Machine Translation in the European Union
The European Union (EU) is one of the most important political, social, and economic aspects of the European continent in the present day. One of its vital issues has been linguistic policy, as it is currently a body of 28 nations with 24 official languages which each enjoy a differing status even among the “procedural languages” of English, French, and German. This linguistic policy is not one which is without controversy or public debate, as demonstrated by the extensive reporting on the prospects for the English language after the 2016 United Kingdom vote to leave the EU. Furthermore, for quite some time the policy has functioned as a political issue, particularly around the critique concerning the expense of translation. This argument is oft dismissed due to the minimal costs involved (a slight 2 euros per citizen), however, to its critics appears in absolute terms as a large multi-billion euro expenditure. This topic has even functioned as a national political debate within the EU, such as the Scandinavian push for more efficient transparency and communication -- this forming a useful and effective campaign against French opposition to the Scandinavian elevation of the status of English within the EU.
One of the vital parts of the European Union’s language functionality, its translation and interpretative services, despite being the largest in the world, has received a comparatively lesser degree of scholarly research.,, Furthermore, most research authored upon the matter has looked at it through a model of language influence instead of examining the ways in which the translation and governance models of the European Union’s institutions affect the issues of language. And yet, the EU has pinned much hope upon technology as a way to resolve problems with its Tower of Babel, as shown by decades of research onto the topic. The functioning of the EU has been impacted deeply by the influence of machine translation, and yet its possibilities and limits are as much constrained by the cultural-political ideology espoused by the EU as well as by its inherent technological features. Indeed, machine translation might have its greatest importance in regards to ideological and normative aspects. Its utilization has been consistently used as a way to attempt to salvage a translation system which has appeared constantly in crisis, especially as a growing gap between official EU policy and EU reality has grown ever wider. Rather than being a transformative tool, it has served a role of preventing the genuine change needed to confront the translation problems which actually do face the European Union.
Translation in the European Union is carried out by the Translation Center which is tasked with servicing all EU agencies. Interpretation in the European Parliament is carried out by its own service and there are separate services for the European Commission and the Court of Justice of the EU. In fact, there are a total of 9 different translation services which vary quite greatly in internal ethos, objectives, size, and function. Interpretation is the process of converting thoughts and ideas from one language into another language in spoken or signed form. Translation is the same but for writing. Naturally, they have very different processes, impacts, and effects and are discussed separately in this paper. Principally, translation is used by the European Commission for written documents, while interpretation in the European Parliament is concerning spoken language.
Machine translation is the automatic translation of texts by a computer. It does not refer to assistance by machines such as dictaphones for recording, or improvements in transferring techniques, but only to the automatic process of translation itself.
In theory, all communication with a government by the European Union must take place in that government’s chosen language. The citizens speaking regional languages which have official language statuses in their own countries can choose to communicate in them with the EU, such as Catalan in Catalonia, Welsh in Wales, and Galician in Galicia. Thus, communication between the EU and a country like Finland must take place in Finnish. In practice this standard is not always met, requiring Germany -- supposedly one of the three working languages in the European Union -- to request in 2010 (repeated in 2011) that documents submitted to the Bundestag no longer be submitted in English, as was standard for the majority of them. Furthermore, despite all languages being equal, working documents are not translated into all languages due to high budget costs.
This shows a dual-sided picture. Simultaneously, the European Union is a bastion of multilingualism in theory, while in practice, if it is not a monolingual institution, it is increasingly a single-language centric one, in contrast to its formal rhetoric. The formal rhetoric hides and actually legitimizes this transformation by enabling discussion and debate, and preventing the formal marginalization of previously prestigious tongues. Machine translation is an excellent example of this. Theoretically neutral and often lumped together with the idea of multilingualism, in reality its presence helps to justify lack of alternate changes to the European system and preserves a model which is in discord with official European Union rhetoric and policy but very much in accord with an increasingly monolingual reality.
