|
Implementation |
In the first teaching period of 1999/2000 (October till February), 20 second year students of German (on BSc programmes in Modern Languages, Translation Studies, and Combined Honours, respectively, plus three exchange students from France) participated in the project as part of the module ‘Intercultural Text Analysis’. In the first weeks, students were introduced to key concepts of textlinguistics, such as text, text type, text function, genre, coherence, cohesion, intertextuality. German and English exemplars of a variety of genres (cooking recipes, instructions for use, short news items, job offers, patient information leaflets, advertisements) were compared in class in order to identify their similarities and differences at macro-level (information arrangement) and micro-level (lexical and grammatical structures).
In their first year, all students had introductory IT sessions and were familiar with word processing systems, they also had an introduction to using the Internet. In one session during the module, they were introduced to translation systems which are available on the Internet (e.g. Babelfish, Free Translation, Systran, cf. http://www2.echo.lu/libraries/en/mling.html#tra) and we tested their outcome with examples of instructions for use. A special link was created on the School’s Intranet to allow students easy access to these machine translation systems. At the end of the course, each student wrote up the findings of his/her projects, i.e. 2000 words in German on conventions of a genre (of their own choice), including an evaluation of the machine translation output.
|
|
Outcome |
The students actively participated in collecting and analysing parallel texts, i.e. German and English texts which belong to the same genre without being themselves translations. Based on the results of the comparative analyses we did in class and of the students’ projects, the students have realized that genres are more or less conventionalized, and more or less culture-specific. They have thus acquired an awareness of textual conventions of the target language, which will eventually result in an active mastery of textual expertise when they are writing texts in their L2 (i.e. German) or when they translate into German.
Translation as target-text production means - for a large number of texts - adopting the target text to the genre conventions of the target culture. Based on the findings of the comparative analyses, human translators can decide on the most appropriate translation strategies. Machine translation systems on the Internet, on the other hand, have in general not been successful in complying with genre conventions of the target culture. This can be illustrated by the following examples Appendix 1 (most of them taken from the students’ final projects).
|
|
Evaluation |
It became clear from the project that a comparative text analysis can enhance the students awareness of culture specific genre conventions. They have been encouraged to continue parallel text research, i.e. compiling an archive of parallel texts, that can serve as a valuable resource for final year modules on translation theory and practice (this applies above all to students on the translation programme). They have become even more aware that human translators are experts in text production, which machine translation systems are not (yet). The future development of machine translation systems will need to account for genre conventions too, beyond their current focus on lexical and grammatical issues.
Returned evaluation forms indicated that students found the task stimulating and challenging. They enjoyed testing machine translation systems (in particular when the lexical choices resulted in some funny or nonsense information). All in all, the objectives of the project have been achieved.
Findings of the comparative analyses we did in class were put on the School’s Intranet, thus accessible to the students and interested colleagues. These data will be expanded over the next years to serve also as information for translation modules. A copy of this report on the WELL project will also be put on the School’s Intranet.
|
