Interview with Sabine Kirchmeier-Andersen (EN)

Interview with Sabine Kirchmeier-Andersen

Sabine Kirchmeier-Andersen has been Director of the Danish Language Council since 2006. She is in charge of the daily management of the Council’s research institute and represents the Council in relation to national and world-wide cooperative partners. Sabine’s most important research areas include language technology, linguistic strategies of companies and institutions, and valence theory.

Sabine holds aSabine Kirchmeier-Andersenn MA in Danish language and literature and BA in German language and literature from University of Copenhagen in 1987. Furthermore, she has a PhD in Linguistics and computational linguistics from Odense University (now University of Southern Denmark) in 1997.

See full curriculum vitae.

You are the director of the Danish Language Council. What is the Council’s most distinguished purpose, and wherein lies its strengths?

The Danish Language Council has three important tasks:
– to follow the development of the language
– to answer questions about Danish language and language usage
– to draw up accepted conventions of spelling and punctuation in Danish language usage and edit the official Danish dictionary of spelling

You are also the Deputy President of EFNIL. What importance do you ascribe to this organisation in its efforts to promote linguistic and cultural diversity within the EU?

EFNIL is an open forum for all official institutions for the national languages of Europe. We continuously follow the language policies of the countries and keep a close contact with the EU’s language institutions, arrange conferences and launch projects which promote linguistic diversity not only in the member states, but also in other European countries. Topics that we have had on the agenda, include terminology, translation, interpretation, language teaching, etc. The next conference will be about language usage in university teaching and research.

You deal with computational linguistics. What does it bring to the discipline of terminology?

Computational linguistics is a discipline that deals with formal language description in order for language to be handled by computer programs, e.g. machine translation systems, speech technology and databases for dictionaries and professional terms. First-rate systematic descriptions of multilingual terminology are important building stones in most language engineering programs. Computational linguistics can contribute to a better understanding of what is needed in order for the programs to handle the languages. Traditional dictionary and terminology works are based on large quantities of implicit knowledge which we humans have, e.g. in terms of our own experiences, but which a computer does not have. For instance we know that a bicycle has two wheels and a saddle, but this the machine does not know. Consequently, we can create meaning in the sentences: Han kom på sin cykel. Hjulet var fladt [He arrived on his bicycle. The wheel was flat], even if the rule in Danish dictates that new references must be introduced in the indefinite form. To us the wheel is not a new reference because we know that it is a part of the bicycle; hence it is okay to break the rule here. Computational linguists can make this knowledge explicit and formalise it in order for the programs to use it. Consequently, we are always in the borderland between language and artificial intelligence.

What is the latest or most important research that you have done in the field of computational linguistics?

The most difficult of them all: machine translation. Here the machine not only has to master one language, but it also has to transfer the meaning into a similar expression in another language, and we of course know that it is not possible to simply translate one-to-one. So it is difficult to formalize. The new statistical translation programs such as Google Translate, which is still the best of the ones available in Danish – and unfortunately of a rather fluctuating quality, seems to have reached the limit of its capacity. Google only deals with the surface of the text and contains no deeper analyses, which is why it often makes mistakes. The latest projects that I have been involved in have sought to incorporate more knowledge about the syntax and semantics of words in the statistical processes in order to improve the quality.

As well as coordinating the terminology database IATE, the mission of the Terminology Coordination Unit of the European Parliament’s is to help translators and make terminology research and management more efficient. As a linguist, what is your opinion on IATE and the terminology work in the European Parliament?

I believe that terminology work is a completely indispensable and central task. A couple of years ago, I heard a presentation on the terminology work in Nokia which described Nokia’s terminology base as the biggest asset they had. This is because the translation of all manuals, product descriptions, websites etc. must be of a high quality, and furthermore the process must be as efficient as possible. There is a huge potential in excellent terminology work and a well-organised terminology database. Essentially a good terminology database can function as a knowledge bank for the whole organisation because it is possible to attach lots of information to each word, and because the base can easily be linked to other language engineering tools such as machine translation, spelling control, search engines etc. I wish there were a national term bank in Denmark just as there is in Norway and Sweden. Indeed, IATE contains lots of different types of information from databases which have become integrated, but this can be confusing. Continued clean-up should be carried out, and also in this matter a national term bank would help.

