Method for extracting Basque terminology from translated legal texts
This paper is intended to make known the method used to locate and analyse legal terminology in Basque in a complete corpus for a particular field. Discarding the methods usually used in terminology processing, we based our research on three points: computer resources, the corpus linguistics and the science of translation. Since legislation itself is the basis of written law and is therefore highly important in legal terminology, we took as our framework for research the whole body of law produced by the Basque Parliament. Since the Basque language versions are translations of Spanish originals, we based our study on those originals and then found their Basque equivalents, in the sure knowledge that legal terminology in Spanish is sufficiently well consolidated and set down in dictionaries. After obtaining the full texts in both languages on a computer data carrier, we proceeded to compare and set out in parallel the two versions so that we could identify with numbers those paragraphs which had the same contents. Using a special application we then established connections between the appearances of the main terms in the Spanish list already drawn up and the context paragraphs and paragraphs with the same numbers in the Basque version. We then located the segments of the paragraphs in Basque which contained the term equivalent to the Spanish one, and transferred the whole thing to a relational database. This allows us to determine what terms appear and what terms do not appear, the area of law to which they belong, the degree to which the Basque terms are consolidated, the syntactic structures of the Basque segments equivalent to the Spanish terms, etc. Phraseology, which is so important in specialist language, could also be treated in this same way. |  |