The Explanation Of Scalar Implicatures With Default Theory

The concept of scalar implicatures, introduced by Horn (1972) as generalized implicatures based on Grice's first maxim of quantity, has proven to be a powerful tool in the semantics and pragmatics analysis of modal expressions like some and all. This concept has been considered as the paradigm of pragmatic inference and an important testing ground for competing pragmatic theories. The recognition of such scalar implicatures not only aids the understanding of semantic meanings, but also the 'logical' expressions in a natural language, especially the connectives, the quantifiers and the modals. However, due to the complexity, there are considerable disagreements in understanding the definition and condition of scalar implicatures. To the arguments applied in this thesis is default theory, one of the new logic approaches, to dealing with the problems of scalar implicatures. After introducing the default inference as the theoretical framework, as well as sample analysis, we focus on three aspects of scalar implicatures: (1) What is the relationship between weaker item and stronger item? (2) What is the condition of scalar implicatures? (3) What is the nature of scalar implicatures? Is it general conversational implicatures (GCI) or particular conversational implicatures (PCI)?The thesis first gives a general survey on scalar implicatures by introducing basic conception of scalar implicatures and dividing previous researches into three aspects: theory review, experimental review and logic review. About the theory review, Grice's approach, Neo-Gricean's approach and Post-Gricean's approach are presented to explain scalar impliatures. As for the experimental review, some experimental study on language acquisition and some experimental study to test the features of scalar implicatures are collected. About the logic review, circumscription and accommodation are recently applied to the explanation of scalar implicatures with logic approaches. Then, the thesis introduces default theory as theoretical framework to interpret scalar implicatures. As one of significant forms of nonmonotonic logic, default inferences appear to go through in the absence of information to the contrary; but additional information to the contrary may be quite sufficient to cause them to evaporate. Thus, the mode of inference appears to have two important properties: one is a default mode of reasoning, and the other is defeasibility. Since scalar implicatures have the same property with default inference, it is possible for us to divide those scalar items into several groups according to their complex relationship and adopt default theory to explain them in order to solve the dispute in understanding scalar implicatures. Therefore, we try to modify the previous definition and conditions of scalar implicatures, and provide our own views with the default approach. Besides, scalar implicatures have great potential in the research of Artificial Intelligence (AI). The purpose in writing this thesis is not only to illustrate one sort of implicatures, but also want to prove that linguistic theory is not always the mere method when we deal with linguistic problems. Logic approach, sometimes, can play a significant role in solving those problems.