Technological History of Translation in the European Community and Union
Historically, the translation processes for the European Union (then the European Community) involved translators either working directly on a mechanical typewriter or dictating text to a typist. This was supplemented in the mid 1960s by the inclusion of dictaphones to record translations on plastic disks, which was mostly of benefit to those with poor typing skills, although simultaneously increased the workload of typing pools. Commencing in 1964, terminologists were needed to establish harmonized vocabulary for the various languages. This process involved setting up units with collections, lexicons, glossaries and vocabulary lists, a situation which continued until the introduction of the internet.
Machine translation of an effective nature began to appear in the the 1980s. Of course, it was not capable of translating documents perfectly, but it was used increasingly by large companies for rough translations. Furthermore, in the 1970s, after the abandonment of its ambitious objectives of providing perfect translation and replacing translators, a reversal occurred and in the 1990s software started to appear to aid rather than replace translators . This software did not come from translation departments and industry, but instead from software localization, as software development had to translate repetitive phrases into a host of different languages which tools such as sentence translation memory systems were highly useful for. This was integrated into the European Parliament in 1993. Funding since the 1980s had been granted to every official language, but in a population distribution. Thus, small languages like Danish were only entitled to 2% of the total expenditure, while large languages with extensive expenditure otherwise like English or French gained much more.
By the 1990s, changes stemming from translation had made no major alterations and only mild qualitative impact upon the general EC mostly altering output rather than input needs or radically restructuring the process. This had not changed by 2004, when continuing emphasis upon machine-aided translation was still a principally qualitative rather than disruptive change. However, the introduction of the internet did have a radical impact on the work of various translators themselves. Terminology departments had been a vital part of the translation process, providing translators with up to date terminology in the constantly evolving language framework in the diverse, expanding number of languages making up the European Community (and then Union). As the internet began to develop, locating appropriate terminology became much easier. However, the problem was that there was inadequate terminology in the language itself for translating certain terms, and so the process moved from one of terminology compilation to terminology creation, an issue especially pronounced given the rapid evolution of EU legal structures. Some examples of unique EU-derived terms include comitology, additionality, creeping competence, derogation, and acquis communautaire, all of which have meanings particular to the EU context. The installation of computers and of word processing equipment meant that the old stencil equipment and ronio copy machines became obsolete and the secretarial profession underwent radical changes as translators could now input text themselves, becoming much more autonomous. It should be emphasized however, that fundamentally the role and use of the translator had not changed despite the marked evolution of the work.
In this period, translation issues continued to negatively affect the efficiency of European institutions, as seen by delays during legislative activity occasioned at Brussels waiting for translated pieces. Translation seems to have almost perennially been an area which has been unable to keep up with demand, from the 1950s when linguistic services were overburdened with work, to the 1970s with actual and projected overloads. Then, in the 1990s steps undertaken in response to major expansions in translation work included new services. These included summaries, oral translations, post-editing machine translation, and translation outsourcing and led to the fear of a drop in the quality of work. Today, the issue continues, such as European Commission documents not being submitted to Germany in German, which has impeded the functioning of the Bundestag. Every expansion of the European Union has been met with the problem of translating the vast amount of extant material into the new language, and although preparations have become better over time, the inherent nature of the work is one which cannot be transmuted way. When machine translation appeared in the European Commission in the late 1990 period (in contrast to the European parliament), its objective was to develop raw translations of large amounts of material for usage in executive work and things like anti-dumping agreements. Despite immense technical changes, it appears that there has been no significant change in regards to actual productivity per translator, which has held steady around 930 pages per year per translator, although declining hours of work per translator does mean an increase in hour terms. Furthermore, even in the European Commission, machine translation has only existed simultaneously with regular translation. To date, the hopes to use it to solve the constant shortage of translation services for legislation has not been fulfilled.
