next up previous contents
Next: Application to Spanish Up: Language-specific applications Previous: Application to English

Preliminary Recommendations

 

Application to Dutch

Most of the tags described in the sections devoted to application to Dutch are selected from two different CELEX lexical databases: some of them are taken from the syntactic information columns of the Lemmas Lexicon and some from the inflectional information columns of the Wordforms Lexicon. To understand the term column it is necessary to understand that the databases consist of rows divided into numerous columns, each column containing a value tag. Each row contains in its columns the information belonging to one lemma (Lemmas Lexicon) or to one wordform (Wordforms Lexicon). The syntactic information in the Lemmas Lexicon is represented by Yes/No tags or by abbreviated name tags. As well as such Yes/No tags, the Wordforms Lexicon presents a different kind of tag, called Flection Type tags. These are single or composed tags, each element representing one aspect of the inflected wordform.

The examples given in the CELEX tables of the Dutch application are deliberately taken from Burnage (1990). In cases where there were no examples, then examples were provided. This is always marked.

The CELEX system allows the attribution to one word form of double or triple tags, separated by slashes, such as: `geo./pers.' and `hebben/zijn' and `intrans./trans./wederk.'.

It should be noted that there are two representations of each syntactic tag in the Lemmas Lexicon: a numeric one and an alphanumeric one. In the tables of the application only the alphanumeric tags are found.

Some categories are not present in the CELEX databases mentioned above. To fill this gap, some distinctions are presented. These are labelled in the applications as `Proposals for Dutch' and are either based on the German tables or on the MULTILEX table, and always based on Dutch traditional grammar.

There is another CELEX database, containing still more detailed information on subcategorisation and subclassification of a syntactic as well as a semantic nature. This database is not yet fully described, but we did use this source for some tables in the application.

The CELEX tags are in the Dutch language, whereas the tags in the ten `Proposals' are Latin-derived or English.

The Dutch section takes into account feedback (Heylen, 1994) which as well as commenting on the application to Dutch also raised some general considerations and questions.



next up previous contents
Next: Application to Spanish Up: Language-specific applications Previous: Application to English