The European Union Translation ? A New Frontier Martin Kay Stanford University and The University of the Saarland Martin Kay Machine Translation 1 Martin Kay Machine Translation 2 T h The European Union e c o s t o f t r a n s p l u a b t i l i o c n a , t i i o n t n e , r a p n r d e t l e a I g t i r a o i s l n h s , e a r n Danish v o i c f e f s i c i n i a v l o E l v Bulgarian u U e n d d l e a i n r n € Dutch g m 3 u a a . g k 5 e i n m i s g i l l e Czech i o s n t Maintenance Manuals i m a a t y e e English a d r . a t j u Estonian s Operation and Troubleshooting t Finnish 300 authors and Guides Hungarian French illustrators Disassembly and Specifications Irish Manuals German 800 English pages per Latvian Assembly Manuals 23 languages day Greek Lithuanian 2,500 (12.5%) of 20,000 staff Testing and Special Instructions Translation into 14 Maltese Italian Adjustment Guides 1% of the annual budget languages Polish Portuguese 40% of administration costs. Systems Operation Bulletins Romanian Spanish ( ) = 253 Slovene 23 Swedish Slovak 2 Martin Kay Machine Translation 3 Martin Kay Machine Translation 4
Assimilation Dissemination Indicative Informative Hard Belles Lettres Language But this is Advertising where the There is a lot of stuff action is! in this corner Scientific Papers Source Difficulty Manuals Weather Source Sound reports Easy Meaning Target High Low Target Quality Martin Kay Machine Translation 5 Martin Kay Machine Translation 6 What is Translation? Generally Speaking— • A text that is based on a text in another language and which — has the same meaning Foreign reader's experience should be as close — conveys the same information as possible to that of a domestic reader. — has the same effect on its readers — gives the gist of the original They should pass through parallel mental — explains the original states. • It depends what you want N o m e n t i o n o f m e • Generally a mixture of several of these a n i n g ! Martin Kay Machine Translation 8 Martin Kay Machine Translation 7
Reference He sat down Martin Kay Machine Translation 9 Martin Kay Blood from a Stone 10 Dateline: Situation 1 Damascus Athens Moscow Language Situation 2 Situation Updater Update semantics Martin Kay Blood from a Stone 11 Martin Kay Blood from a Stone 12
There are a lot of things Problem 1: Keep track of the things She took his wallet. — I saw it in her hand — I saw her do it — I could not believe it Reference Martin Kay Blood from a Stone 13 Martin Kay Translation 14 but only a few things — She took his wallet. — I saw it in her hand • are on stage the thing • are in the spotlight — I saw her do it • preserve the momentum of what is being that said. the deed — I could not believe it what she did Martin Kay Translation 15 Martin Kay Machine Translation 16
She took his wallet. — I saw it It was in her hand She took his wallet. She was holding it — I saw her holding it — I saw her do it These do not mean the same thing out of context — I saw it happen I saw her take it They do mean the same thing in context — I could not believe she could do it She took it as I watched — It was unbelievable (to me) She did it before my very eyes — I could not believe it I did not believe she could do such a thing Martin Kay Machine Translation 17 Martin Kay Machine Translation 18 The sense of a word tells you what properties a thing must have for that word to refer to it. Types have senses Things can be referred to by words with Tokens have referents different senses. Maybe choose the one that gives the most fluent result Martin Kay Machine Translation 19 Martin Kay Machine Translation 20
Syntactic Processing Syntactic Processing • Find words with similar senses • Find words with similar senses • Put them in a fluent order • Put them in a fluent order • Translation in Language teaching • St. Jerome: Vulgate (4th Century) • Popular perception of translation • 20 th Century • Machine translation • Translator is anonymous Martin Kay Machine Translation 21 Martin Kay Machine Translation 22 Trouble is Different grammatical structures give different meanings. • Words often have several senses. The dog saw the cat • You often cannot find words with similar The cat saw the dog senses. Der Hund sah die Katze • Even when you can, they often cannot be Den Hund sah die Katze ordered fluently. Martin Kay Machine Translation 24 Martin Kay Machine Translation 23