Influences and Present Structures
The main influence upon translation and communication in the EU up to this point had not been new technology, nor even English as a lingua franca, but rather the expansion in its membership, which has driven a change in policies as well as the translation process itself. Originally in the European Community, translation had been possible with only a small team, such as 8 translators for the original four languages. In the EEC, which had 9 official languages, there was a need for 27 different translators, with an expansion from previous two language pairs to three language pairs groupings to provide translation. This made direct translation no longer fully possible as a hub and spoke system was needed to pass through the big languages to connect small languages without direct links. This hub system of big languages (typically French and English serving as the translation median) through which small are translated may reflect linguistic reality, but it results in reduced quality. Besides the obvious negative of requiring yet another layer of translation, other problems asserted themselves - - greater translator difficulty in the environment due to less time spent actually broadcasting what was translated (falling from ½ to ⅓ of their time), greater cultural gaps between the different languages, and decreased political cooperation as the large translating teams made events more formal and less capable of political maneuverings. Naturally, having a group of 27 people following around any meeting, larger than the actual participants, and with significantly longer transmission times, meant that such communication felt awkward and unnatural. As a result, much political affairs took place behind the scenes in a lingua franca, outside of the translation services. From this came a service policy change that only some Council (of ministers) working groups would have interpretation offered, as part of a quota, in an effort to reduce the amount of effort they had to expand. Effectively, despite the continued existence and expansion of the translation service, it meant a reduction in the breadth of translation at the Council of Ministers. Although this favored the adoption of a singular lingua franca language, the policy was not created by this motivation but rather by internal European Community dynamics.
The increasing cultural gaps are an extremely important matter and the difficulties experienced in translating legal texts very much relate to them. Ever since the beginning of the EU, terminology issues have been a major difficulty. As expressed in the technological development treatment, they have long been at the centerpiece of translation efforts. While at first this was simply ensuring that translators had access to terminology terms, the struggle currently is far more problematic because quite often an appropriate term in the target language does not exist. Hence, this requires an approximate serving a non-identical role. For example, in Regulation no. 1 which establishes language equality between official languages, the English translation reads as languages to be used and in the French term as régime linguistique; ie, linguistic régime. Different legal traditions like Civil and Common law (the latter principally an English language tradition) may produce irreconcilable differences. Common law makes no distinction between administrative and constitutional law, while these are clearly delineated legal branches in the French and German legal traditions. Translating between the two is thus extremely difficult, and requires that a translator have knowledge of the legal traditions of both common and civil law in order to be able to appropriately translate English-language terms to other languages. Administration translates as contrôle administratif in French or Verwaltungs Kontrolle in German, both meaning administrative control. Law and order meanwhile, when translated into the German term öffentliche Ordnung means a broader sense of respect and obedience to law rather than the prevention of criminal offense. Drafting rules are equally problematic in translating between different legal traditions and languages. Although these may seem like small issues to the casual reader, for international law and legal affairs, they hold great importance, as small differences such as these can produce intense controversies.
In light of these issues, the Court of Justice of the European Communities ruled that “the necessity for uniform application and accordingly for uniform interpretation makes it impossible to consider one version of the text in isolation but requires that it be interpreted on the basis of both the real intention of its author and the aim he seeks to achieve, and in the light in particular of, the versions of all four languages.” These regulations have also contributed to the sense of translators that they feel too much like simple translating machines, are uninvolved in the results or the reasons for their work, and receive little positive motivation.
Despite the formal commitment of the EU to translation and to multilingualism, translations themselves enjoy a low status in their categorization by the EU. All texts produced are classed as “originals” and there are no texts classed as “translations.” This serves important objectives in the EU of preventing a language hierarchy from being officially promulgated, but in non-official practice it actually does the reverse by obscuring the relative influence of certain languages within the European Union. More importantly, it also demonstrates a portrayal of translated texts as being of an inferior quality and authenticity as compared to the “originals.” Both of these have significant effects upon the position of translation and are not always accurate categories. Original source documents written in English are often written by non-native English speakers, and then are translated back into their own language. In such a case, why would a re-translation of a document written in native tongue be used when it is inherently of poorer quality than the “authentic” “original”? Indeed, this issue of inadequate grasp of English and the creation of English-language legal texts by non-native speakers in the EU has had deleterious effects upon the precision and clarity of texts, compounded by an insufficient knowledge of the English common law system.
