grant rights. They are directive texts, which in Jakobson’s (1960) model have a conative function, and in Halliday’s (1994) theory an interpersonal function. Regarding density of information, due to the double need of statutory texts to be both concise and at the same time abstract, in the sense of wide extension (Gunnarsson, 1987) (or all inclusive in Bhatia’s (1993) terms), we can discern text characteristics which are at the opposite end of the scale, and such that render text comprehension exceptionally difficult. One of the features is the extreme density of lexical repetition, striving at the utmost preciseness of expression (Danet, 1980; Bhatia, 1993). Due to the repetitive nature and lack of anaphora of legal texts at large, statutory texts are difficult to unpack semantically at first reading. The other characteristic is flexibility, and can be seen in the drafting of a vague and abstract language which aspires to include all possible contingencies, in the implicitness and semantic density that is created by the desire to be overly compact. This feature finds THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 141 07 IBERICA 16.qxp 4/10/08 18:24 Página 141 expression mainly in nominalizations and passives (Maley, 1994; Tiersma, 1999). In the process of interlingual transfer between the intentions of the addresser and the effect of the translation on the addressee, translators are not encouraged to be real mediators. The standardization of translation within the EU context calls for a linear substitution of lexical and syntactic strings and the direct transfer of punctuation from one language version to another. The strategies employed by EU translators in the treatment of parallel texts are text-focused, rather than recipient-focused. In view of the intricate character of the production of these special types of text, translators do not have individual responsibility, but collective responsibility, which is at the level of the institution itself. Moreover, the text that, for example, German translators ultimately endeavor to translate, may not be the original (French) version, but it can be a translation (Spanish) which functions as a link between the source and the target language. At times, separate sections of the same text are translated by different translators from the same source text; at others, different translators translate the same text from different source languages (Tosi, 2005). EU statutory instruments are influenced by the specific EU context, and all these factors are embodied in the discoursal features of the texts produced within the supranational EU context: there are few archaisms, borrowings and technical vocabulary; texts are replete with non-standard syntactic structures and in general display extreme textual uniformity. In monolingual statutory interpretation, the challenge is to interpret a word or a phrase expressed in one language and embedded in a particular context. In the case of EU legislation, the interpretation of words or phrases is aggravated by the number of languages involved; having in mind that meaning is distributed differently in the diverse language systems. Therefore, it is very difficult to preserve both the denotative and connotative meaning of words and phrases in all the 23 languages. Conclusion How can translators deal with the novel aspects of the communicative situation within the EU context? In order to understand and properly process the text under Kintsch’s (1988) model, legal translators should be skilled in the analytic and empirical methods of comparative law, in addition to having broad knowledge of both the source and target legal system. In DIANA YANKOVA IBÉRICA 16 [2008]: 129-146142 07 IBERICA 16.qxp 4/10/08 18:24 Página 142 most cases they cannot rely on bilingual or multilingual law dictionaries, but need to consult the original sources of law. Within the EU context, there are no guides for strategic choices of the translators apart from guidelines for EU terminology, but the collective and intertextual nature of EU translations ensures no deviations from the general trend. Translators, however, are not free to use just any strategy; they are faced with institutional constraints, which inevitably affect their work, on top of the fact that as a rule, statutory texts call for an exact and faithful translation in order to preserve the intent of the legislative institution. The cultural context within which a text is consumed is also of importance. Currently, European languages show a tolerance for foreign elements owing to the large number of translations that are being done in relation to the ongoing social, political, economic and legal integration in Europe. Therefore, the prevalent general translation strategy is oriented towards acceptance and adoption of internationalisms and neologisms. In specific cases, a felicitous approach is for the translator to construe a term from the context and the dictionary entry, then make certain that it is in compliance with the legislative intent of the source legislation and that it corresponds to the source term in its denotative and connotative meaning, and finally to make sure that the term can be applied consistently in all parts of the target statutory instrument. It is important to keep in mind that translations which are part of the harmonization process become integrated in the national legislation of the individual country and therefore, the choices that are made will become obligatory in future translations of EU legislation. Inevitably, these choices will leave an imprint on the evolution of the national language and will influence the discourse patterns in the receiving culture particularly in varieties of language that are being freshly introduced, such as the new, supranational European legal discourse. (Revised paper received February 2008) References THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 143 Alexieva, B. (1994). “Types of texts and intertextuality on si- multaneous interpreting” in M. Snell-Hornby, F. Pöchhacker & K. Kaindl (eds.), Translation Studies. An Interdiscipline, 179- 187. Amsterdam: John Ben- jamins. Alexieva, B. (1999). “Under- standing the source language text in simultaneous interpreting” in A. Riccardi & M. Viezzi (eds.), The Interpreters’ Newsletter, 45- 59. Trieste: Universita degli Stu- di di Trieste. 