What do you think about proposals to standardise the use and the creation of databases? Do you think this is feasible and adequate?

It would be excellent to standardise the functionality of databases in relation to its users, i.e. having the encountered screen displays and the visible types of information follow the same principles. Consider accessing a web shop in order to buy something on the internet. Each place has its own principles on where the products are placed, and what they are called. When you are buying garden furniture, you also have to search for outdoor furniture, terrace furniture, balcony furniture etc. And while on some websites they are placed separately, on others they are found among chairs, tables and benches. A flowerpot holder is called a flowerpot bowl or a cachepot or a planter and is typically to be found under miscellaneous, other articles, applied art, and what have we. It is a mess – and a waste of time.

Neologisms come into existence constantly. What, in your opinion, is the best standardisation policy to be applied to them?

Please note that I do not think that language should be standardised, but rather the way you find the information. It is obviously more convenient and efficient, especially if wanting to find something automatically, that you consistently use the same way of spelling for the same words, and when it comes to terms, it also makes sense to find a standardised definition of the content. But language is constantly in motion, and we need new words and expressions in order to be able to understand the world around us and to express our experiences, thoughts and ideas. Our world view is moving, and with this the content of words, too. This you realise when you go back in the dictionaries. Hence, there is a continuous consensus process when it comes to language, in the different groupings that we find in society. Therefore we might encounter that some groupings have developed a new meaning or a new use of a word – which then causes surprise and sometimes anger and indignation among those who have not been part of the process.

What new trends have you noticed within the past years in the Danish language? In which direction is the language moving in the future?

It is not the language, but we who move, and within the past years we have taken a giant leap when it comes to communication forms – and language follows. Due to new media we are able to communicate with each other at an accelerated speed and with this also the possibilities of developing the language we use, for the purposes we wish. But the discussion of correct language usage has not become less significant for this reason, and it takes place on a large scale on the social media.

One of your greatest wishes for the future is to create a Danish/Nordic LSP terminology database. What is your opinion on the progress that has been made in the field of Danish terminology in recent years? Do you think Danish terminology is getting enough recognition in Denmark and in Europe?

I think that things are developing far too slowly. Since the first language status, Sprog på spil [Language at Stake] in 2003, the Language Council has addressed several times that there should be more focus on this area. We managed to include the national term bank in the preliminary work for the national plans about research infrastructure, and it has been included in several parliamentary motions proposed by the Danish Parliament. But unfortunately there has been no political will to take up the challenge, and this is why we lag far behind the other Nordic countries in this regard.

How do you see the future of terminology as a discipline, and what innovations do you expect in the future?

I think it is important to our Danish language usage that we work with the professional language and with a coordinated effort for a standardised use of technical terms. We see the need as early as in primary school where Danish is integrated more and more with other disciplines, and the focus is more and more on professional Danish. Students should learn to express themselves clearly and precisely and to choose the right words not only during Danish lessons, but in all subjects. And this continues throughout the educational system onto the highest level. But there is a lack of LSP dictionaries with explanations of words in many professional areas, and often it is also hard to find translations of technical terms into other languages, which means you have to feel your way around. I believe a national term bank with a solid linkage to the EU is a brilliant project of knowledge sharing which could save us billions of kroner. Some of the innovations which I expect to see in the future are methods that automatically collect and structure terminological data, such as the ongoing DANTERMcentre project at Copenhagen Business School (CBS).


Interviewer: Claus Skovbjerg

clHaving first completed a three month traineeship translating for the Danish Language Unit, Claus joined the TermCoord Team for another three months. He has completed an MA in English, Spanish and Physical Education from Aalborg and Copenhagen University, including a Tourism & Communication semester in Melbourne and a Journalism semester in Washington, D.C. He loves psychology, foreign cultures and languages, besides his mothertounge, he also speaks Swedish, Norwegian, English, German and Spanish as well as a bit of Italian and French.