In fact, translation itself is heavily affected by the requirements for textual identification. Texts translated for the European Union are in a certain sense quite unique and distinct from regular translations and are instead mirror images, as they are all supposed to carry identical weight. This has important ramifications. Translators have very strict rules concerning sentence structure. For example, they are unable to break sentences into two even when it may improve readability, although the proffering of a semicolon is acceptable. Since content must be identical between different versions, translation into target languages does not conform to target language rules as is typically the case. Instead, they must fulfill the conventions and norms of EU texts and the EU institution. Beyond any normal government tendency to excessively complicated and incomprehensible language, this has naturally meant that translated documents are extremely hard to understand and are not truly written in the target language. Instead, the oft raised criticism is that they are “eurojargon”, or “eurospeak”, and are incomprehensible not just because they come from a different register of communication which is difficult for average citizens to understand, but rather that the language utilized in EU addresses to the public are all but nigh-incomprehensible, conforming to alien language rules which must necessarily be maintained to match EU translator regulations.
As the principal administrative unit of the European Union, the European Commission is a useful case study in regards to the impact of machine translation and linguistic shifts in a purely textual environment. It is also one which is marked by a less formally equal language regime than other elements of the EU, having three working languages; French, English, and German. Thus, its translation services are principally directed for translation into it -- providing documents and information for usage internally -- or translation out, to the general public, rather than being translation within the institution.
The number of source documents of English origin in the European commission has risen quickly over time. In the 1950s it was 0% (French represented 100% of such documents), in the early 1980s 30% and 60% respectively, the late 1990s 42% and 40%, followed by 55.1% in 2000 and 74.6% in 2009. This has relevance as the domination of English as a source text produces a negative bias against other languages, due to the way in which European Union translation is set up - already convoluted and difficult to understand texts are then translated in a process which leaves minimal agency to translators and which stresses above all else conforming to EU language guidelines: Furthermore, the European Commission is the largest user of machine-translated documents in the EU -- in 2005 the Commission used 671,500 out of 860,314 pages. The Directorate General of Translation which is responsible for translation for the European Commission employed 1,750 translators as of 2005.
EC translators have a marginal and imperfect relationship to their institution, with little communication compared to other departments. As noted orally by a member of the Finnish unit of the Directorate General for Translation:
I used to be in [another EU institution] before and we had there a much closer relationship to those, those, actual, committee members that is actual text users and producers and, and, politicians we had their e-mail addresses and telephone numbers and, and well if we had translation problems we could directly contact the commission, erh, committee members. Here it, [one cannot imagine] that –
Indeed, translators working for the European commission are marked by their separation from other officials, limited engagement with them, and their compartmentalization into mono-lingual environments. Translators are essentially invisible in this arrangement : like many of those who write the text for the European Commission, the document they translate will not mention being written or translated by any individual, but instead being the product of the European Commission. The translators upon which writers depend are visualized as little more than machines in such a framework.
All Members shall have the right to speak in Parliament in the official language of their choice. Speeches delivered in one of the official languages shall be simultaneously interpreted into the other official languages and into any other language the Bureau may consider necessary. Rule 146, European Parliament Rules of Procedure
As a multilingual body where every nation that composes the European Parliament has a chosen official language, the European Parliament has inherently been deeply affected by the issues of translation/interpretation and of lingual changes within it. Existing since 1952, initially as the Common Assembly of the European Coal and Steel Community, in the legislative branch of the European Union the number of languages in the European Parliament has grown dramatically over time, from the initial four - - Dutch, Italian, French, and German - - to the current 24 official languages. Naturally, being the European Parliament, the rôle of interpretation rather than translation is more pronounced, although there is still a significant degree of translation utilized to produce hundreds of thousands of pages of work per year.
