Silvia Bernardini graduated in translation in Forlì in 1997 and, after a short spell in the UK (where she got her MPhil from the University of Cambridge), came back to Forlì, where she has worked as an assistant professor, an associate professor and now as a full professor. By the time she returned to Italy, she had also completed her PhD in translation studies (Middlesex, UK). She has taught a variety of courses (technical and scientific translation, translation methods and technology, and English (corpus) linguistics) and has coordinated the Master’s course in specialised translation. Her research interests are in the fields of translation technology, translator education, English as a lingua franca, documentation and corpus linguistics. She has given invited talks on these topics in Belgium, Brazil, China, Germany, Norway, Serbia, Spain, Sweden, the UK and many Italian universities.
1. You are a professor in English Language and Translation with several years’ experience in teaching and research. Your PhD focuses on a corpus-based study of collocations in translated and original texts. What inspired you to examine this particular subject in depth?
My PhD research brought together two topics that are very central for me, namely phraseological tendencies in language use and typical features of translated language. As a student in the 1990s, prior to starting my PhD, I had the good fortune to meet John Sinclair (creator of the revolutionary COBUILD dictionaries and one of the fathers of corpus linguistics) and to witness the development of the British National Corpus. The 1990s were a period of great enthusiasm for the ‘corpus revolution’, and translation studies scholars like Mona Baker and Sara Laviosa also saw the potential in this area.
I found the corpus-based quest for typical features of translation fascinating, and in my PhD I set out to discover whether any such features could be identified at the collocational level. In case you also want to know whether my quest was successful, let us simply say that research is always more complex than you think at first – this was in itself a valuable lesson for me! However, I did observe some interesting regularities, including (and here I am generalising somewhat!) a tendency for translators to favour plain and common collocations in their target texts, sometimes choosing not to reflect unusual or creative ones found in their source texts.
2. In October 2014 you participated in the conference Terms and Terminology in the European Context with Erika Dalan, Adriano Ferraresi, Eros Zanchetta and Marcello Soffritti and presented a very interesting paper entitled ‘Terminology in international academic settings: The case of English native and lingua franca course unit descriptions’. What advice would you give to ELF (English as a lingua franca) universities when creating or improving their catalogues in English? Do you think that universities in the EU could benefit from retrieving information from shared corpora?
As early as 2001, in an article for The Guardian, translation studies scholar Juliane House suggested that English was the de facto lingua franca of Europe. Universities across the continent use English to connect with their stakeholders (particularly students), and yet terminological inconsistencies are frequent and may make communication problematic. Our study focused in particular on course catalogues in English, given their role in ensuring student mobility and the internationalisation of higher education institutions, but several conference participants confirmed our fears: terminological inconsistencies plague institutional communication regardless of the languages and genres involved!
I think it is crucial for universities to understand that sound terminological work is a prerequisite for mutual understanding. There is no doubt in my mind that grassroots collaboration in collecting corpora, developing terminological databases and sharing such resources as widely as possible would bring an extraordinary return on investment to the EU higher education area as a whole.
3. Consulting corpora is proving to be crucial when it comes to translation. How do you think translators should approach corpus analysis and how does it affect the quality of translated texts?
I tried to answer this question during a talk I gave in October 2016 in Pisa to an audience of over 100 professional translators, both budding and seasoned. For a number of reasons, corpora have so far been used less than other technological aids, such as translation memories. For all their merits, TMs have certain disadvantages: they make translation faster, but not necessarily better, and are not fully adequate for every type of text and task. With machine translation improving by the day, I believe there will be less need for fast translation, and more scope for high-quality translation (and authoring, transcreation, post-editing, you name it). Corpora can help us to evaluate creativity in source texts, enlarge our pool of target language equivalents, make informed decisions and defend them when necessary. That’s why I think they are worth going the extra mile for!
4. At what stage in your career did you first become interested in terminology? And what is the most recent project you have been involved with in the field of terminology?
