Preliminary Recommendations

Bottom-up approach

In agreement with the general EAGLES methodology, we have adopted a bottom-up approach, taking as our starting point existing practices in the main lexical and textual projects world-wide (both in lexicon specifications and in corpus tagsets).

The procedure followed was to:

1. Survey the main encoding practices for morphosyntactic description in lexicons and in corpora, with the aim of deriving a consensual proposal from their comparison;
2. Test the proposal, by applying it concretely to the European languages.

This way of working allowed us to highlight areas of common ground and some aspects of discrepancy between the different systems for classifying morphosyntactic phenomena, and to provide a common nucleus of morphosyntactic distinctions. These are proposed in a feature-based notation, in the form of attribute-value pairs.

The present set of recommendations was tested with respect to all EU languages. Moreover, the specifications were adopted -- among others -- within the two large EC projects LRE-MULTEXT and MLAP-PAROLE, and -- as a validation task -- have been translated into formal and explicit guidelines for morphosyntactic encoding in lexicons for four European Languages: English, French, German and Italian.