The European Parliament has been marked by increased utilization of English, or more rarely French, as a language of discourse by non-native speakers, often of variable skill levels. This has resulted in comprehension difficulties for native speakers and incomprehension and marginalization on the part of interpreters. Interpretation difficulties have long been a problem for official purposes, not even counting of course that much of the work of the The first major shock was the arrival of Denmark into the European Parliament in 1973 and the lack of sufficient interpreters between Danish and the Romance directly. This led to a joke that the Danes always got the last laugh, as translation had to go through a relay system with German or English used for interpretation, meaning a joke would be transmitted to the Danes last. But perhaps it is in another linguistic area that the Danish example showed much better the challenges facing European translation : when the formal translation infrastructure is bypassed, such as when the Danish European President chose to speak English while the Danish delegation minister continued speaking English in 1987 - a common occurrence now of course, but showing that the translation system was effectively being bypassed. Translation infrastructure in the European Parliament is designed to fulfill, in essence, an illusion : that European representatives exist in a monolingual environment, without linguistic contamination through contact with other languages.
For both the European parliament and the European Commission, as well as for the broader European Union in general, a noticeable gap exists between EU rhetoric on the surface, and EU practice in reality. This chasm is acceptable to some extent, but the constant difficulties which have emerged in the translation service over the decades as well as internal political disputes over translation and language matters mean that some sort of policy is necessary to resolve them. Obviously, there are many more than two, but for the purpose of simplicity, a comparison is useful between the ideals espoused by machine translation and its achievements (which do, as previously noted, exist, even if it has not proven so far to be a panacea), and the proposals which have been voiced by translators.
The Machine Translation Solution
The EU of course, formally supports multilingualism, and its actions demonstrate that in a specific style this is more than rhetoric. With nearly two thousand working translators and extensive interpreting services, it is the largest of its kind in the world, and shows an ongoing commitment to the utilization of translation to ensure communication in a multilingual Europe. However, at the same time, these translators can appear almost invisible in government pronouncements. A 2005 action plan for European communication bemoaned the fact that there were a lack of communication specialists at the European Commission -- this despite the 1,750 translators which it employed! This attitude is one which translators and interpreters themselves demonstrate, frustrated over being bypassed, ignored, or undervalued, despite the formally vital nature of their work which they believe to be important to the European project. This is deeper than mere carelessness, but instead speaks to a certain image of the translator in the European project; a limited role, mechanistically shifting documents from one language to another and principally concerned with purely terminology and language-based elements rather than functioning as an intermediary and communicator. This view of translators is one which receives support and legitimation from machine translation which promises to provide the capability to make a dysfunctional system work without a need for any systemic reform of the actual role of translation and communication.
Competitivity and efficiency are widely used in regards to the idea of machine translation, despite the important cultural, linguistic, legal and political problems which it does little to solve. A 1984 General Secretariat report worded itself as the following ; “afin de trouver une solution aux problèmes pratiques que pose la diversité linguistique représentant, entre autres, un obstacle au transfert technologique et une entrave à la compétitivité de la Communauté dans le monde, la Commission a présenté une proposition au Conseil en vue de la création d'un système européen de traduction automatique de conception avancée (EUROTRA)[…]” (so as to be able to find a solution to the practical problems which are posed by linguistic diversity, including, among others, a barrier to technological transfer and a constraint upon global Community competitiveness, the Commission had presented a proposition to the Council concerning the creation of an advanced conceptional European automatic translating system (EUROTRA)[...].” The focus displayed is one upon efficiency and the economic status of the European Community in the world and the linguistic diversity possessed by the European states as a hindrance to this. EUROTEA, intended to be a fully automatic system at its debut, was proposed to solve this and obsolete the translation intermediary. Even though it did not succeed, its intention was clear.
Similarly, the 2005 stated mission for the Directorate-General for Translation for the European Commission stressed the following areas for amelioration:
“Pursue the Commission’s strategy designed to match the supply of
and demand for translation in a cost-effective way;
Promote inter institutional cooperation in the translation field by
optimising the use of resources for internal and external translation,
recruitment of staff, compilation of terminology and the development
of multilingual IT tools;
Strengthen cooperation with the other Directorates-General and Services
of the Commission in the area of prioritisation of documents;
Further improve the quality of both internally and externally translated
documents and raise productivity;
Develop its role as the language service par excellence in Europe.”