The first time I came into contact with terminology was in 1994, more than 20 years ago. Franco Bertaccini, who taught me translation from French into Italian then, introduced us to the basics of the discipline. He was keen to stress that terminological/terminographic work is an inherent part of specialised translation work, whether as a separate professional task or as an activity that is carried out while translating. It was during this course that I started to become acquainted with translation as a profession – over and beyond translation as an academic subject – and I think it helped me lay the bases for my own approach to translation research and teaching.
Coming back to the present, in the last couple of years I have started to work on institutional academic terminology. Academics and admin staff at the University of Bologna are acutely aware of the need to make sure that communication with our national and international students and potential partners is effective. We have consequently started to develop a database of institutional academic terminology in English and Italian for internal use, drawing on available materials produced at EU level. We hope to be able to share it soon with colleagues and continue to develop it as a joint effort. Contacts have already been established with several research groups across Italy and Europe.
5. Would you recommend some best practices for extracting and managing terms from a domain-specific corpus?
Corpus query tools provide several functionalities for term extraction and management, offering an empirical basis for selecting terms and recording their co-textual patterning (e.g. wordlists and keyword lists, cluster and n-gram lists). I would recommend that terminologists who wish to explore these possibilities download and play with AntConc, Laurence Anthony’s simple yet fully functional corpus query tool. Open-source or free tools also exist for extracting parallel terminology from aligned corpora automatically. For instance, we obtained promising results with anymalign. It should never be forgotten, though, that these tools simply provide starting points for analysts: the bulk of terminological work remains manual, requiring the insightful analysis of highly trained professionals.
6. Together with other professors and researchers, you founded the Terminology Lab (LabTerm, Laboratorio di Terminologia e Traduzione Assistita) affiliated with the University of Bologna’s Department of Interpreting and Translation (DIT). Can you describe some of its activities? Did it help to establish cooperation between students, teaching staff and the professional world?
The Terminology Lab, or LabTerm, is an important hub for all departmental activities connected with technical and specialised translation, terminology and terminography, the translation technology industry and the world of work. Together with its research division, CoLiTec (Corpora, Linguistics, Technology), the LabTerm runs a wide range of activities, including continuing education seminars for professional translators, and coordinates extended internships with local companies for MA students (e.g. through the language toolkit programme, run jointly with the Forlì-Cesena Chamber of Commerce).
But the one event we are especially proud of is the TeTra Conference, which brings together in the same room software developers, researchers, students and professionals, and lets them talk (literally!) to each other. The last one, on 30 September 2016, was a big success: more than 120 people turned up despite a train strike!
7. TermCoord is dedicating much effort to constantly improving the EU’s interinstitutional terminology database (IATE = Inter-Active Terminology for Europe). What is your opinion on this concept-based multilingual database?
IATE has always been a model for anyone working in the field of terminology. We encourage our students to become acquainted with it, use it as an example of terminological best practice and turn to it whenever they require a reliable documentation resource in their daily work. ‘Reliability’ is the keyword here: there is a lot of information out there on the web, but it is very hard and time-consuming to decide if one can trust it, particularly (but by no means only!) for novices. Databases like IATE are therefore more essential than ever, as are the constant efforts to enlarge them and keep them up to date.
8. In conclusion, what advice would you give future terminologists, based on the trends and perspectives you have observed within this field?
I believe it will be important for future terminologists to combine a firm grasp of the theories and principles of the discipline with computational resources and methods borrowed from neighbouring research fields (such as computational linguistics, machine learning and distributional semantics). In this way it will be possible to maintain the high standards associated with a manual activity and combine them with the faster turnarounds characteristic of semi-automatic, streamlined processes.
Interviewed by Giorgia Lopez
Born in Vicenza, Italy, in 1992. She completed her BA at the Advanced School of Modern Languages for Interpreters and Translators of the University of Bologna and holds an MA degree in Professional Translation from Swansea University (UK). Her working languages are Italian, Spanish and English and she takes every opportunity to improve her French and Portuguese. She is interested in translation technology, audiovisual adaptation tools and project management, hence her role as Translation Project Manager with a British language service provider (LSP) for one year. As a trainee in TermCoord from October 2016, she is hoping to expand her expertise in terminology while making a positive contribution to the team.