ABSTRACT
Both principle-based parsing and probabilistic methods for the analysis of natural language have become popular in the last decade. The former borrows from advanced linguistic specifications of syntax, and constructs its theories on the basis of native-speaker intuition rather than samples of actually occurring language use. This is partly as a result of the competence–performance distinction (Chomsky 1986) which states that the language that humans produce is not a pure manifestation of the knowledge of the grammatical rules being put to use.