As previously, the focus here is one which places itself principally upon efficiency, which sees translation as a mechanical matching of supply and demand, and which specifically calls for the development of multilingual IT tools to improve efficiency. Other stated objectives included legitimacy and transparency, but these are minimally included in the suggested improvements. Although later additions expanded these objectives, it fundamentally kept the same structure, and clearly represents the EC’s perspective upon translation and efficiency. It is one which ignores the simultaneous supposed European objective to focus upon clarity of translation. In effect, it is one which is much more a way to guard existing practices without any need for reform and to elevate the powers of the European Commission, rather than a genuine attempt to solve translation difficulties in the European Union.
The Translator Solution
A constant theme raised by all translators is increased collaboration between translators and their counterparts, and to give translators a wider rôle as cultural communicators, so that translators can be able to bridge cultural and linguistic concepts that aren’t strictly covered by the translation process and to produce more readable and comprehensible texts for their target audiences. This ranges from lawyers, to translation theorists and ex-translators, to journalists. All emphasize that their problem is not the number of texts produced by the European translators, but rather the difficulties attained in reading the produced texts, in assuring that there is appropriate clarity, and ensuring that translation is more than just a pro-forma, but rather a tool to enable communication in the European Union and to better link the EU to its citizens - as it officially has declared its objective to be.
In what ways has such a policy objective been theorized by this alternate strategy? For one, it has consistently stressed the integration of European translators into broader European institutions. As the situation currently stands, translators are de-facto segregated from other European officials, and proposals for improving the translation service coming from translators have often proposed changing this. As stated by ex-EU translator Monique Scottini:
The Translation Service should become more responsible. It is the Directorate General with the highest percentage of officials with a university degree. Translators are very competent in language matters, but they are too shy and too passive in general. [. . .] they do not see themselves as potential managers, as people who should think and take full responsibility for the texts they produce in view of the decision-making process of the Commission. In many DGs, even C-grade secretaries have more responsibilities than translators do. I think it is very important for the future of the Translation Service that translators become active partners.
Further models have included the idea of translators being incorporated into the process of the drafting of legislation, the utilization of freelance translators, not for the usual objective of cutting costs but instead for that of providing a better link to “authentic” language speakers less seeped in the EU’s particular speech, and training for translators to help them be prepared for a more active role in participation in the drafting of legal documents.
Thus for translators, proposed improvements to the European Union’s translation and intercommunication relies upon a fundamental re-visualizing of how one views the process of translation and communication in European Union institutions. If machine translation plays a role, it is a subsidiary one, for the main improvements necessary are ones which can come only through a fundamental change in the way in which EU translation is carried out.
Rather than being a transformative measure as hoped, and despite constant attention and technological development through the decades, machine translation in the European Union has failed to change the impact and utilization of translation in a transformative way. For decades, the European translation resources have faced a constant problem of insufficient services and difficulties in translation between languages. These range from establishing initial translation between documents in the 1950s, to overwork in the 1970s, to attempted reforms in 1991, and the need for various delays and exceptions to ensure that appropriate translation could be carried out for the inclusion of Eastern European nations. It is clear that although the degree of translation failing to keep up to demand has varied over time, it is a gap which has always been present, and one which the European Union has failed to address adequately.
Translators themselves have been well aware of the problem and have made suggested solutions. The 2003 book Crossing Barriers and Bridging Cultures: The Challenges of Multilingual Translation for the European Union was written overwhelmingly by translators, and none of which expressed any faith in machine translation as a solution to their problems. It was not translation departments which developed machine translation software - - assisted translation came from software localization companies.
