Dilemmas Of Chomsky's Feature Checking Theory And Possible Way Out

This thesis presents comprehensive theoretical analyses and empirical testments for Morphosyntactic Feature Checking Theory proposed by Chomsky [1] [2] [3] [4]. It is found that Morphosyntactic Feature Checking Theory runs into a wide variety of theoretical dilemmas and empirical quandaries, which become even more acute when this checking theory deals with languages almost without inflectional morphology like Chinese. This thesis attempts to formulate Categorial Feature Checking Theory on the basis of the theoretical devices provided by Haeberli[5] [6] as a way to maintain or even extend the empirical coverage on the one hand, and to get out of the dilemmas incurred by Morphosyntactic Feature Checking Theory on the other hand.In the course of investigating Morphosyntactic Feature Checking Theory, the thesis discovers the devastating problem posed for Chomsky's checking theory, namely, the morphosyntactic features, the very building blocks of the system, are neither empirically justifiable nor theoretically well motivated. To convincingly argue for the existence of the dilemmas of the Morphosyntactic Feature Checking Theory and attempt to find a way out, the thesis mainly revolves around the following two research questions. The first research question is to investigate whether the postulation of the morphosyntactic features is justifiable empirically and theoretically. The second research question is the attempt to give a more parsimonious and elegant checking theory which is free of the dilemmas incurred by Chomsky's checking model and which retains or even extends the empirical coverage.As for the first research question, the thesis bases its argumentations on the elaborations of the morphosyntactic features assumed in Chomsky's checking model. The thesis comes to the following suggestive conclusions:(1) The morphosyntactic features assumed in Chomsky's checking model lead to a series of theoretical and conceptual quandaries such as circularity problem, dubiousness of their identification, no genuine theoretical motivation and too stipulative.(2) Most of the morphosyntactic features assumed have no overt morphological realization in many languages. Thus, the stipulation that the morphosyntactic features are universally existent in human languages regardless of their different morphological behaviors in different languages is in great contradiction with one of the basic assumptions of the Minimalist Program. The MP holds that the morphological properties of the languages are the...