07 IBERICA 16.qxp 4/10/08 18:24 Página 143 DIANA YANKOVA IBÉRICA 16 [2008]: 129-146144 Bhatia, V.K. (1993). Analysing Genre: Language Use in Profes- sional Settings. Longman: Lon- don. Braselman, P. (1992). “Der Richter als Linguist. Linguistis- che Uberlegungen zu Sprach- problemen in urteilen des eu- ropaischen Gerichtshofes”, Sprache und Literatur in Wis- senschaft und Ubterricht 68: 71- 81. Danet, B. (1980). “Language in the legal process”. Law and So- ciety Review 14: 460-556. Dickey, C. & M. Meyer (2002). “Is Europe broken?”. Newsweek, 12 August: 14. Flesch, C. (1998). Fighting the fog. URL: http://www.europa.eu.int/comm/t ranslation/en/ftfog/flesch.htm [12/12/01] Gunnarsson, B.L. (1987). “The abstractness and implicitness of Swedish legal language” in A.M. Cornu, J. Vanparijs, M. Dela- haye & L. Baten (eds), Beads or Bracelets? How Do We Ap- proach LSP. Selected Papers from the 5th European Sympo- sium on LSP, 219-229. Oxford: Oxford University Press. Halliday, M. (1994). An Introduc- tion to Functional Grammar, 2nd edition. London: Edward Arnold. Horst, K. (2000). Did you say Sprachbarrieren? URL: http://ec.europa.eu/translation/r eading/articles/pdf/2000_tp_van _der_horst.pdf [08/03/07] Jakobson, R. (1960). “Linguis- tics and poetics” in T. Sebeok (ed.), Style in Language, 350- 377. Cambridge, MA: M.I.T. Press. Kintsch, W. (1988). “The role of knowledge in discourse compre- hension construction-integration model”. Psychological Review 95: 163-182. Kintsch, W. (1998). Comprehen- sion: A Paradigm for Cognition. Cambridge: Cambridge Univer- sity Press. Koskinen, K (2004). “Shared cul- ture? Reflections on recent trends in translation studies”. Target 16: 143-156. Maley, Y. (1994). “The language of the law” in J. Gibbons (ed.), Language and the Law, 11-50. New York: Longman. Martin, T. (2000). Image and self image: public and private per- ceptions of the translator. URL: http://ec.europa.eu/translation/r eading/articles/pdf/2000_tp_tim _martin.pdf [08/03/07] McAuliffe, K. (2006). “Transla- tion at the Court of Justice of the European Communities”. Paper presented at the Language and Law Conference. Düsseldorf, May 2006. Mezzarese, T. (2000). “Legal in- terpretation as translation: three readings of a current analogy” in G. Zaccaria (ed.). Journal of Le- gal Hermeneutics. Translation in Law, 161-188. Münster: Lit Ver- lag. Newmark, P. (1986). Approach- es to Translation. Oxford: Perga- mon Press. Nida, E. (1975). Language Structure and Translation. Stan- ford: Stanford University Press. Schäffner, C. (2002). “Strategies for translating political texts” in A. Trosborg (ed.), 119-143. Solan, L. (2006). “Statutory in- terpretation in the EU: the Au- gustinian approach”. Paper pre- sented at the Language and Law Conference, Düsseldorf, May 2006. Tiersma, P.M. (1999). Legal Lan- guage. Chicago: University of Chicago Press. Toscani, G. (2002). “Translation and law – multilingual context of the European Union institu- tions”. International Journal of Legal Information 30: 288-307. Tosi, A. (2005). “EU translation problems and the danger of lin- guistic devaluation”. Internation- al Journal of Applied Linguistics 15: 384-388. Trosborg, A. (1997a). “Translat- ing hybrid political texts” in A. Trosborg (ed.), 145-158. Trosborg, A. (ed.) (1997b). Text, Typology and Translation. Ams- terdam: John Benjamins. Van Dijk, T. & W. Kintsch (1983). Strategies of Discourse Com- prehension. New York: Academ- ic Press. Wagner, E. (2000a). “Quality of written communication in a mul- tilingual organization”. Termi- nologie et Traduction 1: 5-16. Wagner, E. (2000b). Translation of multinational instruments in the EU. URL: http://ec.europa.eu/translation/r eading/articles/theory_and_prac tice_en.htm [12/12/01] Wagner, E. & T. Martin (1998). “Fighting the fog at the Euro- pean Commission”. Terminolo- gie et Traduction 2: 131-139. Yankova, D. & I. Vassileva (2002). “Supranational legisla- tive texts: a new challenge for translators in Europe” in H. Kelz (ed.), Die sprachliche Zukunft Europas. Mehrsprachigkeit und Sprachenpolitik, 145-158. Baden-Baden: Nomos-Verlag. 07 IBERICA 16.qxp 4/10/08 18:24 Página 144 Diana Yankova, PhD., is Associate Professor at the Applied Linguistics Department, New Bulgarian University, Sofia, Bulgaria. She is the author of monographs and over 30 articles on legal language, culture and genre- specific characteristics of Common law and Continental legislation, convergence of legal studies and linguistics, EU legal terminology, teaching EALP. NOTES 1 A shorter version of this paper was presented at the International Conference “Law and Society in the 21st Century” held at Humboldt University, Berlin July 25-28, 2007. 2 Movement started by translators within EU institutions providing guidelines on how to write clearly when drafting or translating. For further detail, see http://ec.europa.eu/translation/writing/clear_writing /fight_the_fog_en.pdf 3 In the common law tradition, legal fictions are suppositions of fact which are not necessarily true, but are taken to be true by the courts of law. 4 In 2001 the European Commission services submitted a proposal to the Commission President to use only English in some specific cases. A joint letter was drafted in French and German by the then Foreign Ministers of France and Germany, declaring the proposal unthinkable and one that would promote unilingualism in the Commission. 5 All the numbers in this section are quoted from “Translation in the Commission: where do we stand two years after Enlargement?” EU MEMO/06/173, April 27, 2006. 6 A temporary waiver from a regulation or a directive, normally granted only by unanimous agreement of the Council of Ministers, and for a limited period: in the case of Maltese for three years beginning on May 1, 2004; in the case of Irish for five years as of January 1, 2007. 7 The most problematic for a translator would be Step B: the poorer knowledge base. THE TEXT AND CONTEXT OF EU DIRECTIVES IBÉRICA 16 [2008]: 129-146 145 07 IBERICA 16.qxp 4/10/08 18:24 Página 145 07 IBERICA 16.qxp 4/10/08 18:24 Página 146