What role then has machine translation served in the EU? Although it has increased efficiency and provided for greater flexibility, it has not resolved the problems of translation, which have persisted constantly throughout the entirety of the period. Instead, it has served as a way to legitimate the language policy of the European Union, or more precisely, the disconnect between this language policy and reality. For translators, their challenge has been this chasm, which has simultaneously continued their symbolic existence while undermining their value in the European project, as the multi-lingual ideal of the European Union clashes with the reality of an increasingly monolingual environment -- one where the rôle of translators is marginalized or reduced to a political tool. In this sense, the vision of machine translation as a way to provide for efficiency in a way which does not require any change in the way that information is transmitted and translation is handled, is one which serves as a legitimating and normalizing narrative for a particular de-facto monolingual vision of the European Union.
In the future it might be that machine translation will be effective enough to render it capable of being utilized by European Union institutions in a truly transformative way, although most likely this transformation would serve largely to simply replace the existing translation departments without changing their role. For now, machine translation has played a distinctly subsidiary role to other changes, and it is one which has principally been of use for legitimizing the rupture between the policy and the reality of the European Union’s language use.
Ammon, Ulrich, and Kruse, Jan. “Does translation support multilingualism in the EU? Promises
and reality – the example of German.” International Journal of Applied Linguistics 23
Bandov, Goran. “Language Policy of the European Union – Realization of the Multilingual
Policy of the EU.” Međunarodne studije 13 no. 1 (2013): 64-84.
Cavoski, Aleksandra. "Interaction of law and language in the EU: Challenges of translating in
multilingual environment." Journal of Specialised Translation 27 (2017): 58-74.
Coulmas, Florian. A Language Policy for the European Community: Aspects and Quandaries.
Berlin: Mouton de Gruyter, 1991.
Directorate-General for Translation Communication Unit. Translating for a Multilingual
Community. Luxembourg, Office for Official Publications of the European Communities. 2007.
European Communities. Translation at the European Commission: A History. Luxembourg,
Office for Official Publications of the European Communities, 2010.
Hale, Sandra, Ozolins, Uldis, and Stern, Ludmilla. The Critical Link 5: Quality in interpreting –
a shared responsibility. Amsterdam, John Benjamins Publishing Company, 2009.
Kaisa, Koskinen. ”Institutional Illusions: Translating in the EU Commission.” The Translator 6
no. 1 (2000): 49-65
Koskinen, Kaisa. Translating Institutions: An Ethnographic Study of EU Translation.
New York, Routledge Publishing, 2008.
Leal, Alice. “Translation at the European Union and English as a Lingua Franca: Can erasing
language hierarchy foster multilingualism?” New Voices in Translation Studies 14 (2016)
LE SERVICE LINGUISTIQUE DU SECRÉTARIAT GÉNÉRAL DU CONSEIL DE L’UNION
EUROPÉENNE: Assurer le bon fonctionnement du multilinguisme. (Brussels, General
Secretariat of the Council Directorate-General A - Personnel and Administration
Directorate 3 - Translation and Document Production, February 2012).
Mamadouh, Virginie, “Dealing with Multilingualism in the European Union: Cultural Theory
Rationalities and Language Politics.” Journal of Comparative Policy Analysis: Research
and Practice 4 (2002): 327-345.
Philippe, Van, Parijs. “The Ground Floor of the World: On the Socio-Economic Consequences of
Linguistic Globalization.” International Political Science Review / Revue internationale de science politique 21 no. 2 (2000): 217-233.
Phillipson, Robert. “Globalizing English: Are Linguistic Human Rights an Alternative to
Linguistic Imperialism?” Language Science 20 no. 1 (1998).
Slocum, Jonathan. “A Survey of Machine Translation: Its History, Current Status, and Future
Prospects.” Computational Linguistics 11 no. 1 (1985): 1-17.
Sosoni, Vilelmini. "A hybrid translation theory for EU texts." Translation Studies 5 no. 5 (2012):
Tosi, Arturo, “ Crossing Barriers and Bridging Cultures: The Challenges of Multilingual
Translation for the European Union.” Buffalo: Multilingual Matters, 2003.
Vieilledent-Monfort, Catherine. "Études sur la traduction et le multilinguisme. La traduction à la
Commission: 1958-2010." (2009).
© 2018 Ryan Thomas