Versichern Sie Sich daß Sie nichts in Versichern Sie Sich daß Sie nichts in dem Zug vergeßen haben dem Zug vergeßen haben Assurez-vous que vous n'avez rien Assurez-vous que vous n'avez rien oublié dans le train oublié dans le train Martin Kay Machine Translation 25 Martin Kay Machine Translation 26 Versichern Sie Sich daß Sie nichts in Versichern Sie Sich daß Sie nichts in dem Zug vergeßen haben dem Zug vergeßen haben Assurez-vous que vous n'avez rien Assurez-vous que vous n'avez rien oublié dans le train oublié dans le train Martin Kay Machine Translation 27 Martin Kay Machine Translation 28
Versichern Sie Sich daß Sie nichts in Versichern Sie Sich daß Sie nichts in dem Zug vergeßen haben dem Zug vergeßen haben Assurez-vous que vous n'avez rien Assurez-vous que vous n'avez rien oublié dans le train oublié dans le train Martin Kay Machine Translation 29 Martin Kay Machine Translation 30 Versichern Sie Sich daß Sie nichts in Versichern Sie Sich daß Sie nichts in dem Zug vergeßen haben dem Zug vergeßen haben Assurez-vous que vous n'avez rien Assurez-vous que vous n'avez rien oublié dans le train oublié dans le train Martin Kay Machine Translation 31 Martin Kay Machine Translation 32
Versichern Sie Sich daß Sie nichts in Versichern Sie Sich daß Sie nichts in dem Zug vergeßen haben dem Zug vergeßen haben Assurez-vous que vous n'avez rien Assurez-vous que vous n'avez rien oublié dans le train oublié dans le train Martin Kay Machine Translation 33 Martin Kay Machine Translation 34 Versichern Sie Sich daß Sie nichts in Versichern Sie Sich daß Sie nichts in dem Zug vergeßen haben dem Zug vergeßen haben Assurez-vous que vous n'avez rien Assurez-vous que vous n'avez rien oublié dans le train oublié dans le train Make sure that you have forgotten nothing on the train Make sure that you have left nothing on the train Martin Kay Machine Translation 35 Martin Kay Machine Translation 36
Versichern Sie Sich daß Sie nichts in Semantic Processing dem Zug vergeßen haben Assurez-vous que vous n'avez rien oublié dans le train Replace words and phrases by words and phrases that mean the same Make sure that you take all your Reorder belongings with you Pragmatic Processing Replace words and phrases by words and phrases that will effect the same changes in the scene Martin Kay Machine Translation 37 Martin Kay Machine Translation 38 Pragmatic translation often involves addition or removal of substantive information from what is explicitly expressed. Est_ce que ce train va a Perpignan? Martin Kay Machine Translation 39 Martin Kay Translation 40
Est-ce que c’est ta cousine? Does this train go to Non, je n’ai pas de cousine. Perpignan? female No, it stops in Beziers. Is that your cousin? Fährt dieser Zug nach female No. I don’t have a cousin. Perpignan? Nein, er in hält endet Is that woman your cousin? Béziers Martin Kay Translation 41 Martin Kay Machine Translation 42 Est-ce que c’est ta cousine? Non, je n’ai pas de cousine. Abstraction Is that your cousin? No. I don’t have a cousin. girl Is that woman your cousin? Martin Kay Machine Translation 43 Martin Kay Machine Translation 44
Structural Linguistics Dimensions Linguistic Chemistry front back Elements high front Phoneme Morpheme rounded Sememe (?) low Phrase unrounded (Distinctive) feature Martin Kay Putting Linguistics back into Computational Linguistics 45 Martin Kay Putting Linguistics back into Computational Linguistics 46 Phonology Generalization One statement covers many instances. North American Flap læt ə r → læ ɾə r Martin Kay Putting Linguistics back into Computational Linguistics 47 Martin Kay Putting Linguistics back into Computational Linguistics 48
Recommend
More recommend
