4 June 2020
The SENSE 2020 Conference celebrates the Society’s 30th anniversary and has been months in the making. Maastricht was to be our host city and the Amrath Grand Hotel de l'Empereur our venue, but fate had a different idea in store for us. Nevertheless, thanks to the efforts of the conference team, the show must go on(line) – no mean feat with such little precedent having been set for online conferences in our sector.
SENSE Chair, Mike Gould, opened the conference with a reminder of the tough market conditions that freelancers face, which we often feel are beyond our control. He shared a pertinent statistic: 40% of the Dutch workforce consists of freelancers, a stark reminder of the vital part we play in the economy. Mike sees a role for SENSE in this context, namely as a society that offers opportunities for its members to up their skills to match demand.
Joy Burrough-Boenisch delivered the first session, contrasting the way in which brackets are used in Dutch and English and offering strategies to translate phrases dominated by brackets from Dutch into English. In English, we use brackets to add information without which a sentence would still make sense. In Dutch, however, brackets can also be used to mean ‘or’: ‘a (folding) bicycle’ describes a bicycle that either folds or does not. The problem is that this difference is not always translated accurately into English, particularly if the translator is not a native speaker of English.
Joy showed us some creative examples of similar bracket usage in English by native speakers in newspaper headlines, such as ‘the li[k]es of Facebook’ and ‘Big(ot) Brother’, despite these going against the authoritative sources on bracket usage. In essence, writers must be aware of misapplying the Dutch bracket convention in English, and translators and editors must not simply ignore bracketed material or transpose it directly, but identify what the author intended by it and translate it accordingly.
Tony Parr led the second session on translating texts for museums, as iconic places with very visible, public-facing translations. Despite institutions such as the V&A issuing and following guidelines for museum texts, a number of Dutch museums are producing substandard texts for exhibition spaces and individual exhibits. This is possibly owing to different approaches adopted by curators (who want to include a lot of detail in exhibit texts) and communicators (who seek to keep texts short and sweet). Regardless, the lack of sound language practitioner intervention seems to be a contributory factor.
Tony proceeded to show us examples of poor layouts of museum exhibit texts and their translations, like using the same font and colour for multiple languages with little space in between, or texts where the translation font is miniscule. He also presented mistranslations such as ‘entrance disabled’ and instances of misused tense, as when the present tense is used in a text referring to a historical event (a clear influence from the Dutch language). Tony also stressed the importance of offering additional information in museum translations; a Dutch speaker, for example, would know that ‘het IJ’ refers to a body of water in Amsterdam, but a non-Dutch museum visitor, not necessarily a native English-speaker, is unlikely to. However, after carrying out surveys that revealed how few people actually read exhibit texts, the crux of Tony’s message was that a shorter, summarised translation might actually be a better fit.
Emma Hartkamp gave an interesting overview of the European Commission’s DG Translation (DGT) work. MT is used at the DGT under a neural and deep learning-based system known as eTranslation, available in all official EU languages and Norwegian & Icelandic (650 combinations), including to public administrations in Member States. Great investments have been made in eTranslation within the Commission’s wider role of creating a Europe fit for the digital age. The European language industry survey 2020 confirmed trends regarding MT, and identified AI as the strongest newcomer. Technological change is not seen as a major stress factor, yet the survey revealed that there is not enough adequate training for translators in technology.
The DGT is collaborating with a network of Master’s programmes in Translation (the EMT network) to improve the quality of training and helping young graduates enter the translation job market. The EMT Competences Framework 2009-17 served as a leading reference standard for translator training and competences:
- Language and culture: transcultural and sociolinguistic awareness (not just language learning)
- Translation: strategic, methodological and thematic competence
- Technology: tools, applications and MT
- Personal and interpersonal: soft skills and collaborative & multi-cultural environments
- Service provision: client awareness, negotiation, project management and ethics & professional standards
These form the basis for the updated competence profile for future EU translators, which will also see a sixth area added on terminology and data management, and the translation area expanded to include revision and editing. New tasks will be integrated, notably PEMT and project management, as well as new roles, such as that of language technology coordinator.
It was interesting to learn, too, that only groups of freelancers may pitch for EU/EC tenders, because it is felt that individuals will not be able to cope with translation jobs of any magnitude. The alternative is for them to continue working through the translation agencies that win the current round of tenders.
In his session, Brian Mossop outlined how we can define the amount of time that quality requires. He referred to the problem of not spending enough time to achieve quality (caused by over-confidence, distractions and not taking the job seriously) resulting in errors, literal translations and failure to identify ambiguities. Spending too much time, on the other hand – owing to either a lack of confidence or perfectionism – can produce synonyms that are no better, too much of a focus on minor aspects to the detriment of major ones and ultimately a loss of income.
In practice, translators and editors do not have much of a chance to influence the deadline, so our time must be used wisely, and we must develop a realistic concept of the quality achievable within that time. However, there is no agreement on how to define good quality. Brian shared ten criteria on how to define a good-quality translation, summed up nicely by his last point: the output is fit for purpose. This naturally hangs on the accuracy of the text, which can be interpreted differently in different contexts. Sometimes an approximation will do, sometimes omissions are appropriate and sometimes intelligent guessing is perfectly suitable. Equally, universal advice on good writing – avoid the passive, be concise and avoid noun strings, among others – should not be applied without discrimination.
Brian presented a table of four levels of quality: intelligible, informative, publishable and polished. Each has its place depending on the situation, so not every text needs to be translated to the highest level. In order to save time, Brian’s advice is: don’t get hung up on a phrase, come back to it. Use the first word that comes to mind and don’t search for synonyms for no good reason. When checking, the focus is on reading, not writing: don’t make a change for the sake of it. Check monolingually if you’re confident about the accuracy but not with the formulation, and seek to make small changes, not re-write. Ultimately, we should ask ourselves ‘why do I need to change this?’, not ‘how could this be improved?’. Brian affirmed that that we cannot spot every mistake or eliminate every chance of a misinterpretation.
Jennifer de Beyer of the UK EQUATOR Centre gave the last talk of the day. She spoke about the utility of reporting guidelines, which provide the minimum information required in an article for a reader to understand it. The aim of these guidelines is to improve writing quality and ultimately forge better research. No official body has been appointed to oversee the formulation of these guidelines, and reporting quality is inadequate – many authors have never even heard of them, choose not to use them, or do so infrequently or inconsistently.
The EQUATOR Centre teaches academic writing skills and embeds good reporting in the writing framework. That results in a clear process for writing manuscripts and an understanding of what reporting constitutes, to the extent that students can then apply these standards to other texts. Essentially, she argues, they should be applied from the get-go of research design, not only from the writing stage, when it is usually too late to apply them. Also, researchers who have shaky academic writing skills need to polish those first, when interpreting the guidelines tends to become more meaningful. Speaking to a group of language practitioners, though, she admitted that the development or fine-tuning of the reporting guidelines – or their pruning – has fundamentally lacked the input of copy-editors, translators and other wordsmiths.