| Abstract: |
This article presents the concept of the Internet as a lexicographical corpus, which implies three hypotheses. Firstly, although there are many corpus-based and/or corpus-driven dictionaries, e.g., the Macmillan English Dictionary, the concept of corpus developed in Corpus Linguistics is not adequate for lexicography per se. Consequently, discussions on aspects such as corpus composition, corpus representativeness and corpus annotation are secondary in the realm of lexicography. Secondly, using tools such as crawlers for extracting terms from the Internet with the fi nal aim of constructing dictionary articles automatically does not deliver adequately yet. And thirdly, the Internet is a repository of data that can be used for lexicographical purposes. The paper shows how this last idea works in an dictionary project: the accounting dictionaries, a set of online dictionaries that are constructed under the tenets of the function theory of lexicography and which makes use of the Internet as a corpus for three main lexicographical tasks: refi nement of expert knowledge; updating language knowledge; and completing and checking dictionary articles. These three tasks are integrated within a user oriented framework that sees lexicography as an independent science whose essence is its capacity to provide quick and easy access to dictionary data for being used by specific users in specific use situations. For instance, a specialized information tool must offer users ways of disambiguating meaning in a precise way, as what makes specialized language different from general language is its desire for univocity and monosemy as well as the reliance of its users on expert knowledge. |