A Cognitive Approach To The Semantics Of English Prepositions

As an old word-class, prepositions have been studied in linguistics for a long time. Some research in this field has been carried out by generative grammar, case grammar, etc. It has been shown that these traditional approaches, in which the role of syntax is central, do not reveal the true nature of the semantic structures of prepositions. By assuming that the meaning of natural languages is rooted in the human capacity for conceptualization, this thesis argues that three aspects, i.e., topology, function and force-dynamics, contribute a lot in the construal of situations and meaning formation. It will deal with the semantics of English prepositions under the framework of cognitive linguistics. The motivation for embarking on this topic lies in the desire to explore the nature of semantic structure and the way of meaning extension in English prepositions.In cognitive linguistics, the conceptual system that emerges from everyday human experience has been shown to be the basis for natural language semantics. Prototype theory is adopted in the categorization. Cognitive linguistics holds that the different senses of a polysemous lexical item are related in non-trivial ways. Metaphor is a basic cognitive mechanism in meaning extension. Spatial concepts are of great importance in human conceptual systems and in their experiencing the world. Prepositions as spatial expressions are highly representative of the nature of linguistic meaning. It becomes a canonical research topic in cognitive linguistics. At the same time, they have strong expressive force in denoting the relation between things.Cognitive linguists have made great achievements and contributed a lot in the field of prepositions. Most of the cognitive studies (such as Lakoff, 1987) show that the use of prepositions is a highly motivated system. The process, which extends senses from a central prototype to form a radial category, is systematic and widespread. The polysemous nature of prepositions could not be accurately described using semantic features or definitions; instead it requires an...