A PLACE FOR CASE1
Stanley Starosta · Language Learning · 1976
Chomsky's configurational definition of grammatical functions is incorrect as a representation of surface grammatical relations and inappropriate and clumsy as a means for representing grammatical functions in deep structure. Fillmore's case grammar framework is a vast improvement, but it still requires a deep– surface distinction and thus does not contribute to a solution of the problem of the excessive power of transformational grammars. By representing ‘case’ as a feature of lexical items, the deep– surface distinction can be eliminated, and the resulting ‘lexicase’ framework looks promising as a basis for the design of language teaching materials which is much less abstract than Fillmore's approach, and which avoids the theoretical and practical deficiencies of the other transformational models.